Rethinking Mobile First

I wrote the Mobile First Web Second blog post a few years ago. In that post, I talked about apps that were designed to be used on mobile primarily with the web as a companion.

There have been a number of startups that have taken that approach and done well with it. Most notably Instagram, and also our portfolio company Foursquare. It has become a bit of a orthodoxy among the consumer social startup crowd to do mobile first and web second.

But is it the right thing to do? Vibhu Norby, co-founder of Everyme and Origami, wrote one of the most thought provoking posts of the past month arguing that mobile first is a recipe for failure for most, if not all, startups.

Vibhu makes some excellent points:

All in all, mobile service apps turn out to be a horrible place to close viral loops and win at the retention game. Only a handful of apps have succeeded mobile-first: Instagram, Tango, Shazam, maybe 2 or 3 others.


You have an entirely different onboarding story on the web. You can test easily, cheaply, and fast enough to make a difference on the web. You can fix a critical bug that crashes your app on load 15 minutes after discovery (See Circa). You can show 10 different landing pages and decide in real-time which one is working the best for a particular user. You can also close a viral loop: A user can click an email and immediately be using your app with you. You can’t put parameters on a download link and people don’t download apps from their computer to their phone. Without the barrier of a download + opening the app to try your product, you can prove value to the user immediately upon their first impression, as is with Google. In addition, the experience of signing up for a service is superior in every way. Typing is easier. Sign-up with OAuth is faster. Tab to the next field. Provide marketing alongside sign-up as encouragement. Auto-fill information is a feature in every browser. The open eco-system of the web and 20 years of innovation has solved many of the most difficult parts of onboarding. With mobile, that kind of innovation is lagging significantly behind because we create apps at the leisure of two companies, neither of which have a great incentive to help free app makers succeed.


I use my phone more than anything else. I just don’t think that an entrepreneur who wants a real shot at success should start their business there. The Android and iOS platform set us up to fail by attracting us with the veneer of users, but in reality you are going to fight harder for them than is worthwhile to your business. You certainly need a mobile app to serve your customers and compete, but it should only be part of your strategy and not the whole thing.

Vibhu also takes a stance against the ad-supported, privacy challenged, free consumer app world. I respect that stance and every time I upgrade from a free ad supported app to a premium version (advertising free) via the in app upgrade on mobile, I express my solidarity with him on that one. But as a business person, I have and will continue to advocate for a free tier with a premium upgrade (or just entirely free) because as I have written many times on this blog, I think that is the value maximizing approach and it also allows the greatest number of users to access your product or service.

But I don't want to focus on business model in this post. We are at the start of what will be a long MBA Mondays series on business models and will be talking a lot about that.

What I want to focus on is the paradox that mobile is where the growth is right now and that mobile is very very hard to build a large user base on. Everything that Vibhu says in his post is right. Building an audience on mobile is a bitch. I talked about that in my what has changed post:

distribution is much harder on mobile than web and we see a lot of mobile first startups getting stuck in the transition from successful product to large user base. strong product market fit is no longer enough to get to a large user base. you need to master the "download app, use app, keep using app, put it on your home screen" flow and that is a hard one to master

But just because something is hard doesn't mean you shouldn't try to do it. I am convinced the next set of large and valuable consumer facing services will be built with mobile as the primary user interface. You can see it in the success of Uber and Etsy this holiday season. That's where you users are most of the time. And if you don't design your products and services for what is rapidly becoming the dominant UI, you will not maximize the success of your business in the long run.

So do I disagree with Vibhu? Not at all. I think he makes some great points on why you might not want to go mobile only unless you are in the games business. But I differ in two important areas. First, I think you can't abandon mobile. It is the future like it or not. And second, I think it is critical to design for mobile first and then build a web companion. If you design for the web and then port to mobile, you will find that it is really hard to fit your UI onto the small screen. Better to design for mobile first and then build a web companion. Mobile first, web second. But as Vibhu points out, the web can't and should not be ignored. It is valuable in many many ways.

#mobile#VC & Technology#Web/Tech

Comments (Archived):

  1. Guest

    As a Mobile App developer

  2. dovcohn

    As a Mobile App Developer it is extremely frustrating that Lean Startup principles of iteration and testing are so difficult/impossible on the mobile platforms. The fact that I can’t do legitimate UI optimization testing in a native iOS or Android app without a new build and submisssion to iTunes/Google Play is excruciating. Building a web-app or using web views is a option for more dynamic UI, but not when you want to take advantage of native features like GPS, etc.

    1. William Mougayar

      It can be done if you have a web mobile app.

      1. StartUpJerkFest

        u mean your web app is designed to work on a small device screen? responsive? i agree if that iz wot u ment

        1. William Mougayar

          yes. html5 based with responsive part of it. check and compare the mobile to desktop on what we have done. we push changes weekly, sometimes daily.

          1. JimHirshfield

            Agreed. HTML5 solves many of the aforementioned issues.

          2. Mike Bestvina

            And then introduces a whole new set of issues.

          3. JimHirshfield

            Life is like that sometimes. 😉

          4. StartUpJerkFest

            using Twitter Bootstrap v1? isn’t it on v2.2.1 now ? anyhoo, very nice site, nice design, nice resizing.

      2. Barry Nolan

        True but mobile apps are used five times as much as mobile webstes

    2. bsoist

      When we launched Archetype, we were picked as App of the Week and that gave us a flood of users. We had a couple of serious issues to recover from, but we were very fortunate that it was all on the server side. If we had to wait for submission and approval, it might have been over.

    3. dmackerman

      Use Cordova to bridge these features.





    5. fredwilson

      yes. this is the frustrating dichotomy of mobile apps

  3. pointsnfigures

    We are in the spring training of mobile. We haven’t had opening day yet. I get worried when people want to conclude things, become experts in things or control things that are in their infancy. Let the games get played and the market will tell everyone what the right path is. For some apps, the phone will be the place to start. Logging in and authorizing etc is just a problem for an entrepreneur to solve. I saw a business yesterday that has to start on mobile. It’s integral to their strategy. If they started on the web they wouldn’t get out of the batter’s box (to follow through with the baseball analogy-117 days to opening day!http://www.bleedcubbieblue….

    1. Brandon Burns

      post new link? 404 error…

    2. kidmercury

      agreed on the spring training point — still got a long ways to go.



      1. falicon


  4. Pravin J

    Just last week I was asking for update on Mobile First post. Thanks for posting this Fred.

    1. JimHirshfield

      Nice, crowd-sourcing blog post ideas. Gotta listen to your audience, right?

    2. fredwilson

      what more do you want me to talk about?

  5. Brandon Burns

    I feel like a point is still being missed: mobile is a channel, not a product strategy.Instagram worked mobile first because you use it to take pictures, which you do on your phone. You also use your phone to order a taxi, and to connect with friends when out — Uber and Foursquare.Google still has more activity on the desktop than mobile; why? Because people don’t like executing intensive searches on a small screen. Viral content driven platforms do better on desktop; why? People look at and share content mostly while goofing off at work, on the computer on their desk.Your product defines how you execute it, not the channel.Chase user behavior, not press headlines that say “X sector is growing!”Think like an experience designer, not like a VC.

    1. kidmercury


    2. falicon

      Agree 1,000%.It’s not a matter of mobile first or web first…it’s a matter of ‘problem frist’, ‘experience second’, ‘environment last’…meaning, figure out the problem you are really trying to solve, define the experience you really want a user to have/accomplish, and after you’ve got all that in your head take a look at the tech. and hardware and see how you can hack each one to best fit the ultimate experience you are trying to get at.BTW – I just had a sim. conversation with a few people over on Branch about this very topic just yesterday (… )

      1. Jonathan Drillings

        I think you guys are on the right track here. Mobile is most certainly a channel – it all depends upon the business as to where the focus really should be. Mobile first worked for instagram and foursquare because that was the appropriate channel for the service – they are inherently mobile products. For a product like evernote, for which a principle feature is cross-platform/cross-channel syncing, having both mobile and desktop/web is was critical. There are plenty of businesses for which desktop/web makes sense as a starting point – it really all depends upon the product.I do agree with Fred’s last point about designing for mobile first – or at least thinking about mobile. The smaller screen size forces you to really think through the most critical components of your product, and how to really lay-out the best user experience. Once you’ve thought through a smaller screen experienced, it’ll be easier to build a much more effective and intuitive user experience for the larger screen. I’ve done this before and it really helped.

        1. falicon

          agree. mobile first forces you to really focus on the ‘why’ of every part of your product and experience…but it doesn’t mean you have to actually force your product into mobile first.Just design and think about it that way is the point I think…

        2. Brandon Burns

          mobile-first design is a design principle that is automatically incorporated by any UX designer who legitimately knows what they’re doing. its second nature.the premise is to reduce all design elements down to what will work in the smallest space or channel where someone will see your product, and turn that into a system that can be executed in *all* spaces and channels where people will see your product.what it does not mean is to put your design on a phone screen first.

          1. takingpitches

            Brandon, you have a great feel for writing about user experience first design that comes out in your comments. You should blog if you don’t already!

          2. Brandon Burns

            thanks!i’m still trying to train myself to blog. but i’m horrible with routines.but i am going to teach more of my UX class on skillshare. i’ve gotten really great feedback on it.

          3. takingpitches

            Very cool. Is it a one session course?

          4. Brandon Burns

            for now, yeah. but people have said that they want to drill down into specific topics, so instead of one catch-all session i will probably break it out into several classes around types of platforms — e-commerce, content (blogs, news, video, etc.), utilities, etc.the real goal is to teach overarching principles, instead of “you must do this specific thing.” everyone wants to be told what to do when, but the most important trait of a good designer is the ability to respect general truths but execute flexibly.

          5. takingpitches

            I would love to learn the overarching principles, so i can actually hire and interact more productively with our user experience design folks, i.e., I am someone not going to be a user-design person, but will have very strong views on it, and want to understand when to push further and when I may just be plain wrong..

          6. takingpitches

            thanks for the rec!

          7. LE

            “Don’t Make me Think”That’s a great book. Much of the advice wouldn’t be needed of course if developers ironically put some thought in simply thinking like a normal or observing their interaction with their site design or paid attention to the questions that people raised when requesting help.Or even simply reverse engineered and surveyed the site design of the major players out there.I mean if you want to know how to design a fast food restaurant you can either read a book, go to expensive seminars, hire a consultant, or simply observe how the major players have done things as a starting point. There really isn’t anything “under the hood” with ui/ux it’s all out there for anyone to see.Disqus still hasn’t fixed the elements that appear only when you hover over them (same line as a commenters name that allow you to collapse or flag) even though this has been mentioned many times and it’s quite obvious that while this may be cool and possibly elegant it’s not functional at all. Taking action is not covered in any book.

          8. Elie Seidman

            great book

          9. ShanaC

            I love this book. Definitely one of the best books out there. Though I think Norman is good too!

          10. FAKE GRIMLOCK


          11. Richard

            Will do.

          12. ShanaC


          13. pointsnfigures

            would love to see you do that on

          14. Brandon Burns

            Next time I do the class, I should definitely list it on both platforms. Thanks for reminding me!

          15. FAKE GRIMLOCK


          16. fredwilson

            so true

          17. Brandon Burns

            true… and unfortunate. unless you resist being forced into bad things!

          18. leigh

            i like the myth that we only use a 10% of our brains and not surprisingly only 10% of most features of most products. Build a brilliant 10% has always been my philosophy but it’s hard to keep focused on it.

        3. ShanaC

          most products are cross channel. And I don’t believe that small vs big makes you build more intuitive experiences. You let that big screen feel mobile, you end up with the wrong product.

          1. Jonathan Drillings

            Most are ultimately cross-channel, but there is often a more natural channel to start with (sometimes not, but most of the time).I agree that making a big screen feel mobile ends with the wrong product – thats not what I meant, so perhaps I should expand. Thinking about a product with respect to reduced real estate forces you to think about the critical elements and flow of a product, which in turn helps you to better design your information architecture. When you move out to start thinking of a larger-screen experience, with greater real estate, that thinking should carry through. Ultimately, it should help eliminate a lot of extraneous elements (which may have made it in if you started thinking big screen) – making it a cleaner, more intuitive design.

          2. ShanaC

            ok, true, but the underlying element is that the design should be responsive and that all main tasks should be easy to get done irrespective of interface

      2. Brandon Burns

        Interesting convo on branch. more people need to have it.but even more importantly, people need to get the overarching point — don’t follow buzz words and headlines, follow user behavior and good product design principles.when entrepreneurs sell buzzwords like “mobile first” or, today’s favorite, “enterprise,” they get the funding, press and attention — no matter if the products make sense for the pigeonhole they’ve put themselves in. then people who have a perfectly good product vision with tons of potential end up mucking it up because of where the buzzwords and VC clamor have swayed them. or those who refuse to be swayed, and do things the right way, can’t move as fast because they’re neglected by the investors and press who could and should help, but don’t because they’re too busy chasing buzzwords.sure, maybe the strongest survive but, either way, everyone loses.

        1. ShanaC

          Oh, the problems of marketing. You have to be what they want, without actually being what they want. The money is in being tease.

        2. JamesHRH

          @fredwilson:disqus is still right – Mobile is (the) First Problem.

          1. kiran bhanushali

            Maybe most common everyday problems have been solved on the desktop and web? A sign of maturity of the platform? Looked at from a Maslow’s hierarchy of needs the web has reached the top of the pyramid while mobile still in the bottom rungs.

          2. Emil Stenström

            I don’t think so. The fact that mobile is where the growth is just means that lots of new problems are solved using mobile tech. That’s not the same as YOUR idea necessarily working out on mobile. Mobile is just a channel, although one where there’s lots of innovation is happening…

        3. Steve Haney

          I should have just read your post and @falicon:disqus’s reply before sounding off above. I basically restated what you both said. Great insights! +1

      3. ShanaC

        I would stay take a step back. You need to ask more about your customers/clients/user first. Maybe you think it is a bug that you are solving, but to them it might be a feature.

        1. Brandon Burns

          well said

      4. AskarH

        That’s a great quote there @falicon:disqus Every (most) solutions out there started off with a problem for sure. As soon as you have identified a problem these days the next factor to consider is the platform, mobile or web.I agree with Fred and I learnt it the hard way. I hacked together an iOS app,, (check it out at to solve the problem of Muslims keeping track of their everyday prayers to get some insights on how they are doing with their prayers.I took the “mobile first” approach and thought it will take off as these devices are everywhere these days. The problem was the target audience that I went for was not that mobile-savvy (for the most part) and they would rather prefer to see some sort of a tangible thing, aka web, to go back anytime to see their stats and stuff. Also I would have automated an email reminder service to remind them everyday to click on a link to get this chore done. After some analysis a few other points are making the “web-first” approach a more viable one and these mobile apps would certainly augment the web experience for sure. But in my case basing it off solely on the mobile experience seem like a mistake.Even though I slightly disagree with your quote, it’s a great quote for sure.

        1. falicon

          Thanks…finding the right problem to focus on is actually a *very* challenging task…so many things look like they could be interesting and valuable things you can quickly/easily fix with a little bit of tech…but turn out to be very difficult to turn into a business or grow out of a very tiny niche…I think it takes a lot of trial/error and experience hacking away on simple ideas before you start to get good at identifying the things (and reasons) why something has a solid chance at taking off…at least I know I needed that (and though I *think* I’ve gotten better at picking out what to hack on or not, I still have a lot of improvement on that area to go).I think there was probably a lot of great learning and experience that came out of your efforts…I don’t know that a mobile vs. web first approach was the real problem here…reminder and tracking services even when viral (like diet plans) are very difficult to gain adoption and stickiness with…Anyway, I hope you can use all that you learned from the experience for the next idea/attempt you go after….

      5. ceonyc

        “It’s not a matter of mobile first or web first…it’s a matter of ‘problem frist’, ‘experience second’, ‘environment last'”+1

        1. falicon

          Thanks! 😉

    3. ErikSchwartz

      “mobile is a channel, not a product strategy”This. Absolutely.

    4. William Mougayar

      same points i made as well. agreed!

    5. fredwilson

      Its a channel but its going to be the dominant channel and Google (or a competitor) will solve the problem of searching on a mobile device

      1. Brandon Burns

        maybe, maybe not.remember when people thought the new channel of tv was going to be big for listening to music? and where are MTV and VH1 now?i’m not saying search will never be mobile friendly, but i am saying that not all behaviors work on all channels. and just because a channel is dominate doesn’t mean you can force your product on it and expect your users to follow suit.

        1. falicon

          I’ve struggled a lot with trying to figure out how to make search truly a useful experience on mobile…honestly I haven’t come up with a solution at all yet.My gut says that it will need to work more like a conversation, because most mobile at it’s core is built to be conversational with…something akin to a text message back and forth until the answer you seek is presented…but that’s not completely right yet either…which leaves me still pondering it quite a bit…

          1. JamesHRH

            For an excellent mobile UX, check out on an IOS device.The vertical layout is very useful.

          2. falicon

            awesome thanks!

          3. kidmercury

            voice. whenever i need to do the rapid, deep searching i do on my laptop, i use voice. i think there are behavioral/cultural issues here — i feel a bit awkward and people do look at me strangely when i try to enunciate short phrases extra clearly into my phone — but it gives me the experience i’m looking for and i think the behavior is sufficiently intuitive to gain adoption (IMO).

          4. Donna Brewington White

            What about voice is more effective/desirable for you?

        2. ShanaC

          it wasn’t that listening to music on tv wasn’t successful: it was successful for years. But the real world was more successful because music was a lifestyle thing, and they wanted to show how to be that lifestyle.

        3. fredwilson

          i think we fundamentally disagree on this Brandon. i don’t mean that in a negative way at all. i totally respect your perspective but i don’t really see mobile as a channel. i see it as the dominant way we will interact with the cloud ove the next decade or two

          1. Brandon Burns

            constructive discussions around disagreements is the spice of life. 🙂

      2. aweissman

        I would submit that in a subtle way, google has solved part of the problem of searching on a mobile device through youtube. ie, I wonder if youtube is the most widely used “mobile” service in the world, and the one with an enormous volume of searches going through it on mobile every day.

        1. Brandon Burns

          having a search box on your platform doesn’t make you a “search” is a content platform, with video being that content.a popular use case for watching video is while in transit… thus mobile is a good channel. the experience of watching a video is click play, then sit back and relax… again, friendly on a small screen.using a search engine involves looking a a page of text, trying to find a key word that pops out, clicking on the link, going back to the engine, and repeating the whole process until you get what you want. a necessary evil on a desktop, and an annoyance you can easily rationalize as not being worth it on a hasn’t “solved search” with youtube, because youtube isn’t a search product.

          1. aweissman

            This I am not sure I agree with “youtube is a content platform, with video being that content” because I think more and more people are starting their internet search experience with youtube. this is a behavioral change but I one I think is happening. For example, the largest music platform on the internet is also Youtube.

          2. Brandon Burns

            i also use youtube to “search” for music, more than any other platform.i’m not sure i would go as far as to say youtube is a search engine operating in the vertical of film and music, but you’ve planted that seed… maybe i should…

          3. kidmercury

            i start many of my commerce searches on amazon. sometimes, if amazon doesn’t have what i’m looking for, i’ll turn to google. and a couple times, amazon didn’t have it, but i did click on the text link ads on amazon to places that seemed like they might have it (but they didn’t).

          4. aweissman

            great comment

          5. daryn

            that’s what we like to hear! 🙂

          6. takingpitches

            You’re far from alone. According to the Economist this week:”Some experts think Amazon also poses a threat in this battle to find things. “Google used to be the toll-taker, directing people to Amazon,” says John Battelle, a seasoned Valley-watcher and the founder of Federated Media. “Now people are increasingly bypassing it and going straight to Amazon to find and buy stuff.” He has a point: Forrester, a research firm, reckons that 30% of America’s online shoppers begin their search for a product at Amazon”

          7. takingpitches

            And meanwhile, the number one advertiser for Google paid search is the University of Phoenix, spending over $275,000/day…

          8. ShanaC

            that is so depressing. it makes me think the US is going hell.

          9. takingpitches

            there was a lot depressing in the discussion on Fred’s student debt bubble post earlier this fall.

          10. ShanaC

            This is really product dependent. Nice shoes I don’t always immediately go to Zappos….But yeah, they are walmart.

          11. ShanaC

            what is your basis for thinking this?

          12. aweissman

            anecdotal mainly, a hunch, though I am sure there is some data

        2. JamesHRH

          Andrew – I have never searched youtube on a phone. See split uses comment above. Do they get a lot of phone (not tablet) traffic)?

          1. aweissman

            great q – I am looking

        3. ShanaC

          Excellent question, but is that happening?

        4. fredwilson

          and you can’t change the search app on android from google to duckduckgo. android in some ways is google’s bet on mobile search

      3. Abdallah Al-Hakim

        I think Google already has a much nice search app for the iPhone and their voice search is rapidly improving

        1. Brandon Burns

          i think voice will do good things for mobile search.the problem with the search experience is that its taxing. voice makes it less taxing on a phone, when it works properly.

          1. kidmercury

            yup, just said the same thing in a comment to @falicon. voice ftw!

          2. Donna Brewington White

            okay I see this after asking the question above — I can see why voice is more appealing on mobile (or less painful– I hate typing on my phone) but you said you also use this on laptop

      4. ShanaC

        what is the problem exactly?



        1. fredwilson


          1. FAKE GRIMLOCK


        2. William Mougayar

          Agreed 100%.

    6. takingpitches

      Mostly agree but there are some behaviors that are not inherently desktop, like search or blogging or getting work done (when you travel a lot for work). Those behaviors becoming table/mobile is inevitable as the usage goes toward tablet/mobile and the desktop/laptop are left behind. Some of the lag, imho, is due to the fact that folks like Google, who may be dominant in some of the desktop behaviors, have failed in giving a satisfying user experience on tablet/mobile.

      1. Brandon Burns

        i think the lag is more in the fact that “heavy travelers who do most searching on tablets” don’t represent large enough of a population for google to rejigger its entire experience design to please people on tablets in airports, at the expense of everyone else on a lap/desktop. nor is google motivated to create a unique tablet experience that’s different from the other channels — they like consistency across will probably take a new, smaller upstart to craft a tablet-optimized search experience. and doing it for the travel set would possibly be a good jump point.the key is that it would be a product for the travel set — and choosing the tablet as a channel is a choice to go where that consumer is partaking in that activity. the channel decision follows the product and its user, not the other way around.

        1. takingpitches

          The last I recall hearing of Google searches on mobile was about 25% of all its searches (don’t have stats onhand, but can find them), which represents more than heavy travelers on tablets.

          1. Brandon Burns

            awesome stat. thanks for that! now i’m curious…that’s the problem with this blog. you start talking to people, end up researching something new someone shared with you, and *whoops!* there goes your day!but its a welcomed distraction. :o)

          2. takingpitches

            seriously, too many days like that, but the virtue of this blog is it’s a worthwhile and productive distraction!

          3. takingpitches

            the macro-trend that sits above this is there are many user studies (including by Google – look on for an admob study) that show user subsitution from desktop to tablet once people buy a tablet (i.e., when i am casually surfing in the morning at home including getting on avc, i rarely boot up my laptop anymore), and a lot of the sales data that came out from PC and microprocessor manufacturers this year showing shrinking laptop/desktop purchases as tablet purchases are rising.that reality that eyeballs generally are moving is what i think Fred is talking about and it has to be addressed for many many businesses, although all your great points also have to be kept in mind so as not to go mindlessly trend-chasing when your business might not fit this paradigm.

          4. ShanaC

            doesn’t mean that the behavior behind surfing is different. It isn’t mobile, it is nouveau desktop

          5. ShanaC

            then where else are they searching?

          6. takingpitches


        2. fredwilson

          i am looking at three to five years. that’s how i work. and i think in that time frame mobile will be the dominant way people search and blog and do most things

          1. Kasi Viswanathan Agilandam

            Yep. the next

          2. RichardF

            I don’t even think it will take that long (if you are including tablets as mobile) I don’t know what your device viewing stats are like for AVC but I have a side project that is a tiny niche site with just 5000 uniques per month but 1/3 of those visitors are using tablets with the vast majority (unfortunately) on ipads.

          3. FlavioGomes

            Yes..but maybe not on 4 inch screens or even tablets. Voice activated and wearable?

    7. JLM

      .Well played.One of the all time best comments ever made in Fred’s Joint..

      1. Brandon Burns

        quite a compliment! thanks!

      2. John Revay


    8. Elia Freedman

      Great comment Brandon. I wanted to expand on the thinking with some thoughts of my own from the past year.To consider Instagram, Foursquare, Google, etc., web or mobile companies is a fallacy. None of them are web nor mobile companies. All of them are web companies, meaning those that use the web as the center-point of their products. This does not imply anything about how you interact with the service. Google search is primarily used in a web browser. But I’ll bet that Google Maps is primarily used on mobile devices. Foursquare and Instagram chose to start with mobile interaction models, but they are really companies whose primary business and intellectual property resides on the web.Evernote provides a very interesting insight into this, I believe. Evernote has desktop, web and mobile clients for every conceivable platform. But Evernote’s business is not any of those consumer-facing products at all. Evernote’s value proposition is that they make it really easy to save your stuff and see your stuff. This really becomes clear when you look at Trunk. There are a hundred apps for getting stuff into and out of Evernote. In fact Evernote open sourced its entire desktop client so anyone can see the source code.So my point: the web is the business and then figure out how people will interact with that business, Android, iOS, web client, desktop client, etc.

      1. mydigitalself

        Great point on Evernote.I’m finding it difficult to dissect Vibhu’s article properly as he’s talking about many problems as if they are one problem.We’re just getting up and running at the moment with our own startup, and we’re not thinking web first, nor mobile first, nor tablet first. We’re thinking service first. All of their applications consume or publish to their services in appropriate ways. Now they are starting to shard their apps into little verticals based on use cases, and they are doing a good job of marketing these apps and that isolates a value proposition very well.But the service and how you think about the service and how you think about each device and how it consumes the service is very important. Our users will behave differently on desktops to the way they will on phones to the way they will on tablets. These interactions happen in different environments, for different reasons, and our service is, therefore, device aware and acts accordingly.Now when you end up designing the manifestation of the service on the device, yes, you think about that device first and don’t try have a unified UI that works across all of them because the behaviours and the needs on those devices are different.On the whole point of being able to test, we’re building our mobile app pretty heavily on HTML5 and really trying hard to get that to work for us. This should mean we *can* test out 20 variations of a signup page etc…

    9. Kirsten Lambertsen

      I think Fred’s last point still stands though. *Design* for mobile/tablet first. (i) it’s more efficient than doing it the other way around, and (ii) people are coming to expect that more simplified mobile experience on their desk/laptop. Sites that aren’t adopting the mobile aesthetic are looking very web 2.0.

    10. LE

      ” mobile is a channel, not a product strategy.”A channel can be a product strategy though. Not that I think labels and categories matter that much.”Think like an experience designer, not like a VC.”I think it is also important to serve the master that is enabling you to do what you want to do though.If that master believes that something is headed in a particular direction and is more likely to fund you if you are going after that market, then go for it. Once you get get your money you can always move in another direction.I once did a presentation project for a successful real estate developer that was putting together a pitch for a casino. His theme was called “Carnival Club” and it was a family centered resort back when this was what was becoming big at the time. (Was just starting to be done in Vegas).In doing the project became apparent to me that this was simply a placeholder in order to convince the investment bankers to fund his project. After he got the money he could easily present other evidence to the contrary and go in another direction. Had he started out in the beginning with his original dream he would never get the money.

      1. Brandon Burns

        “I think it is also important to serve the master that is enabling you to do what you want to do though.”i’m betting its not important to appease the investor. it sure as hell is a risky bet, and i very well may fail, but we’ll see.

        1. LE

          “it sure as hell is a risky bet”I think the reason there are so many artists out there that can’t earn a living is a testament to the difference between the way artists think (and I would throw programmers and any other dreamers into that mix) vs. business people. And business people does not mean necessarily MBA’s. Or Wall street. It means people who recognize a need and fill that need and make money from it. And at that point they can do creatively what they want to do.That is not to say of course that there aren’t business people that fail and artists that succeed. But in general there is more evidence of people being able to earn a living by thinking like a business person (and hiring artists) then someone who thinks like an artist. And just believes they can click their heals and wish and that there creative vision is so great (or programming prowess) that the world will beat a path to their door.

          1. Brandon Burns

            you’re right.i come from advertising, where from a mass of artists, few rise to the top to manage large teams and multimillion dolar client budgets. and many never rise because they can’t get with the program.that said, investors are not necessarily the master with the key to starting and growing a good business.we in the tech world forget this, but investor capital funds less than a quarter of all business in this country (if i find the study that shows this, i’ll repost later); most is through personal savings or friends and family. there are other ways to skin the cat.

          2. LE

            “we in the tech world forget this, but investor capital funds less than a quarter of all business in this country (if i find the study….”I’ve never taken any investor money at all and while I’m not sitting on multi millions I’ve had a decent lifestyle since college (which was a long time ago). All self funded with the exception of some machines I had to lease in the early years. And I’ve seen many others do the same. Because they were realistic and not gamblers looking for the easy way to quick riches and hoping to be the lucky one. The difference between business as displayed in the pressm, and lotteries, is everyone knows the actual chance of succeeding with a lottery and they only gamble a few dollars for the dream. Not years. Oh sure everyone says they know startups are risky but they still get sucked in.I earned the money to what I did by being self taught in photography while still in high school and having a darkroom in my basement (in addition to several other things – I’m just pointing out the photography because it’s creative). The photography that I did (for lawyers as well as catalog photography) was strictly based on what I felt I could earn money from. And it was just as enjoyable as something that people probably would feel was more traditionally creative “art”.

          3. Brandon Burns


          4. falicon


          5. ShanaC

            that is because they don’t tell you in art school that the best art comes from figuring out how others see the world and that transposing their visions on new ways of seeing.

        2. falicon

          I don’t think it’s important to appease an investor…I think it’s important to find the right investors, ones that buy into your true vision (then there is no appeasing necessary).

        3. fredwilson

          appeasing the investor is always a bad strategy. you must do what you believe in. i believe in negotiating with the investor but not appeasing them

    11. jason wright

      for success in mobile end the distinction between the channel and the product. Merge.

      1. William Mougayar

        Good insight. access anywhere, and the app adapts. It’s a continuum, not a distinctive wall.

    12. William Mougayar

      Brandon, Web and Mobile aren’t really distinct if you think about it as a Continuum. When the app doesn’t care whether you’re mobile or not, then it doesn’t matter. The app lets you do what you want to do, wherever you are.

      1. Brandon Burns

        “When the app doesn’t care whether you’re mobile or not, then it doesn’t matter.”But I 100% agree with you that you should make your product work on every channel you possibly can. But it doesn’t always work.Before good cameras, 3G wireless, the app store — a lot of products didn’t, and couldn’t have worked on the mobile web.And some never will. Like professional grade creation tools like Adobe Creative Suite and Final Cut Pro. And most developer tools… even mobile ones! We code for phones on a desktop! And always will. Find me a developer who’ll write hundreds of lines of code on a small touch screen. Not gonna happen.Not all channels fit all products. But I 100% agree with you that you should make your product work on every channel you possibly can.

        1. ShanaC

          I disagree with this. To really render film you need a server farm. It might be totally possible for someone to build a web app that basically runs renderman on the cheap on amazon. (… )Adobe is also moving parts of CS into the cloud. it becomes cheaper and easier to do complicated manipulations.

          1. Brandon Burns

            yes, rendering issues are being solved.but servers and cloud storage don’t change the fact that creatives want big screens to design and edit film on.its not about rendering and the file compression part of the experience. its about being able to see what you’re doing in the creation part of the experience.however, products from Mixel to Vimeo are showing that you can have a good creation experience on a tablet, if its for a casual user. but professionals still won’t touch the stuff; they want their Adobe products on their big screens. i know professional designers who won’t even work on a laptop, because its too small. getting them onto a phone probably won’t happen.

    13. raycote

      Analogy ?? —-> admittedly from the know nothing peanut gallery!Apple built out a robust hub of integrated hardware experiences = Web AppsThenApple built out a set of convenience driven retail access channels = Mobile AppsGoogle is doing the same in software.Google built out a robust hub of integrated software experiences = Web AppsNowGoogle is building out a set of convenience driven mobile access channels into that array of software experiences = Mobile Apps

    14. Dale Allyn

      Very well said, Brandon. This is precisely my view as well. Coming late to the party, I’ll just sit back and enjoy the conversation (again), seeded by your excellent comment.

    15. hypermark

      I think that you make a great point, but you also present a false dichotomy.Native platforms in the context of iOS and Android are SIMULTANEOUSLY a discovery mechanism + marketplace, distribution channel, software platform and “billing relationship.” My sense is that good native design factors all of these elements into the equation.Case in point, after learning the hard way about the hit-driven nature and short half life of individual apps, we have started building series of apps that all run within their own “app store” app within THE App Store.This allows us to pursue a land-and-expand strategy whereby we can engage users with multiple freemium offerings that are integrated together in a more than the sum of the parts fashion via data, workflow and reward mechanics.It has also changed the WAY that we build these apps so that core application logic resides in the native runtime, but content, such as text, art, audio, video and other interaction logic can be called upon algorithmically or randomly to create a very rich, dynamic user experience.Because our model essentially shuttles content between the native run time and the cloud, it allows us to achieve a nice hybrid between the real-time robustness of a native app and the just-in-time agility of a web app. As a result of this, changes to content, layout or interaction settings can be made instantly without going through the app store submission process, and in fact, we are able to release new derivative apps from existing templates via this same model, which addresses a number of the issues that Vibhu is referring to.Here’s an example of a system that we recently launched with Macmillan via this approach, just to ground what I am talking about in real world terms:…By contrast, HTML 5 continues to be in the elusive not quite good enough for large categories of applications; namely, those where dynamism, data-driven context and intermittent connectivity are part of the definition of the situation.

    16. ShanaC

      people don’t bother to think like an experience designer. Because it is hard to sit in other’s shoes.

      1. Brandon Burns

        which is never ceases to amaze me, the nature of people to believe they have the ability to do something they have zero training or experience in. just because you use, talk about, or even finance digital platforms doesn’t make you an expert on how to actually make one.but then again, my experience in designing digital platforms does not qualify me to build a business from scratch — they are two different beasts. but i’d like to believe i can do both, anyway.i guess its our naive arrogance that moves the world forward!

    17. JamesHRH

      Nice.Vibhu’s product statements about web v app. are all right but he is very dangerously wrong about the key market statements he makes.People do not want to pay for web content. The cost of privacy is not ‘monumental’. Betting on slower, fee for service growth is very unlikely to succeed, unless he has a unique VP & unique investors.If you write a blog post that could cost you your job, you should expect to lose your job. The web is not private, was not designed to be private and carries no expectation of privacy: privacy is layered on by the users.You send an email to me flaming your boss, you trust me not to distribtue it company wide. Outlook, Gmail, Mail carry no inherent privacy & never have – people getting hammered via non-provate emails goes back forever, to like 1996.To get to how squarely Brandon hit the head of the nail – mobile is a mess & noone has a very solid answer for it, outside of the Fred’s short list.I tweeted this yesterday: ” web not used properly, retailer division: I google you via IOS, go to your home page, where I can like you on FB but cannot find a phone #”Part of the issue is that tablets are media devices and phones are IRW navigation devices. The products may share OS foundations, but the users do not share activities – ever see anyone navigating around town with their tablet?

      1. ShanaC

        how is privacy layered on by the user? The user doesn’t control the technical end…

        1. JamesHRH

          It does not have to be, but if it is, it is done by agreement.Sort of like:”I am going to send you my thoughts about my boss & would like you to suggest some potential courses of action. Please don’t share this email, as he is a thin skinned tyrant & has fired 5 people for anything that even sounds like criticism. Seriously, sharing this email will get me fired. Trusting you here.”

    18. Kasi Viswanathan Agilandam

      81 up votes on AVC damn that is good … I made it 82. Great comment.There is one point missed though in the whole conversation through-out (or may be i missed reading it…i skimmed through all comments though)… there are going to be billions of users of the web through mobile who have never seen a desktop and never will see one…they will see the web only through mobile… I can show you at the least 200million+ mobile only users as of day TODAY in India alone.Well there is no doubt we will have billion mobile ONLY users…but the question is whether solution fits their requirement (OR r they potential customers?).

      1. fredwilson

        i wonder what it means when i fundamentally disagree with a comment that is among the most upvoted comments ever on AVC

        1. Kasi Viswanathan Agilandam

          rebellion attitude .. boss is always wrong :-).I also disagree with that comment but up voted for the conversation it created. Lots and Lots of info there… I think it is one of the biggest thread comment after andy’s comment some months ago.

        2. John Revay

          I read Brandon’s comment and may have even up voted it.I then read through most of the comments on post…Then I think I even more so understand your point, especially w/ the 3 to 5 year view.Mobile will be the main platform, there are unique features of mobile devices, (Camera, Location, iOS APIs etc) best to design some of product’s features to use this functionality now – or someone else will.

    19. Mark Essel

      Right on Brandon, was trying to remember all my comments on this topic and I let remember for me:

      1. falicon

        Nice! I award you a kabillion points for this answer. Well played sir, well played indeed!

          1. falicon

            interesting…do you feel like mobile/app interfaces have continued to be pushed forward since that comment?

          2. Mark Essel

            Yeah, there have been great incremental improvements, and they’ve made there way back to web interfaces. Not too many wow moments, but a few addictive camera/video tools to extend what a smart phone can do

  6. takingpitches

    My personal user experience is that you have to consider the transition to tablets in the mobile/web distinction.As tablet replaces desktop/laptop use, like when I travel and as the data on PC sales suggests it is for other users as well, I don’t know how one cannot put mobile first. I am much more forgiving when something does not work on my phone – such as a Google Spreadsheet – than I am when I cannot make the edits on my tablet (as I wouldn’t be forgiving on my desktop/laptop). Without getting that experience right, I think it will be even harder for new applications to gain traction or existing ones to keep traction because that is where users are or are going.

    1. Abdallah Al-Hakim

      I just wrote a comment about tablet vs mobile phones before reading your comment! I think your point about the expected user experience between tablets and laptops is an accurate one. This is why a web/tablet strategy first will make more sense for some businesses.

  7. kidmercury

    my mantra on mobile is that much of it is very early and its evolutionary path has been a bit distorted by bubble 2.0. i think it will be much easier to understand once the dust settles on the unwinding of bubble 2.0. i think we’re about half way through that; if we apply the kubler-ross model, i think we’re in the bargaining hunch is that more companies will need to bundle apps with hardware and sell the whole package. mobile is a vague term. do you mean the 3.5″ screen in my pocket or the 10″ tablet that never leaves my home? or is it a 5.5″ screen in your pocket? with the new galaxy camera it now also means a full-blown camera complete with a unique form factor and case uses. i don’t even like the term mobile, and prefer infrastructure instead — because ultimately the mobile operating systems and touch screen interfaces will find themselves implanted in a wide variety of stuff, much of which is less mobile than a desktop PC (like perhaps a telephone pole, for instance).even if we disregard this view, when i look at the apps i’ve installed on my phone, i see that i have a certain loyalty with the organization — and that this loyalty was built on the web first. amazon is a prime example. i don’t think i’ve ever bought anything from amazon off my phone yet, but i’ve looked stuff up and i’m sure it’s only a matter of time. but to get on my phone, i need to trust you, and that trust comes from repeated, web-based usage. is another example for me. sure, this will likely change as mobile grows over time, although if the trust is built via a tablet that never leaves my home, i think that introduces a different marketing/distribution strategy. put more directly, i wonder how much trust can be built over the phone. the phone is about instant, personal usage, but to do something instantly and personally, i need to trust you first.

    1. Brandon Burns

      “when i look at the apps i’ve installed on my phone, i see that i have a certain loyalty with the organization — and that this loyalty was built on the web first”yep, that’s standard user behavior. even if you are building a platform that you believe will eventually primarily be mobile (as i am) you have to understand and respect this flow, or get yourself into trouble.its a shame even so called “product” people don’t have legitimate UX design training. these are basic principles to those who know better.

      1. ShanaC

        why do I have to respect the flow

    2. Carl Rahn Griffith

      Spot-on.Regardless, a crap/no-added-value app/web-site/et al, will not gain traction/monetization, regardless of the device it’s on.It’s akin to thinking TV shows on the latest 52″ 3D HD flatscreen TV in your living-room will suddenly become more compelling because of the nature of the contemporary device they are being aired on. The Kardashians is still tripe, regardless of the device it is being displayed on.It’s about the content, not the device…What’s the best mobile app around? For many it’s Twitter. The content? User-generated, mainly.We’re all user-choosers, nowadays.QED.

    3. Michael Elling

      They wouldn’t have to vertically package if google and the droid ecosystem partners built more powerful control layer utilities/features. These would both bridge the disparate networks and devices more efficiently for all apps, and help apps integrate with each other seamlessly, thereby increasing their usability and utility for the end user. The latter does a much better job of defining their set of unique needs.I don’t believe the vertically integrated (and siloed) iOS model is sustainable. It is/was a short-term aberration to get around the vertically integrated monopoly carrier bottlenecks. By reducing the importance of a given network/device, expanded control layer solutions would get around infrastructure bottlenecks in mobile and fixed broadband.I wait to see how google relates the KC fiber efforts (anvil) with the droid ecosystem (hammer) to break the infrastructure monopolies. The answer will lie in the control layers.

    4. JLM

      .Lot of great stuff in there.Well played..

    5. LE

      As I mentioned in another comment I think the dividing line is input device not screen size.I will take a 5 inch screen size and a full keyboard over a 20 inch screen size and a virtual keyboard anytime.

    6. ShanaC

      True about “which mobile”. I hate the way tablets are defined as mobile. it isn’t descriptive as to what user behavior is on these machines.Web usuage isn’t enough of a condition to build loyalty. And what if you can build mobile only loyalty? Then what?

    7. fredwilson

      i agree with “loyalty was built web first” for many apps on my phonebut what about the startup who is getting going now?

  8. Barry Nolan

    Both arguments are correct – mobile is the future of customer engagement, and persistent engagement in mobile apps is hard. It’s especially hard because the tools we use in the physical or virtual world (CRM, messages, updates, tutorials, polls, questions, feedback, dialog, etc), by and large don’t yet exist for the mobile app world. These tools are the grease to iteration to get app engagement to a better place. Today the mobile apps publishers are hermetically sealed from both knowing and communicating with their app users. This is the problem our startup is addressing.Mobile, particularly apps, are the future UI with consumers and customers. As an illustration, a banking client of ours have 9m mobile app users – of which 8.5m are MAU. Because the app experience is so much easier than the desktop, almost to a default, customers rarely go use the desktop app. Utility = engagement = profitability. Design (in the Steve Jobs definition) that serves customers, serves the business.

  9. William Mougayar

    I think a key test is- Is the app inherently Viral because of mobile (eg. Instagram, Foursquare) or is it a Utility that benefits from a mobile convenience?To just say that everything is going mobile is a broad statement. The desktop or the tablet still offer a good experience, although the Mobile native app focuses that experience.Mobile Eventually. Sometimes first, sometimes second. It’s what’s driving the usage that matters.

    1. Barry Nolan

      +1 on utility. Thriving in the new mobile app economy relies on the same YC mantra “make things people want.” It’s especially relevant in mobile apps given most apps are free, disposable and subject the massive churn from download to long-term retention.

    2. Michael Elling

      Across a wired and wireless world I have a filter called the 4Cs of supply and 4Us of demand. Cost, coverage, capacity, clarity (QoS). User-interface, usability, ubiquity and universality. Pricing (be it subscription/prepaid or centrally sponsored/procured) and usage are the keys that drive revenue. Pricing and usage should or will be a function of marginal cost. Marginal cost is arrived at a priori by multiple iterations of supply and demand intersection points. The nice thing about the 8 parameters (best illustrated in a blue ocean strategy context) is that it will help reveal what people want (utility value) versus what they have (or don’t have) a priori. They also work across and relate the upper, middle and lower layers of the stack in a critical and objective fashion.

  10. Michael Elling

    The mobile web is fundamentally different from Web 1.0. The latter was an unintended consequence of a monopolistic pricing reaction to the competitive WAN threat in the mid to late 1980s. Only after the flat-rate, unlimited dial-up networks (layer 1-2) had been developed by commercial ISPs did layers 3-6 develop in the store and forward, database look-up “internet” between 1989-1994. What’s different about the mobile web is that bandwidth is slow and outrageously priced relative to underlying technology. This results in a lot of the challenges you and Vibhu point to.Since 1996 we managed to remonopolize the communications industry that gave rise to the internet. When I look at comments on AVC and the trends in Mary Meeker’s recent report they mostly fail to recognize and understand the import of this. One doesn’t solve a problem by adding to it, but by addressing it. VC’s and the application (upper layer) cos they invest in need to better understand what is going on in the access/transport (lower) and control (middle) layers.As for the mobile vs web argument, I have maintained since the early 1990s that mobile is just access for wired/web/cloud based applications, be they text, graphics, voice, audio or video. My criticism of the mobile web are apps that are silos and don’t work across all platforms (or across each other). I love the fact that I can read a book on a smartphone, then pick it up on a tablet, then read it as an open window on my PC as I wait on some other application to process/upload/download, etc… The user should be able to define their own unique demand curves for any application/use; be it creating complexity or distilling down to simplicity.One control layer function that I would really love would be able to define and set my smartphone OS/environment from my PC. I waste so much time doing it manually on my relatively small screen. Hopefully the google/android folks wake up to this opportunity. Doing so, would immediately scale the app layer at the same time.

  11. StartUpJerkFest

    i hate YouTube’s native iPhone app. I can’t find my watching history, and it assumes i care about popular currated content, which i most certainly do NOT

  12. CliffElam

    You know, I *know* you love uber, but my one experience with them was not positive. It wasn’t worse than other ways of getting a ride, but it certainly wasn’t better.And I’ve never even been on etsy and don’t know anyone who has ever spoken of it. Certainly it’s not part of my social ecosystem like eBay/Craigslist are/were.I always find it jarring when I’m not interacting with what a smart person thinks is a big market. Never felt that way about facebook or twitter (*cough* AOL *cough*), but things like Square/etc just leave me cold. And that’s a bit disturbing.I’d make a lousy VC, there are so many things I don’t understand.-XC

    1. LE

      Etsy is “your place to buy and sell all things handmade, vintage, and supplies”.I don’t have any grandparents to buy things for. But I did try to buy things do try and buy things on a regular basis on both ebay/craigslist that I need both personally and for business.The amount of people in a year that will interact (from a buying point) on etsy is certain to be magnitudes less than ebay/craigslist. It’s addressing the niche of gift buying. That’s fine and there is a business model there no doubt.Etsy could gain in momentum by striking deals with celebrities to design and sell things on etsy. That would garner the type of “dead bodies”[1] attention that gets notice in the media and drives traffic and adoption.[1] My reference to the early days of the web when the strangest things would show up for sale on ebay/craigslist with the resulting stories going viral in the main stream media and driving traffic to the sites.

      1. CliffElam

        Funny you say “dead bodies” stories because the only etsy links I ever get are to really horrible krep.

      2. ShanaC

        I hate to say this, etsy has a lot of momentum if you are their market. Youa re not their market. I check etsy constantly. I go there for all sorts of things. Do I buy, not always, but still.

        1. fredwilson

          do you like their new mobile apps?

          1. ShanaC

            I honestly don’t use it. I use the mobile website instead. 🙂

  13. Anuj Agarwal

    Absolutely agree. Mobile first is the recipe for failure. Pulse, Flipboard etc will fail, even though they have traction and product market fit.

    1. falicon

      I don’t know if Pulse, Flipboard, etc will fail or not, but if they do I don’t think it will be because of the mobile first approach for them…I think it will be because they are mostly just interfaces to a service someone else controls.They clearly add some unique value to their users, but they aren’t taking over or owning those users in any real way…the only company I know of that has done a reasonable job of this so far is (Howard built a value-add community on top of Twitter, then took over ownership of that conversation and community so much so that the people engaged there no longer care if that part of the conversation is a part of twitter or not anymore)

      1. Anuj Agarwal

        Totally agree with you.

  14. Fraser

    This tweet from Adam Nash is my favorite on this topic:”The answer is not “web-first, mobile-second”. The answer is web for viral conversion, native for organic engagement.”(… )Reading the post, Norby shares two problems the company faces: building a funnel and testing conversion are hard on a phone; and, lack of an engaging product.The later isn’t a mobile issue but if the team doesn’t accept it, it will have an impact on how they perceive the former. Sometimes we are too quick to blame a product problem on distribution woes.

    1. ShanaC

      products usually aren’t engaging because the people behind them don’t know their markets…

  15. Julien

    Another key aspect to consider is that it’s extremely hard to distribute mobile apps at this point, because of giant gate keepers. Everybody has seen this HN post about 25 app makers capturing 50% of all app revenues.The truth is that the mobile web is definetely a lot more closed than the “actual” web (I don’t like the term desktop, because I actually visit many websites from my browser on my phone and tablet. As this point, it is a real challenge for anyone who wants to get in this business and no-one is safe from eventually be kicked out from the platform (either directly or indirectly, thru the deprecation of some APIs, the appearance of a competing app provided by the platform itself… etc.

    1. JimHirshfield

      What’s your alt term for “desktop”?

      1. Julien

        No alt for desktop. I think the difference is in the “app web” (mostly mobile + tablets) and the “browseable web”.

      2. LE

        I think the demarc for all of this is the size of the keyboard. It’s really the size and physicality of the keyboard not the size of the screen.

        1. JimHirshfield

          But it also has to do with touch, as in touch screen, tap, swipe, and how much is visible on screen.

          1. LE

            When you think of the advances that have come in the society as a result of communications advances, I think you will find that things that enable people to convey more information quickly, and to be able to quickly edit that info, you will see that the motions that you have highlighted (tap, swipe) are greatly restricting in what they allow someone to convey. (Which to me is the point that I am making.) Think typewriter vs. the pen. And word processor vs. the typewriter and then add the mouse (as opposed to line editor commands to reposition the cursor). Or even the dictation machine used by attorneys. Ability to rapidly take something in your brain and put it in a form that is useful to someone else (a person, a website etc.). Siri is great for taking rough notes that can be edited but it’s still not accurate enough and no replacement for a keyboard. In fact if you are really good at typing even having to use a mouse can be restricting, it’s easier to use keyboard commands (counter intuitively).Despite people’s recent love of the short form message (sms and twitter and @fakegrimlock ) long form communication has not gone away. There are work arounds and crutches for some of this (for example auto suggest, shortcuts, siri as only a few examples) but I still personally feel that it takes a certain amount of words to get the desired action out of another party (whether you are trying to sell them, convince them, teach them, or gather info from them).

          2. JimHirshfield

            OK, you convinced me. Good points.

    2. fredwilson

      yes. so true.

  16. awaldstein

    Been thinking about this.Apps are pie against the wall. They do one thing and you either hit or miss. Hard to get feedback. Hard to connect personally at a tool booth. Hard to know what’s happening.Start with behavior and value is my mantra.Some behaviors live on mobile. The Check In for example. Some pieces of behavior do, like checking your flight on AA app after buying your ticket at the site while at Starbucks.I think mobile first only if your behavior exists only there. If not, i’m slow and it takes me a bit to understand the pure utility part of a behavior be it transactional or not. I like a bigger net till i know exactly what the bait is.

    1. JimHirshfield

      Why do you think it’s hard to get feedback on mobile apps? Do you mean usage stats? Or user feedback?

      1. awaldstein

        jimI just edited my Disqus comment and got:null….displayed. Annoying.I’ll retype my response later.

        1. JimHirshfield

          Oh, I got your response via email notification. Sorry about your edit problems. On mobile? (yes, I appreciate the irony of that questions).

    2. William Mougayar

      And sometimes mobile works well in tandem with desktop. It doesn’t have to do all what the web does, but the part that makes sense in mobile.I think email is the bridge between Mobile and Desktop. It reminds you of actions you can take on the app, whether mobile or desktop.

      1. awaldstein

        Email as a bridge?Email is a channel in its own right. I don’t see this as a bridge to anything except action really.

        1. William Mougayar

          To action is what i meant. Action can be taken on mobile or desktop. Action is called Notifications on mobile, but it can also be interruption with or without value.

    3. LE

      “Hard to connect personally at a tool booth.”Hard to have sex also at a tool booth. (What’s a tool booth?)

  17. John Revay

    I think Mobile first vs web second also depends on the application. Instagram – absolutely – the phone is the camera.4Sq – yes – I always have my phone when I go places.Others might not be as much mobile first re: a simple calendar app – (screen real estate, people may do planning / scheduling on PC vs mobile device.I think w/ the fast adoption of tablets – this may change.

  18. JimHirshfield

    Oh so many fails on the mobile app front…I don’t know where to begin…so I won’t.

    1. fredwilson

      it might be a 1/100 market.

  19. falicon

    Where ‘mobile first’ works, and why it’s still a relevant mantra, is in forcing you to get at the core of your service and value-add…you don’t get time or room to clutter up the experience, so you’ve got to get down to the core of ‘why’ a user would put this on their mobile device and actually continue to engage with it.Where it’s completely failed and caused massive problems is that so many people mis-interprut the mantra…they think it means to start with the device and design for the device…too many people are saying “hrm, I’ve got this phone, what kind of company can I build around it?”. That’s completely wrong.The few successes are the ones where they had a problem first…and it was a problem that was most easily solved by what mobile (finally) brought to the table.Foursquare wasn’t mobile first…it was a bunch of failed web first attempts that Dens kept trying to hack into workable models…only once mobile (with solid GPS features) came along, was he able to *finally* start providing a version of the solution he had envisioned…

    1. fredwilson

      yes, yes, yes.that is the main takeaway from this disqussion

  20. Eric Greenberg

    I’m really glad to see this post (and the upcoming ones on revenue models). In general, I would have liked to have taken the ‘designed for mobile, built for web’ approach with that is described here, but alas, we were well under way with the web (several years ago). However, over the last 3 iterations of conXt we have steadily improved; Not just the UI and design, but also the *underlying concepts* of the application. This is still so much easier to do on the web. If we had jumped immediately into mobile-first, we would not have had the time, money, and probably patience, to fine-tune what concepts work and which don’t when trying to create a useful address book/family CRM. Thanks for the thoughtful post.

  21. RodGM

    Great Post Fred! Thanks.

  22. Ajay Kulkarni

    Vibhu makes a very dangerous point, because he confuses vision and tactics.Yes, mobile poses more challenges than web: it’s harder to release, to iterate, to A/B test, to close viral loops, etc.But starting a company is hard. Building a product that people want to use on a regular basis (let alone a company) is very difficult. I think we often forget how hard this process is on its own. Most of us are going to fail.But if we fail, it’s usually because we didn’t get the product right, because we didn’t solve the right problem – not because our tactics were slightly harder to execute.Uber solves a real problem. So does Foursquare. Neither would have worked as web apps. They had to be mobile.Find a problem to solve and then choose the right platform.That said, mobile tends to have more low hanging fruit than web. It’s newer, and changing much more rapidly.It’s also worth pointing out that mobile is still new and people are finding ways to make these tactics easier: scanning address books, using SMS and download links or mobile URIs in your viral loops, A/B testing via well placed web views or even across iOS/Android, and more.If something is hard, then figuring it out gives you a concrete competitive advantage.And besides, if this whole process of building a company were easy then we probably wouldn’t be doing it. 🙂

  23. raheeln

    from a UI pov 7inch tablets should in time become the sweetspot. companion web could follow easier from this too

    1. fredwilson

      i love my nexus 7

  24. kirklove

    The thing I like about you most, Fred, is you’re super opinionated yet willing to see both sides and give merit to others even if you don’t agree. You know where I stand based on our conversation about it last week. Great post, man.

    1. LE

      “like about you most, Fred, is you’re super opinionated yet willing to see both sides and give merit to others even if you don’t agree.”One of the keys to successful blogging (or column writing) is to be rigid as it tends to polarize and draw out commenters. Being circumspect (even if you are circumcised) tends to water things down and invite less opinion and provide less value overall. Fred is also interested in what everyone is thinking so it makes sense to follow this strategy.

      1. fredwilson

        yup. i want to see the range of comments. that informs my views. Brandon’s point at the top of this thread is not one i agree with but it is firmly lodged in my brain now as a counter point with great merit

    2. fredwilson

      my mom says i was arguing with her the minute i came out of the womb

  25. Mark Birch

    Context and users…that is what the decision should rest on. I wrote the following in April:One needs to consider the situational context and user experience. Mobile phones are great for those in the moment situations, but not great at digging around activities. When I need the answer for one thing immediately, mobile is great. I need to find a restaurant around me, I pull up Foursquare Explore. Someone asks me about some obscure bit of trivia, a quick search leads me there. I want the latest sports scores, boom it’s in there.When activities involve research and analysis, mobile platforms do not perform well. If you need a last minute hotel booking for the night, then a mobile app is great. If you are planning a family trip and deciding between various options, then a mobile app is not going to cut it. You can check out a LinkedIn profile, but doing a broad review of job candidates not so much. Looking up a contact versus building a dossier of contacts for a sales deal. Finding an address versus searching through real estate listings. Looking up restaurant reviews versus writing a review. Sending a payment versus managing your financials. You probably get the point.The full post if you care to read is here:

  26. Tom Labus

    There’s a lot a “Gold Rush” mentality to mobile and this has lead a lot of people to a place where they’re not going to be finding huge chucks of gold.Mobile has seduced many but without much love in the long run.

    1. $27180517

      I think this gets to the heart of the issue where you’re seeing entrepreneurs try to build an app for a device to get in on the next big thing rather than solve an actual problem

      1. Tom Labus

        I see a lot of activity and use of mobile but is it the best route to make some cash. I don’t know for the long run but it does seem pretty elusive but for a few right now.

    2. fredwilson

      so true. but like all gold rushes, there is gold in those hills.

  27. Joseph Flaherty

    Kickstarter, Pinterest, and Airbnb are interesting counterexamples to the mobile first thinking. They are three of the biggest success stories of the last few years. Two have solid business models and seem to be scaling well. Pinterest at least has a path to one via affiliate fees. Of the three KS doesn’t have an app. Pinterest’s breaks a lot of the key functionality and makes it basically read only. Airbnb has done a really good job considering the mobile vs. desktop uses cases and designing appropriately.Mobile is unquestionably the future, but it’s a little surprising that there hasn’t been a huge, non-game, break away money maker. As cool as Instagram was, it felt like a bit of a panic acquisition. Maybe we have to wait for Square to exit to see the first big proof.

    1. fredwilson

      great points. i think Kickstarter’s mobile apps will kill it when they finally get them out.

  28. Jake McGraw

    The phrase “mobile first” refers to more than iOS/Android apps, also covers an application design methodology. When your product manager or developers come to you and say “mobile first” don’t just think app store, see

  29. markslater

    create real well thought through user stories. this will be the clearest indicator of what you develop.We are all in on mobile first web second. But every business is different.

  30. leapingazelle

    Approaching the design in a responsive way is key, even if you have a native app. As a designer I like to see the big picture first and design the layout for all the possible features. Look at Facebook, look at Pinterest, they did’t start off with mobile first and then a web companion.Even though times are changing, I still believe that web first mobile second is critical because our screens aren’t all 320 pixels, they are 600 pixels, 768 pixels, 1024 pixels and now even at 1280 pixels, which are standard devices people are browsing on. If you only design for a mobile screen, you limit yourself to the possibilities of what you can display on a larger screen because you have spent all your time focusing on mobile-only content.I think taking away features and elements is way easier than adding them. However, it is important to keep in mind an adaptive and responsive layout to make designing your mobile app easier from a web version, even if it ends up being native.

    1. William Mougayar

      Yes, as long as you take a user experience view, and not just a design interface view when going across to mobile.

      1. leapingazelle

        Ofcourse, they both have to be assessed separately. It’s two completely different user experiences.

  31. Luke Chamberlin

    Can we clarify some language? “Native mobile apps first” is very different from “mobile web first”. When people like Luke W ( who literally wrote the book “Mobile First” use that term they are primarily talking about web design, not native app design.PS yes, Luke W is my nemesis

    1. falicon

      Doh! Then I’m sorry to have to admit that I love Luke’s blog (great summary posts on talks/events he attends)…I also enjoyed his (and almost all the other) book you mention.All that being said, I always enjoy your conversations here as well…So can I be Switzerland in this situation? 🙂

      1. Luke Chamberlin

        I appreciate the offer but I’ve been a Luke long enough to know that no room is big enough for two Lukes.

        1. falicon

          OK – we’ll call the other one Leia then… 😉

          1. Luke Chamberlin

            This got weird fast.

          2. falicon

            That’s one of my true talents…many many women have told me as such throughout my life! 😉

  32. Teemu Kurppa

    My idea of the mobile-first strategy (or it’s younger sibling: a tablet-first strategy) is that in it’s core it defines how you approach user interaction design. In the future, most of consumer internet services and a significant number of business tools are primarily used via mobile devices and tablets. In my opinion, this is a safe bet to make and startups starting today should make that bet.However, it doesn’t dictate neither your initial distribution strategies or your platform or technology choices today. For example you design a mobile user experience, but implement it with HTML 5 and Javascript and wrap that implementation to an app for distribution (and payment) purposes. Or you implement that as a native app because at the moment you can create a smoother experience that way. Or you skip the app part and build just an easy mobile web experience to make on-boarding easy. Distribution tactics and technologies change as mobile platforms evolve, but the fact you need to design for mobile usage doesn’t.Our new startup is doing serious data analysis tool, but from the surface it looks like a very simple tool. We believe this simplicity is essential to succeed in the world where our service will be mainly used via tablets and mobile phones.

  33. Allen Lau

    mobile first != mobile only. Mobile and web should work in tandem. That said, I still believe Internet companies, especially consumer Internet companies, should be biased towards mobile / tablet as opposed to web (and we should treat mobile and tablet differently, but that’s another topic for another day). That’s where the puck is going. Where the optimal mobile / web balance should be? It really depends. Foursquare’s optimal balance is different than Kickstarter’s. It is hard to generalize and we shouldn’t generalize.

    1. fredwilson

      i think Kickstarter’s mobile apps (when we get them) will be awesome.wattpad has played this very well Allen

  34. brian trautschold

    this is extremely interesting per the abundance of ‘mobile-first’ technologies… i think there is a nice hybrid somewhere in between with some order of the hit mobile (super virality) , quickly hit web (better experience/ user value maybe), followed by test/ measure/ iterate with both – keeping the soul of the UX for each device is tough, but I think there are enough good examples floating around now to learn from…Pinterest comes to mind…

  35. Steve Lowing

    In my experience, there is a simple question a company needs to ask themselves when developing any service: “Where are my users when they are accessing my site?”If their user base is highly mobile and not sitting at a desk or have a laptop at the ready then its very likely having a mobile app is not just the right way to go but a “must have.” Likewise if you just need a subset of functionality on a particular platform but full capabilities on another, then that is likely the right strategy.I think developing for the “next wave” or where the growth or future is misses the point.You need to know your users to maximize eye balls.

  36. Abdallah Al-Hakim

    I think within mobile there is an important distinction between mobile phone and mobile tablet. It is my understanding that tablets are growing in popularity and might become the primary mobile device in accessing the web. This is important for those developing a web first strategy because the experience of browsing on the tablet is similar to the desktop. It is still important to have a nice UI for a mobile phone experience but developing a useful interface that works well on desktop and tablets might be more important for some businesses. Obviously companies such as Uber and Instagram are more geared towards smart phone users and as such they developed a mobile phone strategy first.

    1. fredwilson

      we have seen tablet only apps really struggle to build users

  37. Dave Pinsen

    Agreed that this was a thought provoking post. One issue I had with it though is that it conflates two issues: web versus mobile and paid versus free/ad-supported. Norby says free mobile doesn’t work because the cost of acquiring a new customer via Google ads is $20, and that’s more than you’ll collect in ad revenue per user. But it’s not necessarily more than you could collect in sales (up front + in-app purchases/subscriptions) on a paid app. I think the bigger issue might be another one Norby raises, about the efficacy of Google ads when you’re targeting a higher-income, more-educated pool of prospective clients.

    1. falicon

      I think the $20 cost was relative to his experience…the challenge is that, with a paid app (as you’ve probably experienced) the CPC goes up and conversion rate (to profitable) actually goes down…It takes a lot of work and experimenting to find the right price point for making that engine purr…many simply don’t have the resources (time or money) to put into properly figuring it out for their situation.

      1. Dave Pinsen

        You’re probably right that CPC goes up and conversion rates go down with ads for paid apps. I haven’t figured out how to profitably advertise my paid app yet, though I haven’t spent much time or money trying. Perhaps I’ll spend a little more. But I think my general point stands: it’s at least theoretically possible to make more money from selling a paid app than it costs to advertise it. That doesn’t seem to be the case with ad-supported apps, if you consider that a monster free service such as Facebook is only making ~$6 per user per year in ad revenue (I don’t know how that picture changes if you just look at its mobile app). Of course, Facebook grew virally, so its cost of acquiring users must have been negligible.In any case, how can one make money at this? I can think of two ways:1) Make something free that spreads virally, and then sell it to a bigger company.2) Charge money for something, and spend less than that to acquire users. I was able to do that for a while by writing Seeking Alpha articles, which were free, effective advertising, but I suppose it’s also possible to do this by paying for advertising, though I haven’t figured out how yet. Maybe iAd for Developers might be effective? Apple says the CPC is $0.25 domestic and $0.30 foreign, but I don’t know about its conversion rates or how precisely they target the ads (they apparently do that for you).

    2. fredwilson

      i agree with your criticism but i didn’t want to get into that.thanks for encouraging me to write about Vibhu’s post

      1. Dave Pinsen

        No problem. I’ll throw another post idea out there for you. You tweeted a link to the BI post about Mary Meeker’s State of the Web paper. I wonder what you think about This comment on it. Brief excerpt:”Asset light as she calls it.. is just code for the deflationary death spiral that the US has been going on since 2008. folks can’t afford cars so they use ride sharing services, folks can’t afford their mortgages so they use airbnb to bring in extra cash…”In addition to reading your take on that, it would be interesting to see the Kid weigh in in the comments.

        1. fredwilson

          great suggestion

          1. Dave Pinsen

            Glad you liked it. BTW, just sent you an email with “Fintech Innovation Lab” in the subject heading. I realized as soon as I clicked send that that subject line probably wasn’t the best choice, given your old advice about emailing you. So I thought I’d give you a heads up about it here in the hopes you might get a chance to read the email anyway. It’s a quick follow up question about a conversation we had earlier this year (about discount brokers that might be interested in hearing about Portfolio Armor).

          2. fredwilson

            I saw it and just replied

          3. Dave Pinsen

            Just saw that. Thanks again.

  38. Vinay Pai

    I don’t entirely agree with the “mobile is where the growth is” mantra. The number of desktop users etc. may not be growing in the US and the rest of the developed world, but that’s only relevant if you are Google or Facebook or some other player who has pretty much saturated the market. If you have 1% penetration it do you really care whether the platform as a whole is growing or not? You have bigger hurdles to cross first.Likewise, I disagree with “If you design for the web and then port to mobile, you will find that it is really hard to fit your UI onto the small screen.” The reverse also holds true. If you build an interface designed around a 4″ touchscreen and clumsily port that to a website where you have vastly more real estate, but a keyboard-and-mouse interface, you’re going to have poor results. The problem is with clumsy poorly thought ports, not with which one you do first.Just start with the one that makes the most sense for your application. On one extreme, something like Foursquare clearly makes way more sense on mobile, something involving complex data input like Google Docs makse sense on the web first. Then design the companion website or app respectively not by force-fitting the other UX onto it, but from the ground up.

    1. fredwilson

      it is tougher to take share in a market that is not growing

      1. Vinay Pai

        Do you care to articulate why you feel that way?I think it’s more a function of a /smaller/ early market, when our personal friend, family, and friend-of-friend networks represent a larger percentage that take a significant chunk of the market, show up on “best of” lists and so on, and grow organically from there. Under those circumstances, it could make sense to sacrifice overall user experience for a shot at the new world. I would argue that we are well past that point with mobile, so it’s time to take a step back to first principles and figure out which experience is /actually/ better for any particular application and go from there.

        1. fredwilson

          Good blog post request

          1. Vinay Pai

            I look forward to reading about it in a future blog post!

  39. Orhan Sönmez

    I actually don’t understand what the confusion is all about, isn’t it just like about everything in life: there is no fix, right way to accomplish something?In my opinion, it depends heavily on WHAT you want to build and that you have to figure out which strategy works best for you.There a enough examples to prove Vibhu’s point (web first vs. mobile second) right, but also more than enough to prove Fred’s (mobile-first vs. web-second).Instagram would have NEVER worked out, if you had to upload pictures from your hard-drive on a webpage, to apply filters to them and share them with your network. In fact, there were already a lot of services doing exactly this way before Instagram, but they failed to gain serious traction, because it was simply too inconvenient to use the service.Whereas other services, such as Airbnb, Dropbox or even Facebook did right with their web-first approach. Booking a room for your trip next week, uploading (large) files or consuming a huge load of information about what is going on in your social circles is just so much better on a full desktop environment, if not simply because the screen is bigger and you can view more information at once or your bandwidth connection is usually better than your 3G/4G connection on your phone.TL;DR: I don’t think that there is a right or wrong way to start your service/business (mobile-first vs. web-first), and I don’t think that there should be either. We shouldn’t believe blindly on a mantra that is imposed on us and squeeze or own service onto that boundaries.If their is a mantra that you should adhere to, that would be: Think first. Try to think about (or ask early adopters) how your users will want to use your service? Do you rely on usage-scenarios that are way more convenient on the phone (esp. if your service is even more adapt to the usage of GPS, cameras or other sensors that mobile devices brought along to mainstream usage).

  40. howardlindzon

    Thanks Fred….in financial web, it seems like mobile is all about price, volume, chart, news and watchlists and even mary Meeker neglected the topic in asset light.At Stocktwits we are thinking of all the ways our data and sentiment can be uniquely displayed and integrated in mobile first ways.That said, the desktop still dominates the trading world for now.

    1. William Mougayar

      Yup. It’s a ying and yang. One serves the other. It’s not always first, second. It’s together.

    2. fredwilson

      but it won’t forever

  41. falicon

    BTW – one of the real hidden gems in Vibhu’s post is this truth:”If you paid Google’s $1 CPC for people to enter your funnel, you’re really paying $20 per user and you will never recoup that cost.”I learned this many moons ago with my draftwizard subscription service (killed it the 1st year mostly because of being early to adwords, after that everyone caught on and the CPC exploded to a level that made even just breaking even a HUGE challenge [especially if you don’t have a high customer lifetime value to recoup those costs with])

    1. fredwilson

      yup. dave pinsen learned that too and he was the person who encouraged me to talk about Vibhu’s post

  42. howardlindzon

    I shared with team videolicious because they are a perfect example of a company that has to deeply think about how mobile and web experinece meld together for customers. thx.

    1. fredwilson

      do they have an android app?

  43. andyidsinga

    there is another really cool thing about building a mobile _web_ app — every time someone clicks one of yoir links inside a mobile native app, one that embeds the native web view of that system ( ex. twitter, flipboard etc ) your web code app can still work, launch, be seen, interacted with etc …and you didnt even have to convince twitter, flipboard etc to embed your app!it seems to this sort of indirect embedding is a powerful cool feature of the emerging web.

  44. Matt Mattox

    “First, I think you can’t abandon mobile.”I didn’t see where Vibhu suggested that Origami was abandoning mobile. This would be a very different debate if that were true.

  45. Semil Shah

    Hi Fred, I’ve met a good chunk of startups who have all raised Series A (or are about to) who’ve used the web to hack mobile growth:

    1. William Mougayar

      There is so much of a fragmentation of attention in mobile apps that if your app isn’t inherently viral, it’s very tough to grow the user base.

    2. fredwilson

      i think that makes a ton of sense. it was one of instagram’s best moves.

  46. William Mougayar

    Take Foursquare, the quintessential Mobile company. Maybe their end-game isn’t so mobile. Maybe their monetization is web-based via search or other B2B value. Monetizing the mobile stream with ads or sponsored content on its own merits might be tough.Look at Instagram, the 2nd quintessential Mobile company. They have hooked-up with Facebook because Facebook is their back-end repository. Instagram is just the picture collection app. Mobile in this case serves the web app.So the question isn’t whether mobile or Web is first, but rather: Does Mobile serve the Web, or does the Web serve Mobile?

  47. Matt

    It depends on your business model and goals. Mobile first is great for first to market strategy and quick capital gains but if you’re looking for overall long term growth without a quick exit, then web first is the way to go. Mobile usally costs more today because consumers are demanding more features than ever before…

  48. JordanCooper_NYC

    Future is mobile and mobile first, but starting to believe it ain’t native…have deep thoughts if you want to discuss lmk

    1. Carl Rahn Griffith

      Unless explicitly games for example, native device/apps development is an overhead that makes little sense for most general business instances – I’ve been doing some research on this lately and there’s very good ways to address develop once/deploy to any device in an optimised way. Mobile obsessions should not blind pragmatism. Most of us are here to make a profit for our business, not evangelise mobile as a device. It’s a clever ‘phone, not a talisman to be worshipped, at all cost.

    2. fredwilson

      meaning it will not be html5?

  49. RudyC

    I think the internet community has been since the advent of FaceBook chasing the masses. Usage numbers drastically increased during this time and suddenly all you needed was quantity or quality. Instead of insiders using their heads they chased numbers.Mobile at this point should be separated into 2 parts, smartphone and tablet. Interesting enough I feel as if the tablet will solve a lot if not most of the problems being discussed in this column. Smartphone apps have always had a limited application so they became more features than anything else. With the exception of Pandora and Facebook I don’t really see anything else that has held people’s attention longer than 2 or 3 months.My guy feeling is that the tablet will eventually replace the smartphone and solve a majority of the problems faced by programmers. The tablet as it is right now is just a 3rd device used in homes as mobile, using a wi-fi feature.Eventually the phone will be combined with the tablet, enabling people just to have one device connected to the internet at all times at a affordable price..

  50. Brett Hellman

    Another reminder to take all advice as advice and not fact. Luckily at Hall we thought this through and learned for us what was more important was Desktop First and we weren’t shy about it…Desktop First:

  51. Carl Rahn Griffith

    Ironic timing:So, as far as two of the best social mobile apps are concerned, war is (formally) declared…http://bits.blogs.nytimes.c…Lead with text and maybe a picture, or vice versa – you have to take your pic (sic), now, I guess. It suddenly became very partisan. Inevitable, I guess.

    1. fredwilson

      fuck. that really sucks for everyone, including both of them

      1. William Mougayar

        It’s back to being just a link before Twitter Cards, so Instagram is asserting that they are the photos rainmakers. Really, I’d like a link to go to Instagram so I can manage my photos there. Twitter doesn’t let me manage my photos. This move affects more the photos Apps like PictureLife, MyShoeBox, etc.

  52. Matt A. Myers

    You need to provide value through your mobile offerings, though you need to design first for desktop – while keeping mobile in mind. For most services, mobile facilitates or complements the full cycle of use, as people now have multi-device use in their daily habits / routines.

  53. Peter Cranstone

    RE: If you design for the web and then port to mobile, you will find that it is really hard to fit your UI onto the small screen. Better to design for mobile first and then build a web companionSorry – that’s nonsense. You can seamlessly extend the web to ANY mobile device via the browser. The CORE issue to overcome is that the browser doesn’t have access to native OS API’s like an app, so the argument becomes what you’ve stated above.Except that’s complete nonsense. The browser can easily talk to Native OS API’s and it can do while extending better Privacy and Security to the user. No Enterprise is going to sit around and build ‘apps for everything’ anymore. The form factors and device capabilities are changing too fast.What’s needed (and has already been invented) is a way to share the device capabilities with a web server in real time so you know EXACTLY what to deliver over the web to the browser.There is only one web – you have a lot of different devices connecting to it. All you need to do is figure out how to transmit those ‘DEV and TERM’ capabilities to the server so the appropriate page can be delivered.Doing anything else is simply wasting resources on something that is guaranteed to change in 6-12 months.

  54. daryn

    I don’t like the terminology web v. mobile, as I think mobile web is really the most important piece for a lot of traditionally “web” businesses.With regard to native mobile apps, I think that besides for a few very specific uses and your core apps, they will go the way of the desktop app over time. There’s tons of desktop software still made and available these days, but how many apps do you really use? Compare that to the number of web sites you visit…The reason mobile apps are hot right now is because in general the experience is so much smoother than mobile web. And you get that valuable “bookmark” on people’s home screen. But over time, just like on the desktop, the capabilities within the browser will gain enough parity that the points Vibhu mentions, and not having to get over the installation hurdle, will push development back to making sure the web experience is great for all users rather than the web is great for desktop users and the app is great for mobile users.

    1. fredwilson

      mobile web is great for consuming. but not for engaging.

      1. John Revay


  55. mgwitham

    Thanks for demonstrating how to appropriately argue conflicting viewpoints. You’ve shown us the Venn Diagram of two valid arguments; where they overlap, and contrast.Very thoughtful & respectful. Kudos.

    1. fredwilson


  56. Michelle Sun

    Mobile is growing at a much faster rate than web, and because venture capitalists are betting on high growth, mobile first makes sense for startups that are looking to raise capital.For majority of products, web is still dominant. Even in the US, high speed mobile connection in main cities are still prohibitively expensive (I live in California). Many people spend 2+ hours on commute during the week, then 8 hours at work, how much time is left on the road?

    1. ShanaC

      Who knows

  57. ShanaC

    Why are we talking like this? I don’t get it at all.Mobile is another personal computer with a different form factor. That is it. The most successful, nongame tools are not either mobile or laptop/computer.Think evernote: it is ubiquitous. Wunderlist. Drive, Dropbox. Ubiqutous.The things that make these tools powerful: they function well irrespective of operating system and formfactor. They’re adaptive.The best products are adaptive. Not mobile, not internet. Adaptive.

    1. Donna Brewington White

      Well said.And will I lose my AVC pass if I admit that I just looked at Evernote for the first time today?.Your comment finally tipped the scale. How am I living without this. Oy.

      1. John Revay

        I started using evernote about a week ago

    2. fredwilson

      no, no, nomobile is very different than the desktop webdistribution works differentlyUI works differentlyif you think they are the same, you will fail on mobile

      1. William Mougayar

        In other words, you’re saying that Mobile is NOT a continuum for the Web, but is rather a disruption to it? I think it depends on the Application. Some take Mobile like duck to water & thrive, while others get a modest benefit from it.

      2. ShanaC

        just because distribution works differently doesn’t mean that the apps that function well and gain mass tracation are effectively platform agnostic when it comes to functionality.

  58. Dave W Baldwin

    Good post and intersting comments.My suggestion (I’m assuming this is what Fred is trying to communicate) is not to think either/or, but where you product best fits along with the primary intended revenue model.Just because something will work best on mobile doesn’t mean all things must be mobile. At the same time, it depends on the evolving method of delivery to constomer’s question/interest/need. It is going to change from the “make it fit small, then enlarge” scenario, which of course, is logical at this stage. Think today/tomorrow vs. today/yesterday.

  59. Sebastian Wain

    There are areas where you can’t go web first: editing content. If you want to do a WYSIWYG text editor, an spreadsheet, etc you must first develop for a specific mobile platform. That is because the current state of HTML5 on the mobile is buggy in terms of content edition (contentEditable, designMode). If you want to go web first you must workaround a lot of issues between different devices and browsers that probably will change in the next iteration because this part of the HTML5 is not stable.

    1. fredwilson

      great point. we need to fix that.

  60. Leonid S. Knyshov

    What annoys me with mobile web is the friction for Facebook oAuth. Even though the user is usually logged into native Facebook app on the device, he is logged out in the mobile web browser.I implemented oAuth dialog for Facebook in my mobile web app, and it just takes too many steps. In many cases, people are not already logged into Facebook as they are on their desktop, so that requires typing in Facebook username and password.The same is true for just “liking” a page or clicking through my mobile app.According to this:… it’s potentially achievable now. I will be testing more.

    1. fredwilson

      big problem. we need to fix that.

  61. Steve Haney

    It’s not just mobile first or web first or anything else first. It’s about building an online service, and then building various “windows” into that service that best serve your customer. The most popular of these “windows” right now are PC desktop / web, tablet, and smartphone; there could be many more depending on the company and it’s goals (e.g. retail kiosk, TV screen, etc.) Each of those screens employs different degrees of interaction and functionality that best serve their usage model. Which one of those “windows” you build first is a go-to-market question, not a technology question and not necessarily a business model question. What really matters is the service up in the cloud and the job the user hires it to do. Once you know that, how to initially deliver it to users could well be the desktop, and not the smartphone. It all depends.At the end of the day, yes, you’ll want to be on all appropriate screens, but building for the smartphone first isn’t a hard and fast rule. IMHO.

  62. arustgi

    The key thing to understand is to understand and appreciate the difference between mobile context and mobile interface.”And if you don’t design your products and services for what is rapidly becoming the dominant UI,”… I appreciate that and I agree with this. However, if the sole criteria for your going mobile first, is because of the mobile interface and users spending a lot of time on the mobile interface, you will find it a struggle, as far as your business is concerned.However, mobile context brings up a large set of opportunities. The companies that have address mobile first because of the context ‘Uber’, ‘Instagram’, ‘Foursquare’, are “successful” (assuming we are talking about success in adoption. The verdict for success as a business is still out.).Finally, there is also the consideration that a large part of the worlds population will only ever have a mobile phone. Building creative mobile based solutions for that segment of the world’s population will aways be an attractive opportunity (both as a business and making a difference in the world). Case in point being m-pesa and other mobile payment based systems…

    1. fredwilson

      great comment Aditya

  63. Bill Patrianakos

    This all makes sense if the app in question makes the most sense in the context of mobile. Uber absolutely needs to be mobile first. People are often on the go when they decide they need a cab. However the decision as to whether to go mobile first or not should be made on a case by case basis. In most cases mobile and web are necessary but each app must decide whether mobile or web first makes sense. I don’t think you could agree or disagree with Vibhu if you wanted to because we’re not talking about exactly the same things here. Mobile first is important when its important and web first is important…. when its important. Both are almost always necessary but the order in which one builds differs. I don’t understand how the author can talk about all the great points Vibhu makes then say “I think it is critical to design for mobile first”. It’s only critical when it’s critical if you know what I mean. Abondoning mobile is suicide these days but going mobile first, while often the best option, is not ALWAYS the best option. I know I sound like a broken record but everyone is right, it’s just that you have to know when mobile first is the right strategy (but mobile should still always be part of the strategy regardless).

    1. MogulLabs

      Totally agree. Properly designing for mobile is about conisdering it as the context in which the user is operating. A key effect of this that a users priorities are re-ordered, e.g. speed becomes more important than choice.

  64. khuramh

    In my view it depends on the particular dynamics of each company. If you have plenty of run way and you have raised enough money to take the time to make a good product, create a great mobile experience and have significant customer validation (from real users) then it makes sense. Do not try the lean startup approach on mobile. If on the other hand if you do not have as much cash to get to your next set of metrics then I would always start with the web. Fast iteration, faster testing and quick go to market. Usually you get a few chances to get it right. This won’t happen in mobile you only have one chance to get it right so you need to spend a lot more time in the drawing room getting things right without any investor pressure. I am hoping smart investors would understand this.

  65. Matt Straz

    We’re doing enterprise HR and it would have been insane to do mobile first — for a whole number of reasons. Now that the web app is built, though, we can present key data via mobile apps.

    1. fredwilson

      i hope i qualified this post with the consumer preface somewhere

  66. Anuj Agarwal

    Problem is – Most founders are not empathetic to mainstream usersI think most founders in Silicon Valley are geeks. They are power users. They are highly sophisticated users. They hardly use desktop/laptop because they own iPad2,iPad3, iPad mini, Nexus phone, iPhone, Kindle etc etc. If not all of these devices but atleast 3 or 4 of these. It is not their fault that they are not empathetic to mainstream users. Mainstream users lifestyle is totally different. They do not own so many devices. Its very hard for a power user(founder) who spends most of his time, socially and at work, with other power users. The press who cover their startup is also a power user. How can these people make a product for mainstream users?

    1. fredwilson

      that partially explains why everyone builds for iOS when Android is the dominant mobile OS

      1. Anuj Agarwal

        Fred, im reinventing feed reader. Just like disqus reinvented comments, instagram reinvented photo sharing, im rethinking rss reader and building an entire platform from scratch. I wish to show you this thing im building.

        1. fredwilson

          Can I try it? Does it run on android?

          1. Anuj Agarwal

            Yes. Its Building web app first. Next is Android app.

          2. fredwilson

            i would like to try the android version when its ready. i am not a fan of web feed readers

          3. Anuj Agarwal

            Ok. will follow up.

          4. Anuj Agarwal

            Fred, Buzzfeed published a great story today. I thought i should share with you because it is relevant to our conversation.…They say big companies leave holes, startups fill them, startups become big companies and the cycle goes on. My mission is to fill the hole google left in one of their most respected product, Google Reader, and reinvent this space alltogether.

          5. fredwilson

            The question you need to think hard about is whether people want to subscribe to content and consume in an inbox like experience which makes you feel like you are behind in your reading

          6. Anuj Agarwal

            I agree. the old reader was a inbox like experience. and im changing this by adding a social layer so that users also have access to the roar of the crowd. By design. secondly, data says lot of people still prefer reading news in a rss reader. over 10million rss reader related searches performed globally on google. long back you told me in another comment conversation that even your blog gets good traffic from rss readers. I think mainstream users will start using rss reader once it has a social layer. They are bored with FB. Already 17k mainstream users are using feedspot. though small but we just started.

      2. Sean Hull

        I would argue it is just much more complex to build for Android, and get it right. A myriad of hardware platforms, running different & customized versions of the OS also makes it extremely difficult to properly test for all those devices.I wrote about this – Why The Android Ecosystem is Broken:http://us1.campaign-archive

        1. fredwilson

          just because its hard doesn’t mean you shouldn’t do it

          1. Sean Hull

            True enough. But virtue may not be part of every company’s balance sheet.

  67. anne weiler

    Didn’t have time to read the other 240 comments—late to this party so someone may have said this, but what about features you might need on the OS, security in particular or the ability to use OS features like the camera? Instagram needed the camera so they couldn’t be a web app (at least without a lot of other steps)

  68. Zaid

    First, I think what you really mean by “mobile first” is “app first” and what you really mean by “web first” is “browser first.”Second, let’s take your own blog. At least on my own iPhone 4, it is incredibly difficult to use and never completes loading. I am typing this comment in Let’s take your own blog. At least on my own iPhone 4, it is incredibly difficult to use and never completes loading. I am typing this comment in my Notes app in hopes that when I do get to focus on the laggy text area, I can quickly make this comment. You, and many many other folks, DO need to go mobile. But not with a mobile app. Finally, we are in the business of building mobile sites for doctors. Naturally one of the first questions that gets asked is “why do you guys not make apps” and our answer is “because you don’t need an app.” The point is more easily made when we put their mobile traffic stats in front of them to show the dire need for a mobile site. I think the same applies to many many people who jump on the App bandwagon instead of a mobile friendly site. Mobile is not apps. If it helps, think of your website as your most popular app.

    1. fredwilson

      i want to fix that. it is a problem for AVC

  69. Pete Griffiths

    Related to Vibhu’s excellent piece is the following:…The gist of this article is that a recent study detailed the incredible concentration of revenue in the hands of an extremely small number of companies.The concentration of anything that could reasonably be described as success is striking. Notably, ‘Half of all App Store Revenue goes to just 25 Developers” and “All but one of the 25 developers were gaming companies…” (the exception was Pandora).In other words the prospect of a breakthru is miserably low. This is of course consistent with Fred’s point about the difficulties but it puts a little more gloomy flesh on the bones of Fred’s argument.

  70. Darragh Browne

    How about web first with a mobile optimised, responsive design?This gives gives you all the goodness of a/b testing, continuous deployment etc, while also optimising for a mobile experience.Then focus on building the mobile app when lessons & optimisations are learnt from the web.

    1. fredwilson

      i think that is smart. but it still takes time to build that mobile app once you know what you want to build

      1. Darragh Browne

        True, but at least you will be building the right app with the right features.

    2. William Mougayar

      That’s exactly what we have done. We will learn from our html5 app, while we can continuously iterate on it, and we are able to get to the market faster.

  71. Nico Lumma

    I am all for Mobile first, but it a) fundamentally depends on the product and b) should take into account tablets as well. For lots of products that we considered to be mobile first, web second, we will actually find out that they are tablet first, mobile second, web later.For now, I think it is fundamental to have a great UI/UX on mobile and then go from there. That’s how we did it for Stuffle ( and of course we included the viral hooks Facebook has to offer.

  72. Don Holloway

    Made me think … again. I decided that for most entrepreneurs, prioritization is really the key to what you’re talking about. If you think about mobile as a distribution channel, then it is one dominated by extremely large players. So from that perspective, you are “betting the farm” on your ability to get the support of Apple, Verizon, or other companies in strategic positions. Win big, or go home. And get in really early because it takes years for them to launch.

  73. Murray McKercher

    To quote a famous thinker ” The Media is the Message” is mobile – phones – network- software – apps – the media or the mesasge or both?

  74. Trevor Doerksen

    Great conversation. Web is often, at best, an acquaintance to mobile. In a lot of cases a distant acquaintance. It’s like they don’t even remember each others names.Done well, web is a companion to my home and office life. Mobile done well is not only a companion – it’s a buddy. Love you man.

  75. stevenwillmott

    Interesting & awesome comments. To me it comes down to one thing: businesses need to be platforms and not UIs.

  76. Mike

    I prefer to build for desktop and once the mobile site renders readable text and full screen imagery then that’s fine – for a mobile user. For a desktop user I can focus on design more and create greater impact for increasingly wider screen sizes. If my website can deliver a higher conversion rate on desktop even at 30% of the future market then that’s good for me!

  77. firesky

    I prefer to build for desktop and once the mobile site renders readable text and full screen imagery then that’s fine – for a mobile user. For a desktop user I can focus on design more and create greater impact for increasingly wider screen sizes. If my website can deliver a higher conversion rate on desktop even at 30% of the future market then that’s good for me!

  78. Andy Jagoe

    Mobile first vs web first is a false choice and the wrong framing to focus on. The most successful consumer services build a relationship with users where trust and value increase continually over time. Starting small/free reduces friction and increases acceleration. A critical component is getting permission from users to contact them and having a good reason for doing so, ideally as a byproduct of usage.

  79. gwynnek

    Mobile first makes sense, depending on your definition of mobile. Looking forward, mobile isn’t about the device, it’s about the service. Your service offering needs to be mobile–delivered anytime anywhere on any device in the context of your customer. This means, look upstream at data and creating modular services so you are able to deliver in many channels. Design with flexibility and portability in mind, then work on the presentation layer.

  80. OurielOhayon

    I am late in commenting here but I am with you fred on this. Mobile first is a requirement. The reality is that the very large majority of app developers are unaware of the tools existing to do testing tracking closing loops attributing life time value. This is why the perception it is hard is out there in the wild. Regardless not every developer will make it. The reality is that most developers will not make it. Why? Because most apps are not that good and that in this Darwinian world only the good nd the best are surviving. And those know how to build great apps with the right marketing tools

  81. csertoglu

    This post triggered a long-planned blog post:…and for continuity, here’s the full text:It feels like the overarching theme in technology startups in 2012 was “mobile”, with mobile wisdom and sexy statistics echoing in the news, blogs and conferences. The discussion was marked with numerous milestones, such as the successes of Angry Birds, Instagram, Uber, Spotify, Hotel Tonight, and most recently, the proclamations of eBayand Etsy about how mobile has arrived in the holiday season. We have VCs that put “mobile” on investment strategy slides, and discussions by entrepreneurs about the challenges of mobile development. Thinkers like Fred Wilson are also weighing in with their experiences.I personally find the discussion quite frustrating. It is not clear to me what mobile really is. It used to be a clear distinction: mobile was your phone and web was (accessed through) your computer. This has now changed. We have high-powered computers in our pockets, with improving form-factors (iPhones and most Android phones have large enough screens and pinch zooming) and innovative interfaces. Add to this the tablet, which, in my experience, really belongs on the coffee table at home and the senior executives briefcases in the office.At the product level, here’s further confusion around apps and mobile web. To me, it feels like apps emerged due to the frustration around mobile data speed. We’d seen this on PCs before. The need goes away once the web is fast enough (just ask However, Apple and Google love to get in between us, and the tools we use, and are continuing to encourage the app world. I continue to believe native apps break the web and am counting the days until they go away. (This last point excludes the apps that are obviously native-appropriate, such as mobile games, and pure mobile needs like Moped, Wunderlist, Uber, Instagram, etc.)It will be a good day when the whole noise around “mobile” starts to go away. To me, there are unmet needs and points of friction, and the current paradigm (that Meeker explains so well, especially with Slide 9, which is the reason for this post) with proliferation of diverse mobile devices, will inspire brilliant engineers to come up with products and services, that make use of these powerful devices all around us, irrespective of the form factor, that are connected to the sky with a fat pipe.Through this process, some startups will have to focus on one type of device, and others, another. Lately, the challenge has grown for many startups because they feel compelled to parallel develop on and optimize their product for multiple device types. The ante at the table is now a great website, a great mobile site, and iOS and Android apps. In most markets, the competition is fierce enough that under investing in any of these may cost valuable early traction. If you have ever run a product team, you know how difficult it is to do this.And, it is this challenge that will move us away from multiple development needs for different device types. Engineers will create frameworks that will simplify this task, and we’ll go back to simpler development models (remember X-browser optimization challenges. That’s history now). To conclude, I think a lot of the current mobile discussion in startup-land these days is noise. We are just in transitional period with multi-device product paths. However, the core challenges around product–market fit are similar to what they have always been. If anything, the ways you can solve a problem today are more numerous, with the immense processing power we now have in our pockets and the fattening pipes to the sky.It is a great time to be focusing on what problem you want to solve next, rather than “what your mobile strategy is”.

  82. Mark Birch

    All very sound points. Users and context are a mantra I have been touting for some time when the mobile first chorus became deafening. I focus my investing on B2B tech and a mobile first strategy for many problem areas would be a loser for most audiences. The second issue for enterprises is security, which is extreme tricky and porous in mobile. Therefore the need web first and that will be the case for a few more years.

  83. fredwilson

    android tablets are and will continue to be the cheapest with the most form factor options. they should consider them before iPad

  84. kidmercury

    i do this as well.

  85. Brandon Burns

    i use foursquare for location search on both mobile and desktop. i tend to get better results, and i like the ability to just type in “oysters” and get a place i want. but that’s just me. i used to use google maps for that, too, until foursquare got better.i guess the “go back and repeat” part of the typical search experience is cut out for me, because i get it on the first try when searching places on foursquare.i think the thing i’m seeing here, which aweissman also pointed out with youtube / music “search,” is that people are moving away from search engine products to execute the activity of searching. and that’s interesting.

  86. Carl Rahn Griffith

    I got so tired of Foursquare so often being unable to geo-locate itself I have pretty much stopped using it. Google is always there, no delay.So, Google it is.Waiting for a response on a desktop is bad enough – on a mobile it’s a killer (in a bad sense!).

  87. falicon

    I think it’s a movement about contextual search…On mobile, when I search, I want the results to be aware and reflect that I’m on mobile…show me results that will be mobile friendly….skew towards places near me right now…skew towards apps I might want to install.On sites, skew towards the conversation, the opinion, and the community that’s active within that site (I know, I know…I talk about this one too much)….more and more, just like within social networks and groups, I find people are looking for more meaning, context, and niche…they’ll deal with general search (and there are still many use cases for it), but often they are getting more pleasurable results when they start with a bit of context…Looking for a video to watch for a few seconds? Search youtube…looking for a great conversation around equity? Search equity on (sorry – *had* to work in at least one)…looking for the history of a given person/event? Search on wikipedia.Can google find these things? Yes (and often fairly well), but do you get the same happiness from that search experience? Is it the most efficient use of your time? How much *more* work do you have to do to get at the meat for your specific intent?

  88. falicon

    Perhaps if I had a better, more clear, explanation put together it would have been better to link to the about page…yet another part I’m still working on… 😉

  89. ShanaC

    vertical is unsustainable outside of small niches. hence why google happened in the first place

  90. ShanaC

    It isn’t a or b. I want local or my version of local – my version of local includes my community. In fact, a number of people have picking up on this fact in urban studies. Local in terms of location isn’t enough – people define local by who they are with in that location.Which means foursquare is sitting on a lot of interesting data.

  91. CJ

    With Google Now it’s getting better and better every day. But I agree, currently I have an individual app for just about everything you mentioned.

  92. CJ

    I’m having that same issue with 4Sq lately. I couldn’t checkin at all on my London trip because 4Sq could never find itself even though there were loads of wifi around and I had a good local roaming signal.

  93. fredwilson

    yeah, and i was in a meeting from 8am to 10pm. just getting into the comments now

  94. falicon

    that is officially a horrible job you have there…

  95. AbbeyPost

    Unless you’re Twitter or Instagram/FB.

  96. AbbeyPost

    That is a great point, Shana.

  97. fredwilson

    yeah, fuck that

  98. Mark Essel

    did you and Cynthia see the chrome extension that pulls images back in? Don’t have the link handy but retweeted it yesterday. checking…

  99. fredwilson

    I use twitter almost exclusively on mobile