The Mobile Downturn

I woke up to this post by Chris Dixon in my Tumblr. Chris states:

Apps have a rich-get-richer dynamic that favors the status quo over new innovations.

He’s right about that and he’s right about a bunch of other things in his post as well.

We did a portfolio review at USV yesterday. We spent two hours staring at our portfolios on the big screen in our main conference room. We have three early stage vintages that are fully invested, 2004, 2008, and 2012. We talked about the macro environment during which we invested and developed these three portfolios and there was a palpable sense that the wide open period of innovation during which we invested the 2004 and 2008 funds was not as present in the 2012 fund investment period. We have a bunch of great investments in the 2012 fund, but we took safer bets and the result will likely be more wins and less strikeouts but also less home runs. Did we change our appetite for risk in our most recent early stage fund? It wasn’t conscious. We didn’t change our investment strategy. We didn’t tell each other that we wanted to take less risk. But it sure feels like we did.

My partner Brad hypothesized that it had something with the rise of native mobile apps as the dominant go to market strategy for large networks in the past four to five years. So Brian pulled out his iPhone and I pulled out my Android and we took at trip through the top 200 apps on our respective app stores. And there were mighty few venture backed businesses that were started in the past three years on those lists. It has gotten harder, not easier, to innovate on the Internet with the smartphone emerging as the platform of choice vs the desktop browser.

Chris mentions something else in his post that, to me, is a big bright spot on the horizon:

Most worrisome: they reject entire classes of apps without stated reasons or allowing for recourseย (e.g. Apple has rejected all apps related to Bitcoin)

Let’s go back and revisit the big innovations on the commercial Internet over the past twenty years. TCP/IP, HTTP, The Browser, Search, Social, Mobile, Blockchains. Each one of those innovations drove an investment cycle. Our 2004 fund was built during social. Our 2008 fund was built during social and the emergence of mobile. Our 2012 fund was built during the mobile downturn. And our 2014 fund will be built during the blockchain cycle. I am looking forward to it.

My friend Adam tweeted this in the wake of my The Search For The Next Platform post. I don’t know if it was a reaction to it, but that’s how I interpreted it. I agree completely with Adam.

In tech investing, we get booms when a new cycle emerges and downturns as it matures and consolidation of economics happens (think Microsoft in the late 80s/early 90s). The downturn is always followed by something radical and new that starts on the fringe and becomes mainstream. Timing all of these things is hard. But it is what we have to do.


Comments (Archived):

  1. JimHirshfield

    Great insight. Thanks. So…Kik, it’s a mobile app. You got in on that one.Also… UConn!

    1. pointsnfigures

      Tonight Go ND!

  2. David Semeria

    I would hate to be in a startup whose objective was to either be on the mobile home page or be an “always open” tab in the browser. I think there is more chance of winning the lottery than achieving these goals today.I believe it has less to do with the html / native divide on mobile and more to do with the fact that it’s getting ever more difficult to come up with genuinely innovative consumer facing solutions.However, I see massive value in “horizontal” apps.The two key features about smartphones is that they are geo-aware and inherently personal. I firmly believe we will see the rise of “city apps” where users can find highly personalized content within a geographic boundary.Public transport info, show times, new exhibitions, restaurant reviews, etc, etc.I don’t know whether this will happen at the app or OS level, but the key point is that in the future it won’t be so much about functionality per se, but how that functionality is organized.

    1. awaldstein

      You are reading my mind from Milan.I’ve been thinking about two pieces of this–one that local is still the wild west and that micro marketplaces, not mini communities are going to map this back to consumer need. And that possibly point solutions on mobile will become platforms giving incumbents a huge advantage. Uber announcing delivery services is a case in point and also a bit brilliant.

      1. David Semeria

        The winner will be the one which comes up with an consistent and intuitive framework for presenting the information. People will not want to learn a new interface for each city app — they will want everything to be in the same place.I believe this is the key reason why Facebook beat my MySpace.

        1. awaldstein

          Yup and urban wise, one giant chunk of this is transportation and underlying logistics. Whether this is Uber amassing their drivers to do courier deliver for $15 (really cheap btw) or now AmazonFresh with them going after Whole Foods and owning the delivery channel, this is playing out.

      2. Alex Wolf

        I was at a breakfast with a Google Head of Strategic Planning and data and local came up. I asked if they were looking at a way to mesh this in x format (my vision of it) and she said I should write a white paper on my idea. I’m afraid I’m not up for them eating my lunch. It’s actually a great way to mesh all that together and make it work.

      3. Richard

        Calling a delivery service brilliant is a bit of a overstatement.

        1. awaldstein

          Actually I don’t think so.The simplest things can make the biggest difference.The hard work of connecting the public to the crowdsource of drivers and building the logistics systems was a hurdle or significance.I think it is a really smart move to fill the funnel of activity with delivery using the same merchants, the same customers and adding an entire new level of value and income.Hey, I think little moves that change the world are the biggest thing.$15 to tap a button on my phone and have a box delivered to the upper west side from the financial district is a big deal.

          1. Richard

            My point is that the technology for this is “off the shelf” and for uber its literally on the shelf. I don’t think there was any a ha moment for coming up with local package delivery. The delivery style actually makes more sense for sidecar and Lyft.

          2. awaldstein

            Technology is not the issue.The market is and it is already there if you are using Uber.Who is doing something is more important than who can to me,

      4. pointsnfigures

        Will Uber be able to out execute the delivery services already in existence? Or, would it be cheaper for them to buy them and integrate execution.

        1. awaldstein

          dunno–seems like all they need is drivers.

    2. Javier_Noris

      I don’t buy the “its getting harder to innovate” argument. What makes it inherently harder to innovate? The exponential increasing technological capabilities of smart phones would lead one to believe that if anything innovation should get easier as time passes.Perhaps the platforms/distribution models such as the app store as we know them are what is constraining innovation and all we need is a new model to allow us to break into new kinds of innovation.

      1. David Semeria

        What makes it inherently harder to innovate?Two things: – The low hanging fruit has already been picked (apart from games) – The “rich-get-richer” dynamic mentioned by Chris, making it harder for new entrants to get a toe-hold in native apps.

        1. pointsnfigures

          Unless you pivot the turret to things that have been left undisturbed. Initial innovators tend to go to big markets. Once they commoditize it, they can go for smaller markets.

          1. Jim Canto

            Exactly. There are millions on this planet who have made a VERY comfortable living by simply serving hot dogs… or making house calls as an electrician, etc. It feels like that’s what’s coming on the web; Micro businesses. Developed by DIY’ers (like me).. Browser based (no distribution restrictions), providing a valuable solution/tool/service, to the right market at the right time… with the lowest overhead possible (key principal) for the provider. I’m thinking in the 10’s-100’s of dollars per month to design (without employees), develop (without writing code (don’t hate me :)), and own and operate. Not thousands or millions.This plays well with @awaldstein:disqus’s “micro-marketplaces” mentioned below. Though I think “micro marketplaces” and “mini communities” are synonymous.I don’t know if I’ve expressed my thought clearly here… or if, for the AVC community, it’s like hearing a song for the 15th time in one day…but thanks for listening.If people knew what they could do…and how to do it, what would they build/invent?P.S.; The depth of knowledge and experience I witness within the posts and comments here on AVC is nearly intimidating to a non-coding, DIY’er like me. But always thought provoking. Thank you.

      2. awaldstein

        Scarcity of distribution channels for apps changes the game. This is no different than fighting for shelf space at Best Buy when they were the game in town to sell your products.

        1. Javier_Noris

          Look like we need new ways of distributing apps. I’ll venture to guess we will see a big innovation in distribution in the next year or two, to obvious of a problem.

          1. awaldstein

            The most interesting piece of Chris’s post was that less mobile web solutions were working and that this was pushing everything to closed distribution.Within a year–dunno. You think you could get funding premised that this would change, like we all did with the bet that bandwidth would become ubiquitous?Not as convinced that this problem being seen, like cable content monopolies will find a quick solution.

        2. BlairMacGregor

          Yeah, I think today’s mobile landscape is analogous to the old desktop software ecosystem in a lot of ways. People talk about app discovery and how important and necessary it is to have a PageRank-like system. But the way we search on Google (or Bing or otherwise) and the way we search for apps seems fundamentally different.Most of my search queries are information-based. I’m looking for news about a specific person, gas mileage on a specific car model, information on a new book etc. I can look for those things within a corresponding app, of course. But if I’m not a car buff and the information is for a car purchase that’s going to last me at least five years, it doesn’t make sense for me to take the time to download Edmunds’ app, for example.Conversely, when I search the app store, I’m looking for a piece of software that will do a specific function: a platform game, an app to manage my taxes, a mobile messaging app etc. Either that or I have no idea what I’m looking for but in that case I’m browsing rather than searching.I guess my point is that yes, the app stores need to improve. But based on the fact that our inputs are different, do we really need what works for search to work for mobile apps?

          1. ShanaC

            well, wouldn’t we figure what gets written about on the open wen could be used to find out what would work best.This could lead to gaming content marketing.

    3. Barry Nolan

      It really doesn’t matter if the app is on the homescreen. What matters is that when the consumer has a job to be done, they call your app. That could be 10 times a day for a messenger app, or 10 times a year for your airline app. It really doesn’t matter. Once the app delivers convenience and utility, it will be used.

    4. Matt A. Myers

      I personally like that challenge as it gives one something to think about, create strategy for. With Twitter just buying Cover, I imagine that is one of their goals.

  3. jonwerbell

    will be interesting to see what is next (self-interest has me hoping for lots of innovation in infant care and education support for parents ๐Ÿ™‚ ). and love the pre-6am post.

  4. Mark Birch

    Disappointed that this post is not titled “Mobile is Dead” ๐Ÿ˜‰

    1. fredwilson

      It is not dead. But is is the last big innovation cycle not the next one

      1. Matt A. Myers

        Do you have more on why you say it’s the last big innovation cycle?

        1. Mark Birch

          From a VC perspective, this makes sense. They are looking for the next massive innovations.From an entrepreneur perspective though, there is huge amounts of opportunity and innovation in mobile has only just begun. 2014 will make the year of the next great billion dollar mobile enterprise tech startups. That is not Fred’s focus, but that is certainly my focus and I see wide open opportunities to reinvent the intersection of work, people, and technology.

          1. Matt A. Myers

            I don’t think that it doesn’t make sense, just curious if there’s any end point or signal(s) that causes Fred to say that.I know where an increasingly fluid environment will allow us to move towards – mobile being access to organization and information tools wherever we are – and I am planning for that.I see other massive cycles happening, though perhaps they’re not exclusive to internet technology – however everything is interconnected. Mobile/mobility is part of the idea of being unbound, de-centralized – so the more you can see the intersections here from everything in life, the better ideas and direction one will have.

          2. Andrew Kennedy


          3. ShanaC

            what kind of things do you think will make it big for enterprise tech?

          4. Mark Birch

            Enterprise bypasses the app store lock-in. They are not in an app distribution game so what Apple, Google, or anyone else do has little bearing on what enterprises do. The other thing is that enterprise mobility is disrupting the entire application stack within companies, and even more importantly, the entire underpinnings of work within companies. I have many more thoughts on the topic, but will save for a later post.

      2. Abdallah Al-Hakim

        It is interesting to read your opinion about mobile because just before your article I read anther by a Canadian VC in INOVIA –… – who was pushing the case for Canadian investors to invest more in mobile. He makes good points such the global growth of mobile but I see the point you are making. Interesting contrast!

      3. Ben Healy

        I’m not sure I agree. All of the examples you list for innovation cycles are mini cycles within the broader software cycle. I think the next true tech cycle will be driven by hardware innovation rather than software.

  5. LIAD

    amazing and empowering that a single smart person figuring out a hard problem, in this case how to achieve distributed consensus over an insecure network, can create a macro-trend the ramifications of which will have transformative positive effects for humanity.brainpower is the ultimate leverage

    1. William Mougayar

      Same w the Internet, more or less. It was 3 people that set it on fire: Tim Berners-Lee, Vinton Cerf & Marc Andresssen.

    2. fredwilson


    3. John Rhoads

      brains = good, agreeSatoshi = 1 guy, disagree

      1. LIAD

        you know what they say.the only way 2 people can keep a secret is if one of them is dead.

  6. jason wright

    2004, 2008, 2012. is this four frequency merely coincidence?Phonebloks/ Google Ara and open smartphone hardware architecture could disrupt the mobile space and lead to new innovation. perhaps in partnership with the blockchain.

      1. jason wright

        so are we all meeting up in 2016?

        1. Javier_Noris

          Seems plausible, but business cycles fluctuate in duration after a few cycles typically. I’d be interested to read more about the relationship between the tech start-ups as opposed to the larger economy and the business cycle. I’m sure someone has written this up somewhere.

    1. fredwilson

      It was deliberate. But we broke it this year when we started investing the 2014 fund

  7. Tony Yi

    watched 60 minutes last week about equities market lack of transparency. good example of potential application of blockchain…and plenty of capital to figure it out. #michaellewis #flashboys

    1. pointsnfigures

      How? Interested in your thoughts.

      1. Tony Yi

        part of the issue was lack of visibility into the transactions that allowed flash traders to game the system. so katsuyama created his own exchange with speed bumps to various points to mask the sizeable trade signals from the flash traders. nice bandaid (and a potentially lucrative one). but a more permanent fix could involve creating open blockchain/transparent transactions where everyone could identify and see every transaction/time stamp without heavy manual investigation (that ends up being too little too late)?

        1. pointsnfigures

          Curious why you think its important to know who is doing what? I traded my own money for 25yrs, and was on the CME Board. I love to hear people’s opinions and thoughts on this.

          1. LE

            Ordinary guy says “where there’s smoke there’s fire” despite all the criticism of the author (and the new exchange) as presenting lopsided info that is self serving. Nice PR campaign going on on both sides.Hard for “ordinary guy” to know enough to have a seat of the pants feel for what is correct for sure.In the domain business we had a somewhat similar thing with people locating near where domains were released (by the registry) to shave milliseconds of time and snap up valuable expired domains that were deleted. And plenty of people made money on that. I got checks of over $20k per month (back in the heyday) just from selling 1 connection that I was alloted by the registry. [1] That’s a check for that amount (later dropped to say $5k to 10k still nice, huh) and we are talking straight to the bottom line. No expenses at all. Not wall street big of course. No chicks for free either.[1] You know who got me into that? Was Mike Arrington before he was at Techcrunch. He cold called basically.

  8. William Mougayar

    We’re still in the very early stages of blockchain/crypto based innovations. The platform and its underlying technologies are still raw and not totally defined. Even the amount of VC investments is still small.The outcomes promise to be great, and that’s part of a new trend of the decentralization of everything.Who are the winners that will emerge is still too early to tell.

    1. awaldstein

      Do you think it is too early to tell what the winners will be doing–apps, platforms or exchanges?If you don’t have that belief you are betting not investing.What’s your take?

      1. markslater

        all of the above. There is alot of noisy headless “thought leadership” going on around the blockchain right now. Get your very large combine harvester sized Sythe out and start hacking…….but for sure there is a beautiful plowable field underneath the chafe.

      2. William Mougayar

        I’m a believer 100%, and currently analyzing the segmentation opportunities in this space. I’ll need to write a post on it, this week.

  9. Jordan Thaeler

    “Did we change our appetite for risk in our most recent early stage fund? It wasnโ€™t conscious.” How about the move of VC to PE? PE has outperformed VC for decades and takes substantially less risk. Seed funding comes from incubators or F&F mostly as VC pulls back to Series A: and you had better be cash flow positive. Most funds will deploy as much cash as possible to a growth equity round in Uber, AirBnB, Dropbox et. al. Lean startup means executing small ideas on shoestring budgets. You cannot achieve something large if you’re forced to think small.

  10. JimHirshfield

    We have a bunch of great investments in the 2012 fund, but we took safer bets and the result will likely be more wins and less strikeouts but also less home runs. Did we change our appetite for risk in our most recent early stage fund? It wasnโ€™t conscious. We didnโ€™t change our investment strategy. We didnโ€™t tell each other that we wanted to take less risk. But it sure feels like we did.I don’t think risk analysis in retrospect is accurate. It felt like risk back then; how can you say it was less risky now? Losers tend to say they took risks and failed. Winners tend to say they made smart decisions. Generally speaking. Point is, I think your “Monday morning quarterbacking” on the risk thing.

    1. Cam MacRae

      On the contrary, post hoc sensitivity calibration is incredibly valuable. But only if it’s done well.

      1. JimHirshfield

        Yeah, I agree with you on taking a look back and evaluating. But if it seemed risky when you made the investment, than it was risky.

  11. Alex Wolf

    Fred, The longer tail of my work at na2ure which you’ve seen part of is a new way forward in mobile that’s wicked cool. We’re just not there yet (I’m the product person who wants to hand over the CEO reins and stay on product, but still looking for the CEO). When we get closer to making, it will be fun to show you.

  12. Julien

    The blockchain is probably one of the most innovative “beasts” we’ve seen in a long time, but I also think that the web stack is far from being dry at this point ๐Ÿ™‚ More than anything, I am convinced that whatever successes the app world achieved are in fact areas where the web needs to be stronger.It starts with discovery. App Stores have a lot of issues, but so do search engines when it comes to disocvering content and applications we use an love. The social web needs to step up here.It’s also true when it comes to usability. Apps are on average far prettier and better designed than web apps because the OS + frameworks (iOS, or Android) offer a lot more building bricks than what JS + HTML has to offer. This is chnaging thnaks to initiatives like bootstrap, but we need more of that!Payment is also a big aspect of why mobile wins. Even if it’s outrageous, itunes in-app purchases (or Android’s) are incredibly easier to use that even Amazon 1-click purchase which is arguably one of the best integrations of payment out there. Maybe BTC will help there as well.And there is also notifications, geolocation, offline capabilities… etc.If we want the web to succeed, as devs, we need to start addressing these issues for realz!

  13. Aaron Klein

    Let’s keep in mind that this problem is largely constrained to consumer apps that need to be distributed for free so they can build an audience and then figure out how to monetize.The future is still very bright for businesses that build their own distribution networks. For example, in the segment called “real world enterprise businesses that don’t get covered on Techcrunch,” we use the revenue we collect from our customers to drive adoption by newer customers through a process we call marketing and sales.I know, it almost seems like that’s an innovation in itself. ๐Ÿ˜‰

    1. awaldstein

      Comment of the day my friend.

      1. Aaron Klein


        1. awaldstein

          Selling a product that brings real value that people are glad to pay for can be a real pleasure.

          1. Aaron Klein

            It’s so true. I’m more and more divorced from that moment, but I still love seeing a customer’s eyes light up when they get it. I had one yell out “sold!” in the midst of a presentation last week.The moments that make you smile… ๐Ÿ™‚

          2. LE

            I had one yell out “sold!” in the midst of a presentation last week.Ok question for you then.When they yelled “sold” did you get your ass out of dodge or did you keep talking and finish your presentation?

          3. Aaron Klein

            I wish I could have stopped right there, but I had 50 other customers I was presenting to who hadn’t yelled it yet. :)If that was a 1:1 presentation, I would have stopped and said “so do you want your training on Thursday or Friday?”

          4. LE

            Exactly.This is best described in a scene in the movie “Heat” (Pacino/Deniro) where someone turns someone on to a potential lead by saying “if he hadn’t said nuttin I wouldn’t have thought nuttin”.Meaning once you get the sale you keep your mouth shut. Anything you say after that gives the person a chance to reconsider or raises issues that they might not have even thought about.(In the movie version of your life the person yelling “sold” will be planted, much like a claquier in plays of old time (essentially the laugh track).) I know you won’t want that but you will have given up creative control of the project.

          5. sigmaalgebra

            > Meaning once you get the sale you keep your mouth shut. Anything you say after that gives the person a chance to reconsider or raises issues that they might not have even thought about.Yup! I have an uncle who was a good salesman. One of his sons wanted a similar career, and, when he came home after a day trying to sell, told his father what had happened. And, yes, one day he had the sale, hung around the customer’s office trying to make the customer feel even better, had the customer change his mind, and lost the sale. So, when the sale is closed, be nice and leave. But leave. Yes, be nice but get the heck out’s there.Apparently a universal point in selling.The son now has private label janitorial supplies in half of a major state.

          6. Girish Mehta

            Something my ex-boss often said – “Don’t talk past the close”.

          7. Aaron Klein

            Love it.

      2. Matt A. Myers

        Feature worthy, too.

    2. Cam MacRae

      Bam! Pure gold.

    3. Abdallah Al-Hakim

      Yup – I have worked for both type of companies and definitely the B2B field is satisfying in that people want to pay for the product because they see value! You can do the same in the consumer space but many focus too much on the freemium model and don’t establish a clear business model until they can grow their user base by several folds (which sometimes never happens)

      1. Aaron Klein


      2. John Rhoads

        The complaint is that the App Store isn’t actually a viable growth or discovery channel for mobile appsThe same big boys with incumbency get all the spotlight

        1. Niv Dror

          We are in the pre-PageRank days of app search. Check our Quixey – the search engine for apps. It uses functional search, here is an example I tweeted the other week:

          1. ShanaC

            Interesting, but the mobile phones themselve often use a lockdown mode, and you still need to convince people to switch search methods. App finding a=has to get a lot more spammy (which thanks to company control, it isn;t yet) before it get better.

          2. Niv Dror

            I believe that they now power search on Sprint – or at least has some kind of relationship with the m. Hopefully Quixey’s tech can be integrated in more places… They also just opened a new office in Israel.

        2. Robi Ganguly

          There are two aspects of this that people seem to make blanket statements about: 1) App store search is broken and 2) The app store isn’t a growth or discovery channel for mobile apps.On point 1: from the consumer’s perspective, app store search works reasonably well. There continue to be improvements in both major app stores (unsurprisingly Google is leading the charge here) and the personalization and ranking methods are getting better. However, “app store search is broken” tends to be a complaint by app publishers who want to rank differently.On point 2: the influx of mobile apps certainly makes it hard for many app publishers to stand out in the crowd. However, this does not mean that new apps can’t find growth and discovery in the app stores. We have hundreds of clients who started off small and have grown, mostly through the hard work of building and iterating on a product and improving it as they listen to their customers.

      3. ShanaC

        Yes, but pricing is a complicated question

    4. Salt Shaker

      “The future is still very bright for businesses that build their own distribution networks.”At its fundamental core, isn’t this all about identifying and delivering needs based solutions, irrespective of platform? Consumers are willing to pay for perceived value. Mobile isn’t dead or in a downturn, creativity is. Too many products/services rely on the belief that scaling will lead to monetization, and although it can, there cert are no guarantees, part w/ companies that are ad rev dependent. If your product/service doesn’t deliver significant/meaningful points-of-differentiation and address legit market needs, then fold the tent. You’re salmon swimming upstream. The app store is full of garbage and parity products. Building needs based sector solutions (e.g., energy, transportation, finance, healthcare, food) still holds sig more promise than identifying the next big platform.

      1. Aaron Klein

        Unfettered free distribution is in a downturn.This post is correct. It’s Fred’s mobile version of “don’t be a Google bitch” except there are no diverse sources of distribution right now for mobile apps, so freemium on mobile is getting tougher.My point was that freemium != innovation and there are other ways to build distribution, but they typically involve a more traditional marketing and sales process.

        1. JamesHRH

          Guess what most VCs will not ingest in – buildung distribution.

          1. Aaron Klein

            The smart ones will focus on that in equal amounts to product.

          2. ShanaC

            um, spelling

    5. chris dixon

      It’s an interesting question whether enterprise mobile will be more app or web based. My guess is app based for major apps like Salesforce and Workday. But in enterprise it is mostly just a software development / architecture issue, since distribution is likely to come through other means than mobile discovery.

      1. Aaron Klein

        Right. I think my point is that a lot of the worrisome issues you talked about in your post are mitigated when you’re not relying on mobile discovery and you have an enterprise marketing and sales engine for distribution.

      2. Mike Bestvina

        As someone who built a team who developed mobile apps (native iOS mind you) to the Enterprise, here is my take: it’s still expensive as hell to create mobile apps and security is a pain in the ass. Mobile is built for consumption (i.e. execs love it), but the larger part of the biz workforce is built on production (writing, data entry, developing, etc). Until the costs go down, or it’s pre-packaged with the vendor’s software (Workday does this well), then I still see enterprise app/web adoption lagging behind.

      3. ShanaC

        depends for what – I wish it was “app that integrated into other apps” base.

    6. Pete Griffiths

      I’m not sure this is true, is it?Businesses as well as consumers have been deeply impacted by innovations such as social and mobile and are likely to be profoundly affected by the next wave. Or have I misunderstood you?

        1. Pete Griffiths


    7. ShanaC

      What do distribution networks look like?

      1. Aaron Klein

        That’s a blog post in and of itself.

    8. PhilipSugar

      That’s what we need Aaron: A snappy name with an acronym to say what you said. I need to buy a vowel however.

      1. Aaron Klein


  14. Barry Nolan

    Whatever the merits of web v native apps, the control of stores is of utmost importance to innovation.We have two players, one a gatekeeper, one an open house.Apple give us security and censorship (conforming to rules and subjective definition of an experience).Google gives us no security and open innovation.The No1 app on Google Play this week was Virus Shield. That it was the No1 paid app speaks to Androidโ€™s users fear. That same app was a complete scam – it does nothing. The store’s openness only reinforces Androids weakness. Our smartphones know more about us than any other technology. When malware infects, the consequences can be profound.Iโ€™m all for open innovation. But Iโ€™m also all for a store review of apps to verify security. That seems like a happy medium.

  15. Matt A. Myers

    “Apps have a rich-get-richer dynamic that favors the status quo over new innovations.”This is purely related to the inherent nature of the value and efficiencies that fluidity and frictionless brings relating to accessibility.Accessibility is relating to organization, full accessibility means fully organized (as best possible).Google’s mission is “Organizing the world’s information.” I’ve come to realize recently that my mission is “Organizing the world.”

  16. Twain Twain

    As I wrote in the “Next big platform?” post, I believe the innovation will happen in the structure and quality of data rather than the distribution channels of it nor the volume scale of that (Web, mobile or wearable and “Big” is not where the innovation will be).If we look at cutting-edge AI, Vicarious — which just attracted some more investment from some more billionaires — it’s about a step-change in the structuring of the machines to simulate our brain’s intelligence and its Captcha test is about qualifying bits of information.

  17. brandoncarl

    Thanks for the post Fred.After working on Wall Street for many, many years, I’ve come to realize that entire investment strategies are subject to supply and demand. This occurs due to the fact that many investors operate looking at historical returns. I think that part of what you’re highlighting is attributable to very strong demand for VC strategies over the past 5 years.Case in point: macro hedge funds typically profit from trending market moves, as they need to deploy large amounts of capital across a small number of themes. The late 90s were a boon for momentum investment strategies. However, as the trend stopped, and started chopping, the Tiger funds blew up.Meanwhile, range-bound markets are great for quant and arbitrage strategies. Just such an environment existed for the few years proceeding 2008. As the range broke, we had a less-publicized “quant crisis” in 2007.Namely, as the demand increases for investment strategies (like Tech VC), so does the supply (e.g. incredible numbers of incubators launching). I would imagine that makes for a tougher market, not only for the VCs, but the app developers as well.

  18. Richard

    Innovation always follows a S curve. This time is not different. Remember, the genius was the triumvirate of the microprocessor, the personal computer and the smartphone. The innovations that followed was somewhat predictable and are marginally fewer (not harder) as time goes by.

  19. shareme

    IF a major payer dida mobile web browser with a couchdb engine we would see a swing towards mobile web apps

  20. Bruce Warila

    glad to see “blockchain” is the new noun

    1. JimHirshfield

      Can we verb it? “Johnny blockchained that business model; so radically disruptive”

      1. LE

        A ping used to be just a ping. Then everybody and their uncle started using it.PING ( 56 data bytes64 bytes from icmp_seq=0 ttl=50 time=57.837 ms64 bytes from icmp_seq=1 ttl=50 time=70.354 ms64 bytes from icmp_seq=2 ttl=50 time=53.522 ms

        1. JimHirshfield

          What’s your point?

          1. Cam MacRae

            He’ll ping you and fill you in.

          2. JimHirshfield


      2. Alex Wolf

        love verbing, and while disqussing extra points!

        1. JimHirshfield

          Hahahaha! Ani-gram-it-ed.

          1. Alex Wolf


  21. Brandon Burns

    Focusing on social in 2008 and mobile in 2012 means that’s you could invest in many different kinds of businesses, as anything can be social or mobile.The blockchain is specific to bitcoin. If all your bets in this new cycle will be centered around that one small piece of the world, that sounds a bit dubious to me.

    1. JimHirshfield

      Any business that has transactions can use the bitcoin blockchain. And every business transacts. Same logic back at ya as your logic re social or mobile, no?

      1. Brandon Burns

        Hm… never looked at it that way. Maybe you’re right!I think what’s more interesting is the general concept of a centralized database-of-sorts from which all transactions occur. Fred talks a lot about decentralization, but the reason why bitcoin is do “decentralized” is actually because its very centralized. Its all built atop one blockchain.The concept works in other areas. I’ve been applying this thinking (not consciously, but coincidentally) to shipping and inventory management. The next phase for W&T will see that thinking incorporated in some interesting ways. Stay tuned!

        1. markslater

          blockchain is the conduit brandon. its the ultimate decentralized platform.

      2. ShanaC

        is there a way of using the blockchain without bitcoin?

        1. JimHirshfield

          You’re asking the wrong guy. Not an expert, but I think not.If you do, you can use an infinitesimally small amount of bitcoin with each transaction.

        2. Matt Zagaja

          I thought blockchain was just a technology bitcoin was built around and you can create independent blockchains that have other currencies like ShanaCoins.

  22. whitneymcn

    Sorry, but my father and his editing pen are (figuratively) standing over my shoulder, so I have to say this: it’s “fewer” for things that you can count, “less” for things that you can’t.

    1. JimHirshfield

      Ah, so “fewer strikeouts” and “less risk”?#TIL

      1. whitneymcn

        Exactly so.

    2. Grammar Gal

      One of the tricky ones — and this is right in Fred’s wheelhouse — is coffee. You could want less coffee in your cup, or order fewer coffees for your group, if one of your buddies doesn’t drink coffee.

      1. ShanaC

        thank you grammar gal ๐Ÿ™‚

  23. Semil Shah

    It’s so hard to reach scale on mobile, but the scale of mobile is so huge. It’s so hard to build an audience on mobile, but it’s where all consumer attention is. In terms of early stage investment, you’ve nailed it – most players have chosen to stop and just wait to see what breaks out and then compete to pay up into growth.

    1. LE

      but it’s where all consumer attention is.What are you basing that on? Particularly the “all” part?

      1. Semil Shah

        I’m being colloquial. If you read Chris’ post and see the charts, the data is pretty clear that consumers are focusing on mobile/apps and not the web.

        1. LE

          Ok but Chris saying”People are spending more time on mobile vs desktop”is not the same as:”but it’s where all consumer attention is”Not to mention the fact that it’s not that I doubt that some of this is true but not knowing the methodology of the underlying data I’m not as quick to accept the conclusions. Especially when people (like Fred, investors, developers) are making business decisions based on the conclusions. (The echo chamber at work).From Dixon:”What wins mobile, wins the Internet.”I don’t agree. What does that even mean “wins the Internet”? Sounds like the LBJ character from “The Right Stuff” talking about the space program and why we had to beat the Russians.Let’s just take amazon for example. They are growing. Do we know how much business Amazon is doing by mobile vs. by web (which is how I order from them?) [1]Or do we know how many people pay their online bills on a laptop on the web vs. on their mobile phones?[1] 2013 data I pulled up. I’d hardly say this is close to even “most”…

          1. sigmaalgebra

            LE, so cruel, how can you be so cruel? :-)!I mean, this is an echo chamber party in honor Dixon and all things mobile (it’s eating the world, at least, right, even the outer planets?). We’re all supposed to join in the social conformity! I mean, the lines on the graph went up and to the right with one of them more upper and righter than the other, and that’s supposed to be deep enough to be totally convincing, right?Then along you come raising questions!Ah, come on, junk your desktop computer! Get a smartphone, an iPhone 21 or some such, and be happy. You’ll never be sorry, never want to go back, and will be happy, right? I mean, I mean, I mean, just what could you ever want in computing or your life that an iPhone and the pick of the app store would not be good enough for you?Join the party, take another pass by the punch bowl, or ask Fred to fill your glass again, and in the morning just toss your funny paper party hat! You really want to be a party pooper??? :-)!

          2. LE

            Yeah I just don’t get how someone can use a word like “all” and then claim that they didn’t mean “all”.It’s as if people get in this mania and then fail to see anything but the halo of the group opinions and thoughts which confirms some pre existing bias.Imagine if I said “besides what a woman really wants is to stay at home, have kids, and take care of her husband”. God knows I’d get into trouble even if I inserted the word “some” into that sentence.

          3. sigmaalgebra

            It’s simple: Growing up we miss some lessons and get some others wrong. One of the lessons we need to get is that the two words “all” and “all” have very different meanings. “Some” women can be especially good at teaching this lesson! E.g., (1) “all” you ever do is work. E.g., (2) “all” you care about is making money. E.g., (3) “all” you ever do is watch sports on TV and drink beer. And there are related lessons: “Yes” and “No” can mean the same thing. One of the classic lessons is, she spent $600 at a 40% off sale so saved $400 so that if she does this 100 times she will have saved enough for a year of college for one of the kids! There are a lot of lessons to learn when growing up, and somehow schools don’t teach them very well!

          4. LE

            Reasons why men work all the time.a) You need money because the wife is looking at what the other wives have and what the other guy has that you don’t. The jones.b) You are avoiding undesirable things at home (wife should just hire a handyman and stop hocking you).c) You like to work, combined with a and b above.d) You need to work because god knows people really need money.

          5. sigmaalgebra

            Of course work and money are very important. I mentioned those two just as a joke to illustrate how a cartoon style parody of a nagging wife could use “all” in contradictory ways!

          6. Amar

            I was feeling lonely and now I don’t ๐Ÿ™‚ I don’t know the two of you and we may not get along but at the moment I am terribly glad that world has people like @LE and @sigmaalgebra:disqus ๐Ÿ™‚ Love the challenge to status quo.Reminds me of this venn diagram by Jessica Hagy, I cannot find it online but I took a screenshot when i first encountered it.

          7. sigmaalgebra

            > Love the challenge to status quo.Remember what wise guru say: “Sacred cows make best hamburger!”.Also, finding solid truth is something like finding solid ice on a lake: Have to stomp on it really hard to be sure it is solid.

  24. pointsnfigures

    Blockchain is interesting-not because of consumer. It’s really hyper interesting when one starts to brainstorm B2B uses of the block chain. I don’t know what the critical mass or big pivot that will cause businesses to engage in block chain usage with other businesses-but it’s coming.

    1. Stephen Bateman

      Could you give an example?

      1. pointsnfigures

        For example, when a farmer transacts with a grain elevator, or a cattle rancher transacts with a packer. There are issues around who owns what when, that relate to insurance. (had this problem with finishing grass fed beef) It’s not always clear-and contracts are set up to take care of that. Then who hedges what when. Who’s on the hook? How is the risk distributed? Blockchain transaction might make that a lot clearer.

        1. Stephen Bateman

          Awesome thanks!

          1. pointsnfigures

            Also think that contract and patent law could be first disrupted.

        2. LE

          I don’t know much about how farmers operate, but I do know how businesses operate. I can’t really see any of the issues as not already being solved [1], at least not to the level that you are making them out to be.The marketplace already has figured out a way to deal with this that fwict it is working.Not only that but things that would have been unthinkable years ago are happening now. Used to be getting signatures (so you know when title passes) was pretty much absolute with almost all deliveries [2]. I regularly get packages delivered to my office from Amazon left at my door where someone else can easily take the packages. Yet Amazon has decided that it’s a good risk vs. the extra expense to have UPS/Fedex/USPS return with the package and get a signature. [3]big pivot that will cause businesses to engage in block chain usage with other businesses-but it’s comingYou are aware of course that many businesses, both large and small, still pay their bills by check? And many municipalities still issue multi page mailed paper purchase orders for small purchases as well? (By that I mean for practically any amount of money, even $50?)My point is this critical mass is a long time coming if at all (for what I think you are describing of course if I have interpreted wrong let me know).[1] If I understand your point as being “when does title to the goods pass?”.[2] Pizza delivery being an exception of course. Or a service like getting your lawn mowed.[3] Of course they aren’t going to leave a laptop doesn’t fit the risk profile.

          1. pointsnfigures

            I think you have to be open to doing things a different way. Just because it’s “always the way we have done it” isn’t an excuse if something is more efficient. Are we still using paper general ledger or is Excel better? It’s hard to come up with a concrete example on the fly because one hasn’t been invented yet. I do know that when we were dealing with grass fed/finished beef, there was a who owned the steer issue at a later point in the fattening cycle. It was solved-but I wonder if it would have been easier to solve with a block chain app?

          2. LE

            Me? I am not talking about me. I am talking about them. I’ve never met an efficiency that I didn’t take up to shave time or effort off of something. (You have no idea it’s practically a game….) I used to get paid piecework as a kid so I’m all in to anything that makes things easier. Sometimes I need to make copies. So I have two copiers next to each other because it’s faster than waiting for one copier (when not doing volume that is for volume you use volume equipment). And the list goes on and on. I have no clue why houses aren’t built with more than one clothes washer or dishwasher.Have you ever sold directly to the type of people that you say should be open to “doing things a different way”?I have. I’ve actually transacted business with these people and this is over many many years. So I know exactly what it is like to try to get them to change what works for them as “good enough”.And you know the reason that it is hard to get them to change (in addition to other things you might think about on your own like the cost etc.)?Because if it doesn’t work it becomes their problem. They are SOL. And it’s their time to learn and use the system and in the mean time they don’t even have enough time and labor to handle the things they have to do when they come in in the am. So they put off making any changes. And if you were them you probably would as well.Small example from the past: When I was in high school I worked as a gopher in a law office. I remember one day where one overworked secretary got in a brand new IBM memory typewriter where she could store template letters instead of doing each one from scratch. That was a complete win. No way it didn’t pay to use it (and she didn’t even have to pay for it). She wasn’t stupid she just was so busy that she didn’t have time to even crack the manual. So she continued doing things the old way. Because if she didn’t get the letters out that the lawyer dictated she was up the creek. No excuses. Same with the lawyers clients. No excuses. (Investors are by the way shielded from all this day to day “the truck broke what do we do” goings on that happens all over the world).Another example: Fred going to WordPress from typepad. It was a big deal. Fred is busy and had no time to manage it and wasn’t going to just trust someone else. And that is the way these things are for the majority of business owners. There are no committees just some guy trying to earn a living. He isn’t going to adopt something unless it’s a super big pain point and he has the time and bandwidth to make it happen and after considering the risk of making the change.

          3. pointsnfigures

            Yah, I have sold. It’s not a snap, and I didn’t mean to give the feeling that it was. But, I agree with Fred, the block chain is a big deal.

    2. Matt Zagaja

      What about government uses? Maybe it’s just because of my current line of work but I’m utterly fascinated by the idea that blockchain could be great for electronic voting. An interesting comment on this at:….

      1. ShanaC

        that could work, but you’d have to get the government involved in development. And that is asking for a fiasco.

      2. pointsnfigures

        I think that is right. But govt will come to adoption slower than business. Just look at how crappy their websites and UX are.

  25. Rick_Robinson

    The new platform needs to merge (neither exclude nor include) discrete apps: http://www.huffingtonpost.c

  26. Elia Freedman

    The pendulum swings. I’ve been mobile my entire career, starting with PalmPilots and Newtons back in the late 90s. Even when we did Windows it was designed for tablets (read: touch, pen), not desktop (read: keyboard, mouse).The last few years, however, have been some of the worst from an innovation perspective. Focused simply from a design perspective, the iOS App Store in particular emphasizes all the wrong things. It:1. Gives us massive scale immediately2. Requires a high level of polish3. Long release cycles4. Connects us only indirectly to the customer5. Offers only vanity metricsBuilding something new requires all of these in reverse. We can’t handle massive scale because we don’t know how to scale yet. We want a rough product to launch that can be revised quickly and easily. We need tons of customer touch points so we know whether we are on the right path. We need in-depth data and knowledge to refine our metrics, not vanity metrics such as downloads that tell us next to nothing.I’ve been working on a new idea. It was all prototyped in iOS. But now that we are moving to launch the product by the end of April, we shifted to developing a web-first version with responsive design. While it has other problems that we will need to overcome, it solves each of the problems we’ve faced above. If our customers keep coming back despite the short-comings, we know we are onto something big. The mobile-specific (offline) versions can come later.

    1. John Rhoads

      Could you explain why you are only indirectly connected to the customer? Is the a comment about oAuth for web vs ?noAuth? for mobile?Also congrats on getting massive scale immediately haha. That sounds like first world awesome problems.Finally, curious why not Android to avoid some of the process and release red tape

      1. Elia Freedman

        Connection to the customer: we have to design ways to get direct customer contact since Apple/Google stand in the middle with the app store.Massive scale: this isn’t about achieving it. It’s about controlling the scale so you can scale as you are ready.Android first doesn’t fix any of the problems except #3. If I had to choose learning HTML5 or Android, which I did, I chose HTML5 because it is more universal. I can take lessons and code from that to both platforms. In addition, most of my current customers are iOS and I’m connecting with those folks for the new product. Designing Android means throwing all those relationships out.

        1. John Rhoads

          Interesting and thanks. Thought Android might address #2 also.The platform app stores are metrics black holes. Not much information coming out of there now.What’s the single most important metric you’d want to be provided to you?

          1. Elia Freedman

            I don’t know that answer yet. Too early. But I know it is not downloads and most of the other metrics I get from Flurry.

  27. John Rhoads

    I think is a fascinating conversation that has stemmed out of Ben Evans’s hard and clear work.The problem with developing a great mobile web is that you get a quazi-rent situation where to build it somebody would need to invest a whole bunch of energy in it and that energy would have to reside right next to iOS and Android.These two entities have a vested interest in a great mobile web NOT happening (arguably more so for iOS). The apps and their performance advantages on hardware setups are actually retention mechanisms for their platforms (not to mention huge revenue streams). Increasing the cross-portability aka decreasing the switching costs of going from one platform to the other would reduce the value of their platform and make some really open-source platform valuable and competitive. Something they will wage a “holy war” against.More specifically the way towards a mobile web would be to create better programmatic access to the really valuable signals that everyone has discussed here and elsewhere (e.g. location, change in location) and also figure out how to do that without taxing the battery too hard. I don’t see Apple encouraging this in the short-term nor does the fragmented Android marketplace make me feel encouraged that the battery problem will ever be solved for those devices (perhaps an argument for Apple to pursue it more aggressively)I do think you’ll see the user process of what it means to get and find and app, install and app, and launch an app change. There is very little reason for it to be the search in the app store bucket approach that it has now. I’d love to hear thoughts about how the app store fails users today

    1. sigmaalgebra

      A ” mobile web” is what? A mobile device has something like a Web browser which uses wireless access to the Internet to get to something like Web servers much like desktop Web browsers get to Web servers except the “mobile Web” also exploits GPS, etc.?Well, exploiting GPS should not be too difficult from, say, just a current Web page with some AJAX software. So, some JavaScript or some such running in a Web page running in a Web browser running on a mobile device could, with an API in the mobile device, read the GPS device and send the data back to the Web server where the Web server could know position, direction, and speed of the mobile device and, thus, give directions to Starbuck’s, warn about police radar traps, etc. Quite generally a Web server, just as a standard post operation, can send data to a Web browser that, presumably, some JavaScript in the Web browser could read.

      1. John Rhoads

        we can discuss what mobile web is today, but perhaps more interesting (and what a16z and Fred Wilson allude to) is what mobile web should be.I’m not sure what the answer is to that, but to describe a simple example that is not in conflict with your description above… I see no reason why saving a responsive site like as a rectangular bookmark on my home-screen and appifying it isn’t a step towards a better mobile experience. The square button app use modality is to limit user keypresses to get to the desired “experience.” I press this and I get this. No need to have dumb text serialization addresses for content. We can use pictures and taglines.For blogs and non-rich media that seems not to far away.To your second point about technical implementation… I’d agree about technical feasibility with some (big) caveats around security and energy management. The larger hurdles are, in my opinion, the incumbent incentives (i.e. the power players never wanting to do this)

        1. sigmaalgebra

          > I see no reason why saving a responsive site as a rectangular bookmark on my home-screenand appifying it isn’t a step towards a bettermobile experience.I use what I believe are better approaches to doingthings than putting an icon or some such on ascreen, especially on the limited space of a smallscreen.Broadly I find icons wildly inefficient; I have alot of software on my computer, but I nearly neverstart a program with an icon and have better means.> No need to have dumb text serialization addressesfor content. We can use pictures and taglines.I nearly never type a URL.I nearly never dial a phone number.The key to my work at a computer is text filesmanipulated with my programmable text editor.E.g., if I need to contact my bank via voice, then Iknow what directory has my information on my bank,have a fast ‘tree walking’ way to get to thatdirectory, use my editor to read the file with theinformation on the bank, use the search ability ofmy editor to find a line with the phone number, andtype one command DIAL to dial the phone number. Itwould be easy enough in my text editor to have thecommand DIAL assigned to a key so that I could diala phone number with one keystroke. So, DIAL is aneditor macro that reads the phone number from theline and sends it with an old AT command prefix tothe old COM1 port which has a FAX modem card whichdials the number. Works great.Still with, say, my bank, for its URL, I have mytext editor find the lines with the URL, my user ID,and my password, swat one keystroke to have my Webbrowser go to the URL of my bank, and with a commandGPS (global put on the clipboard after strippingblanks) one at a time move over the user ID and thepassword.For the Web sites I go to more frequently, I have alittle Web page, that is, a file of type HTM; my Webbrowser starts with that page; and that page haslinks to and a few hundred more.I have several other ways to find things, usuallymaking good use of my favorite text editor, variouscommand line scripts, e.g., for directory treewalking, etc.So, I hate icons and have better solutions.On a mobile device I wouldn’t have my favorite texteditor and likely wouldn’t have a good substituteand, then, would have to do what you do.

  28. Dasher

    Mobile is just an another gate to the web controlled by Apple and Google.

  29. timrpeterson

    @fredwilson:disqus Curious how you reconcile mobile and blockchain both being the focus. Apple bans Bitcoin apps and Android isn’t carte blanche, right?This seems to me to give the web a clear advantage.

  30. jonah

    The perception of appetite for risk has to be, in part, a function of your own cognitive biases. Perhaps the blockchain will not, in fact, be the next big innovation on the commercial Internet, but is only one of several options currently fighting for that designation. Perhaps in ten years all blockchain/bitism investments will go to hell, and you’ll look back at your 2014 vintage as exhibiting extreme risk.Interesting stuff to think about.

  31. brlewis

    Minor point: “the past 20 years” only go back to April 8, 1994. The commercial Internet predates that, and is in turn predated by the browser, HTTP and TCP/IP.

  32. Manuel Molina

    your last paragraph sounds a lot like Carlota’s (Perez) theory. totally agree.

  33. Steven Livingstone

    I suspect the wave of innovation you finish with will be a return to the web – most apps being web based and a few native apps rather than the reverse. Most apps – especially in the consumer space – simply don’t need to be native (in the same way most semi-interactive web pages don’t need a special browser to view them). The problem will then move to contextual discovery and surfacing of “apps” rather than a few wasted apps installed locally taking up space and resource. The app stores are a mess just now.

  34. Matt Zagaja

    I’m going to have to disagree with Chris’s post on a few points. The first chart he posts from ComScore where he claims people are spending more time on mobile than desktop has two axes: “Number of global users (millions)” on the Y axis and time on the X axis. Both desktop and mobile users are growing, but mobile users are growing faster. This has no bearing on which device they are spending more time on, they could be using both. I know many people who own an iPhone that they barely use and do all their computing on desktop, and I know others for whom their Android phablet is their primary computing device. I think intuitively it’s clear mobile is growing and it’s share is growing and therefore you want to say that time spent on mobile, and it’s share is increasing relative to desktop (see… but it’s not clear that desktop use is decreasing, at least from these charts.The second chart refers to app domination and we are treated to two snapshots in time. What we don’t know from the chart is if the first people who came onboard are changing their behavior to use apps over mobile web, or if the new people skew more heavily towards app usage while the older people skew towards web usage. Either way it might not be terrible, if overall mobile usage and time is increasing then 14% of millions of person hours spent on a mobile device is superior to 20% of thousands of person hours spent on a mobile device.I don’t know enough to comment on the rich-get-richer dynamic but as a consumer I’ve noticed a few things that I think might prove informative. First is that in almost every category on the app store there is a lot of competition. Search for apps that turn your camera into a scanner, podcast apps, read-later (aka Instapaper clones), navigation apps, photo filter apps, and you’ll have much variety to choose from. The second is that reviews matter to me more than downloads. People don’t download things because they see an ad or it’s on the front page of the app store, they download because a friend showed it to them.Finally on blockchains I did not fully understand Bitcoin and the advantage of blockchain until I listened to John Gruber’s interview of Glenn Fleishmann on the subject (…. I was impressed. Now if bitcoin pegged its value to the dollar I’d use it in a heartbeat.

  35. willfprice

    There was essentially no dispute that social and mobile were the next thing in their days. To think that “blockchains” or Bitcoin based products is the next thing is quite a leap and something that imo real technical people would warn you is simply not the case. But hey, you said you were worried about not taking enough risks, so have fun with that.

  36. M. Edward (Ed) Borasky

    I just recently returned to smartphones after an absence of nearly two years. Quite frankly, the only apps I *want* on my (Android) phone are:1. Native Google – Google Now, Maps, Google+, Hangouts, Calendar, GMail, YouTube, etc.2. Twitter3. Firefox for Android (Nightly)4. Wolfram|Alpha5. HDR+ camera app.The rest of them are garbage. They increase my attack surface, waste space, drain the battery, gobble bandwidth or all of the above.

    1. ShanaC

      there are a few extra i love – evernote being one of them. but many are a waste

  37. Greg Cote

    Great post. Very interesting take on your portfolio dynamics. You’re right. Its phase II time. Most of the technologies we have seen over our careers have seen a second leg up about 10 years into it. Wifi started with Pacific Monolithics in 91 with the second wave in 99 (300 companies funded). Single Chip Systems (RFID 1993) with the second wave when Walmart adopts RFID in 2003. Geocities (Social v 1.0) sold to Yahoo (99) and Facebook (2nd wave) hits the inflection point of MAUs in 2008. I got that first TMobile HTC smart phone in 2002 ish?? I am wondering if we aren’t on the verge of the next leg right now. It will be interesting to see who reinvents this vacuum and how they do it. I am generally surprised by the creativity of the second wavers. Do we call them fast followers?



  39. sigmaalgebra

    The analysis reported here by Dixon and USV is curious but, I have to believe, nearly irrelevant for the future of computing, mobile, the Web, innovation, entrepreneurship, and venture investing.First, what the heck is the intended point of considering the overall growth rates of ‘sectors’ such as computing, mobile, …?Well, if there were an ‘index’ fund for such a sector, then a rapid growth rate for the sector might be a reason for investing in the fund. But for venture investing, likely even in a sector with relatively rapid growth, such an ‘index’ fund, with investments in essentially all start up or early stage companies in the sector, would still lose money because such a large fraction of such young companies do poorly and the rest nearly never make up the difference.So, we don’t consider such a sector as promising for investment broadly.But as is fairly clear from the history of computing, especially over the last 20 years, such a sector is like a mountain stream: It’s cold, uncomfortable, messy, and full of mud and gravel but also, if careful, and look for streams in especially promising places, one of the best places to look for gold.To me, this stream analogy has most of what can expect to get from considering such ‘sectors’ and is not powerful enough to mean much for venture investing.In short, instead, have to look at projects one at a time paying close attention to the details of each specific project and largely ignoring the sector.Dixon’s graph of mobile versus Web doesn’t say much about the future of either mobile or the Web.Is mobile growing? Sure it is. Does mobile promise to continue growing? Of course: Prices for mobile devices are coming down, way down, quickly; in much of the world, for purposes of voice telephone, mobile is much cheaper to install than old landlines so that for much of the world voice telephone will be from wireless mobile; with a wireless mobile device for voice telephone, can also, at modest additional cost, have sensors for temperature, GPS, audio, video, etc. and offer full Internet access, the Web, and apps.So, the big explosion of mobile is still mostly just for portable voice telephones with the cheap apps along for the ride.In particular that mobile devices are outselling desktop computers is not surprising or important for desktop computers: Oranges, apples, and bananas are no doubt outselling both. Yes. some poor desktop users who used their computer mostly just to check e-mail may decline to buy a new computer and just use a mobile device instead; they never were good computer users.Dixon’s observation that for the apps “the rich get richer” is likely true, at least for now. But, really, there should be a way to let users of small screens still easily get to a few thousand apps. E.g., on my desktop computer, I have something over 150 of each of command line scripts, editor macros, and TeX macros and something over 80,000 files, plus more that are older, and have no difficulty finding what I want right away. If an iPhone or Android device is designed so that only a few dozen apps are easy to access, then the design just sucks and can be improved.The observation that, on mobile, apps are beating Web browsers is not so impressive: One Web browser can easily get to any of 100+ million Web sites, and that’s a biggie.What’s going on here? Web browsers with HTML, CSS, JavaScript, etc. have a big, huge advantage: Nearly every computer has a Web browser that supports recent standards for HTML, etc. and, thus, can get to 100+ million Web sites. Nearly all the Web sites are written to drive such a browser because 1+ billion Internet users have such a browser.Or, all the users have a Web browser because all the Web sites assume one; all the Web sites assume a browser because all the users have one.Net, a Web browser is the biggest example of a ‘network effect’ in all of history, that is, maybe more important than the voice telephone network, the Roman alphabet, and the English language! :-)! A biggie.The Web with its 100+ million Web sites and standard Web browsers have another really big advantage because together they have heavily ‘standardized’ graphical user interfaces (GUI) and, thus, largely eliminated the Tower of Babel possible and inevitable otherwise. To be more clear, it can take full time for two weeks or so to ‘explore’ and get good with the ‘idiosyncratic’ and nearly unique GUI of Microsoft Word. So, to use Word a user has to invest two weeks just to use that one program. But with Web browsers, a user can invest less time than the two weeks to learn to use the largely standardized GUI of a Web browser and, then, quite quickly use any of a large fraction of the 100+ million Web sites.Imagine if each user had to invest two weeks of their time, like they do for Word, to use each of the Web sites they want to visit? If so, then the Web would never have grown to anything like its current importance.So, on mobile do apps threaten to kill off Web sites? In two points, no:(1) So far a large fraction of Web sites have not been written to look good on small screen mobile devices. Indeed, a large fraction of Web sites do not look good on laptops or anything! In particular, at present there is an OCD ‘JavaScript’ riot to make each Web site have an idiosyncratic GUI, e.g., to make its screens jump around, that takes too long to learn, and too often is not usable at all, and that nonsense will pass. E.g., (1) far too many Web pages have single lines of text 120+ characters long. The Web developers believe that all their users have screens the size of outdoor billboards? Demented. Deranged. Dysfunctional. Destructive. Where do they get that really strong funny stuff they’ve been smoking? E.g., (2) “Do you really want to leave this Web site?”. So, click, click, etc., and finally maybe just chop power on the computer — that’ll stop that nonsense. Too many silly hackers in the third grade are having way too much fun with JavaScript:”Look, Ma, what I can do with JavaScript and the DOM!!!”.”Right, Son. You can totally torque off 99 44/100% of the users who might have returned to your Web page for a second time.”.Darwin with just A/B usability tests stands to be on the way except he is so busy with his huge backlog of other such much needed work!(2) If each Web site had to be accessed via an app unique to that site, then that would be too many, 100+ million, apps to write and install, and we would be back to wasteful, often unusable, idiosyncratic GUIs (assuming that an app for the GUI is more difficult to write than the HTML, etc. for the GUI), quite different for each app and Web site (assuming app coding for a GUI has less discipline than HTML, etc. coding). Bummer. That dog won’t hunt.Yes, there are many mobile apps for this and that little thing. Right, in grades 5 – 12 and in college, boys can get SnapChat for selfies from girls looking for attention. While I don’t have SnapChat, I have to believe that a major fraction of the person hours at the NSA is looking at such SnapChat intercepts; I mean, what else are they going to do to have more fun than that?Still, and a biggie, now and coming rapidly, for each user, necessarily nearly all the important data is not on their mobile device but on servers on the Internet, and, then, the question is what GUI to let the user get to the data? Due to the biggest network effect in all of history, far and away the easy answer will remain the GUI of a standard Web browser.Apparently many people have long seen this point of the importance of a standard Web browser with full clarity: Sure, some one company might want to try to arrange that to use the Web everyone must have their operating system and Web browser connecting to their Web server. Tried? Yes. Successful? No. Why? Because far too many people saw the danger and worked to keep the Web ‘open’.Now the Web with ‘open’ standards and standard Web browsers is far to big to attack. Better try something much easier like trying to level the Himalayas! :-)Yes, the world of small screen mobile devices wants “their story told”, and far too many in the media desperately “want a story to tell”. So, the story is, “Desktops, PC, and the Web are dead; mobile is replacing everything!”. Anyone really believing that nonsense should be especially interested in that great deal on a bridge over the East River.In simple terms, what small screen mobile devices are doing is providing portable telephones with some additional utility and a lot of ‘gimmicks’, but in no important way will a small screen device without a good keyboard replace a computer with a much larger screen and a good keyboard. Instead, in the industrialized world people still need to get actual work done.Since the smart phones, etc., are so new, and with amazing audio, video, GPS, etc., it has been possible to write a lot of apps.Gee, maybe write a tachometer app! Aim the camera at a rotating object that puts out a periodic signal. Well, that signal, with astoundingly meager assumptions, can be represented to essentially arbitrarily high accuracy by a sum of sines and cosines of a ‘fundamental’ frequency and whole number multiples of that frequency. To find these frequencies, take the signal and take a Fourier transform and the power spectrum and, then, look for the peak of the lowest frequency. Presto: A tachometer. Do have to worry about Nyquist sampling rates!Still, really, (1) a small, mobile device is still mostly just an end user, ‘client’ device for much more on the Internet. (2) For a huge fraction of what is on the Internet, the easy, efficient, and effective GUI is just from a standard Web browser, not an app unique to some Web site.

  40. Druce

    Getting here a little late so maybe somebody’s already said it but isn’t it partly because you can do an Instagram or Whatsapp and get to 450m users, sell out for billions to Facebook without venture investment?Every 2-bit app asks for contact list access and everyone grants it. And with AWS etc. infrastructure is a solved problem. So the top apps can go rocket ship viral without anyone’s permission or a VC investment.Sucks a bit for Facebook though, they were the ones who were supposed to control the social graph and everyone was going to build apps on top of their API. Now they have to shell out billions to buy these rocketships. But I guess it’s to their credit that they are doing it to remain relevant, if not very promising for their investors.Tough for early-stage VCs too, you have to identify those superstar entrepreneurs and get in super early and the power law distribution is even more insane than before. Which favors a YC shotgun model as opposed to a VC model of trying to get behind the eventual winner in each niche and leverage the VC’s network and deep pockets to get them there…but for those top consumer apps super-virality is all that matters.

  41. Joe Michels

    It the bubble of consumer apps is about to bust, fantastic! After the bust consumer apps will return with gusto. Just like post .com bubble burst produced some of the most important web companies, after the app bubble burst, better apps will be produced.

  42. Sofia Fenichell

    Great post. Really great comments. There are only two things that will define a successful mobile app in my view going forward and that is either 1) must have content or 2) must have utility. I agree that the app ecosystem has suffered from too much fragmentation. There will be more rationalisation as returns go down and content providers and VC congregate around platforms and utilities that are more game changing within their respective sectors. Plus frankly the concept of having an app for every content provider or use case is really exhausting for the consumer.

  43. leigh

    i’ve had a couple posts on closed ecosystems related to mobile and looking back i think my thoughts are still true both about android and the carriershttp://leighhimel.blogspot….http://leighhimel.blogspot….

  44. soulgirl*

    I’ve always wondered, are the games that people play on their iPhone regarding things like…FIFA or bejeweled based on plain algorithms or is there some connection between the supporters gained and the people behind the invention of the app? Say…if you play a lot of FIFA, does this mean you are supporting the players in a real life “virtual reality” reality? Or if you play bejeweled do the colors matter in supporting the algorithms…

  45. Jack

    You’re betting the house on Push Technology in 1998.

  46. Aaron Klein

    I’m not saying it doesn’t work for some. I’m saying that the changes in mobile are a disadvantage to freemium, not to innovation as a whole.

  47. Matt A. Myers

    But using the cohort of the fastest growing possible companies – because in part they’re free and mobile – isn’t really the best way to measure where things will go.

  48. JimHirshfield

    Please keep commenting; we need this adoption.

  49. PhilipSugar

    I look at the free model in two ways.1. Free is basically your marketing expense. This isn’t new: Give away the razor sell the blades, free samples, etc.2. I give you value, you are the product. This isn’t new: radio.Now what is new is giving it away free and figuring out the model later.

  50. Aaron Klein

    Freemium != InnovationYou can build a great business without bowing to the powers of mobile gatekeepers โ€” and you can even do so on mobile.

  51. JimHirshfield

    Utterance of the day.

  52. Aaron Klein


  53. Richard

    The total ad spend is 500B, mobile is still just 5% of this. Free still has a long ride ahead of it.

  54. Aaron Klein

    What do you mean in-app isn’t freemium? Of course it is.Install the free app, and that’s how you get to my revenue source.I agree with this post โ€” it’s Fred’s mobile version of “don’t be a Google bitch” except there are no great solutions for apps that need to have free distribution.My point is we should be cautious about considering that innovation as a whole. That’s one small piece of innovation. Where we generate the revenues to pay for building our own distribution, we’re still thriving, even in the mobile world.

  55. Richard

    There is a great algorithm that I use for forecasting. It is called the dynamic linear model. I believe It was developed to forecast Nazi submarine locations in WWII.

  56. Matt A. Myers

    You don’t think those $500B spent will be reduced greatly? Online advertising is cheaper, more effective – if done properly.

  57. Aaron Klein

    To be clear, I don’t believe that at all. I have some designs on consumer stuff that I think could be great. And I deeply respect those who have done it. It’s very hard.But I felt like Chris and Fred’s posts effectively said “because unfettered free distribution doesn’t exist in the mobile world, innovation is doomed.” That’s what I disagree with.

  58. Salt Shaker

    Yeah, but good luck trying to crack the agency inner sanctum. Mobile spend still growing, but increased market fragmentation is the enemy for any ad rev dependent start-up. Ad biz becoming too reliant on programmatic buying/selling, which is commodity based.

  59. John Rhoads

    This isn’t about monetization opportunity but instead how do you grow your installsNo Google Adwords equivalent for mobile really and and an arguably overwhelming advantage to the big apps of yesterday

  60. JimHirshfield

    Launched! I buy you lunch for every 500M PVs/month you bring on-board.

  61. JimHirshfield

    Fries please

  62. JimHirshfield

    Did you ever expect to get rich off affiliate marketing? You strike me as more of an entrepreneur than that.

  63. LE

    You know the idea that I want someone to do is called “tell ceo”.It’s where ordinary users have a channel to the CEO where they can give feedback that in some cases isn’t practical for the CEO to get by walking around.It’s a hub and a filter making sure the CEO is only supplied with grass fed commentarybut not people wondering why they can’t return something 2 days past the return policy (although that data would be compiled and presented in aggregate). [1][1] As a kid I observed several businesses try to post the number to the CEO’s office in the store with “call me I want to know if anything isn’t right”. Those signs almost always disappeared my guess is because it became just a support line which wasn’t the intention.

  64. Aaron Klein

    Oh, I still have lots of customer contact. I just used to have contact with 100% of customers and I sorta miss that. Gotta scale though.Speaking of which, my calendar just buzzed and I need to make a bunch of customer calls now. ๐Ÿ˜‰

  65. Aaron Klein

    I have tried to build this for myself by giving every customer I meet my business card, as well as sending each one a personal note (as personal as can be in these circumstances) with my business card.I don’t get boatloads of email from that, but I do get some, and it’s invaluable.

  66. awaldstein

    People write to CEOs and execs all the time.If execs list their emails they will get them.Whenever a customer got to me when I ran the show, it was a guaranteed response.I wrote to Jeff Bezos a few months ago telling him why I think his wine biz will fail for the third time. He responded.Point–this channel is open to everyone now.

  67. JimHirshfield

    I bought those. They don’t work. I want my money back.

  68. LE

    There is so much friction for someone to get to the CEO. And giving an email address obligates acknowledgment or there is bad will.And how many people are going to take the time and trouble to write a letter? (What office is Jeff Immelt in anyway and when?) And if they do it would have to be something that is painful enough (or they are just looking for positive reinforcement) that they will take the time.You know when my father was in the hospital recently I observed a boatload of shit that was wrong (non medical). And I couldn’t believe that the head of the hospital would let these shitty things go on. And I’m pretty motivated to let my feelings be known. But in the end after the stay ended I didn’t do it.Why? Because I’m not in the business of taking my time to help other people’s businesses (as much as I enjoy it and you I would guess are coming from the same place) and in the end you kind of feel that you will just end up getting some curt corporate answer and end up feeling “why did I spend the time to do that”?As far as Bezos responding either you really hit it out of the park (would love to see that one perhaps you can do a blog post on that or post here a summary). I mean it’s hard to believe that with the amount of orders Amazon does that Bezos has time to review thousands of letters that arrive if statistically as many people write as have actual problems or suggestions with Amazon.Like “hi Jeff, I have a hard time figuring out which indoor security camera I should buy. Here is what I’d like to see… – thanks LE”What do I gain by doing that exactly? My idea is to provide possible benefits to draw more Arnold’s and LE’s out of the woodwork. Not just bullshit karma.Other question really is why did you have to write to Jeff. Why didn’t Jeff have an active program to get info from you? (Not my idea stated just wondering). Maybe he does. Why didn’t they know what you know?

  69. ShanaC


  70. awaldstein

    My take is just different LE.No one writes letters. Many write emails when they are unhappy. Many express on the social channels when they are happy. Each takes a few moments. You want to encourage both.If you want info as a consumer, almost all companies provide channels. Chat, email, call. Real time chat rocks and is way cheaper to provide.My point–the channels are already there. How you present them makes they get used or not.Great companies–and Amazon is one of them–make it easy and they get used all the time.

  71. ShanaC

    some people do, though I still don’t fully understand how

  72. awaldstein

    There are a bunch of levels to this:-access to company execs. Almost everyone makes themselves accessible,-feedback generally. Completely different, One of the regulars here has a company that just may crack this open. They’ll chime in if they please.

  73. PhilipSugar

    Not sure what you mean. Btw: Your stick your finger in the air and keep flipping the burgers made me laugh so hard I nearly cried.I am spending time in forecasting hell and remembering why I am unemployable.I don’t think my comment of: Which do you think will help our sales: Each of us calling ten new potential customers or spending that time masturbating with ourselves and excel? Went over to well.

  74. PhilipSugar

    I think the problem with not knowing upfront is that you optimize differently for each.The beauty is my team doesn’t get involved, its those above me that hear it, that is why I’m unemployable.