1 Billion Pageviews
Yesterday our portfolio company Tumblr announced that they had quietly done a round of financing with the existing investors. And they also announced that the Tumblr service is serving over 1bn pageviews per month.
I thought it might be useful to put 1bn pageviews in perspective. I went into comScore and looked at how many web properties serve over a billion pageviews worldwide each month. The number is 57. Here are the comScore stats on that. Monthly pageviews are the farthest column on the right.
Now, it is important to state that we are comparing apples to oranges here. Tumblr is not measured directly by comScore. The data in that first chart comes from Quantcast. And comScore does not roll up all the Tumblr pages that run on other domains. Quantcast does that. If comScore did roll up all of Tumblr and if it measured Tumblr directly, it might not report 1bn pageviews per month. Or it might.
But that's sort of beside the point. 1bn pageviews is a seriously big number. Only 57 properties in the world report that kind of number on comScore.
And Tumblr got there on $5mm of invested capital (we just put in another $5mm).
There are several reasons why they were able to do that. First David and Marco are an amazing team. They did all of the work themselves for the first year. And now they have a small team who maintain the service and develop new features. They get a lot done with very little. And finally the service is amazing. The Gotham Gal started a tumblog this past weekend. She has blogged for almost seven years so she is no novice to this kind of software. She got her tumblog up in about five minutes and posted a bunch of stuff. And then she turned to me and said "wow, this is so easy."
There are some lessons here. First, make your software super easy to use. Second, you don't need hundreds of employees to build a big time web service. You can keep it lean and scale if you have the right team. That's how Tumblr got to a billion page views and we just made a bet that they will be able to take that number a lot higher.
Comments (Archived):
I met David, Marco and John just once. They’re among the smartest people I’ve been able to meet. It took me a few minutes t convince them that they needed a PubSubHubbub hub, and a few days later, it was live : http://tumblr.superfeedr.com/ I wish it was as easy to work with everybody!As for the “small” team, I remember reading somewhere that it was something that mattered a lot at GOOG. Rather than 1000 good engineers, they want 10 top-notch engineers. Not only the “individual” performance is much higher, but you also remove all the overhead costs that bigger teams have.
small teams are absolutely the way to go. as you grow, you need to break your engineering and product people into small teams. think of it as a lot of little startups inside a big company.
Let’s push that even further. Don’t you think that at some point big companies (with hundreds of thousands employees) will even disappear? Maybe I’m mistaken, but the “biggest” companies of now aren’t smaller than the biggest a few years back?
i don’t know about that. exxon probably gets bigger
But Exxon is “old”. I think Apple has now almost the same MarketCap as MS but 50% less employees.
and crapple is the worst. imagine a company that truly embraces and empowers the development community. such a company will have a very small core.
Apple’s mandate is not to embraces and empowers the development community. Their mandate is to milk that community on behave of their shareholders & customers while throwing out just enough feed to keep the milk coming, hoping no one else executes a better implementation of that balancing act. Anyway, we have bigger fish to slay.Next up to the milking bar FaceBook with their centralized hyper-commoditization of all possible organic social relationships.It will be all over but the crying before the average netizen becomes viscerally aware that the vast potential for meaningful, effective, organic-social-relationship structures has been abandoned in favor of a centrally-commoditized-profit-driven-privacy-invasion, sponsored & enforced, by a corporate army of robot algorithms that largely leave them and the public interest outside of the social-capital value equation, looking in.Why waste social evolution on stupid stuff like an Omega Point driven Noosphere when with little or no social awareness we can effortlessly stumble into a massive commercial cluster f*** foobar!So yes, I hear you on Apple but they are not the alone, to big to be good for us, ecosystem problem. Lets not get over focused on them specifically.To me this is where small socially visionary developers have an opportunity to reach out and catch the backlash, back wave of web users that will want to pay a modest fair price to escape that commercial over-world.Speaking of Tumblr, I love this thing. Great tipping point between easy and flexible. I love the non-commercialism. It is one site I would like to pay to support but they won’t let you? Please let us pay! I’m worried it will be bought and commercialized to death just as I start to invest more effort into it.ONLY THREE BEEFS WITH TUMBLR SO FAR: Work around suggestions welcome!- cannot make home page point to a static or tag driven page- when going to your dashboard it always returns to your default or master blog instead of remembering which blog you are presently working on- it needs a new posting type = hidden-post with operable tags , then using a tag base page you could create menu driven tag pages that only show given groups of posts via your menu like an ultra simplistic CMS. Even better a user settable pageURL #ing suffix that can be used to control the sequencing of those tag based posts on such pages. Now you have got a very simple way to mimic id base blog post sequences in a CMS like Joomla.
Point taken but it’s apples and oranges. You can’t drill for oil from a computer terminal, and you can’t transport oil via email. And I can tell you from my days back in the Big 4, Exxon is extremely efficient at what they do.Not sure about MS vs. Apple though. Just saw a special about Bill Gates and Steve Jobs on CNBC, which made the point that in the 80’s Apple decided to focus on hardware while MS focused on software. That probably has something to do with it.
but you will be able to…
For tech companies making a ‘soft’ product, then yea they can stay small the more they take advantage of good talent and the tools at their disposal. However, any company that has any hardware, or services a piece of hardware will need to have numerous employees in a variety of positions to handle it. I think you are on to something though. Why can’t these largest of companies scale back on certain aspects with efficient use of the newest and best software. Can’t a thousand engineers at Exxon get by with a lot less overhead if their HR and supply chain departments were run more like a startup?
You got it Scott. I’ve been preaching small and distributed as the advantage of the network economy for a few months. The trick is making a live business example ;), words only go so far.
Couldn’t agree more Julien. I see great advantage for lean, small and distributed companies. The parts of business that need BIG offices will be handled by specialized efficient external entities.Awesome how the Tumblr team could see the advantage of Pubsubhubbub for distribution, but with the type of market they have, why not setup their own publisher?
If they start setting up their own “anything” they are going to need more people and they are not going to be small anymore. It can be the right decision, but they should to be really selective with what they bring inside.
Fair enough, but I was under the impression setting up a Push publisher/hub isn’t a major task. I personally never set one up so I could be mistaken.
I’m afraid I can’t answer that. I have no idea if its a major task or not, it was just a general thought.
Here’s one way to do it, and leverage the Google App Engine as publishing hub.http://intertwingly.net/blo…I suspect for bigger flow models you may need several hubs.
Speaking of ‘specialized efficient external entities’ which I would assume to essentially be really good contractors. Do you think in the future, for large companies, that vertical or horizontal integration will not provide the advantages it has in the past century? Will ‘in-house’ arms of a company, such as legal, HR etc., no longer be worth it?
I suspect any larger revenue business will want inhouse legal & financial services. But the number of required people to perform and maintain the knowledge of these areas may shrink.
…especially with the rise of outsourcing. Those functions may not go completely go away, but the internal teams may become leaner.
That is the way Google does it. They think three is that optimum number. So they have lots of teams that are three strong. Curiously, USV also is a three person team. How did both Google and USV come up with that same number? Why not two? Why not four? That also makes me think, could USV add three teams of three beneath its current crop of the three dons? I have thought about this since I applied for the USV job of Intra-Portfolio Evangelist. http://technbiz.blogspot.co…If you manage 300 million now, maybe you can manage a billion. Give yourself a pay raise. Or maybe that is unnecessary since you have already embarked on your best decade yet. The 2010s: your harvesting season. Wait for your first few IPOs, then compare yourself to those guys at Goldman. Yeah, I read that New York magazine quote. 🙂 http://technbiz.blogspot.co…At this point I don’t know how much of the harvested money goes to those pension funds, and how much stays with you. But I’m sure it is a good cut.Happy 2010s, Fred Wilson!
Julien, Can you describe the roll of Superfeedr and publishers. It appears I’m confused on the model/framework you have constructed. What does it take to setup a Pshb publisher and hub? I used a wordpress plugin and it added Atom feed support and connected me to Google’s publisher for free. How does Superfeedr play in this area?
You could simply have your very own hub, instead of using Google’s 🙂 We host hubs for mant many publishers out there. We hope to have soon a way for bloggers like you to use our hub too!
Thanks, will RTFM 😉 and figure out the details. Appreciate the rapid response and your work to get the web real time asap!
WOW. In the company of Disney.com and Walmart.com. How long since they first started working on tumblr?
we invested in Oct 2007. i think they had been operating as a side projectof David and Marco’s for about six to nine months before that. so maybe theystarted at the beginning of 2007??
Getting to 1Bn views in that short a time period to me is as impressive as getting there in less than 5mil. Implies that they made a lot of great product decisions (starting with the easiest signon process ever) and kept mistakes to a minimum.And of course David is a great evangelist for tumblr: all you need to do is take a look at his tumblr page to want to try. 🙂
David Karp sure has a great tumblog. It is almost like he created the service for his personal use. The dude has an amazing sense of humor. http://technbiz.blogspot.co… (I hope I am not overdoing this link thing here. I say something, and then there are more things I want to say that I have said already, and I just go ahead and link to that particular blog post where I said it.)
he did create it for himself.paul graham talked about that in this posthttp://www.paulgraham.com/o…
Oh my, wow. It’s like you hit in the dark and still hit the sweet spot. Wow. I say, wow. The guy is mighty impressive. David Karp, pass on some of those page hits to my blog, will you pretty please? 🙂
This is a rough comparison. But this reminds me of that Giuliani anecdote. “…. as if he was an altar boy,” he said. Ends up he w-a-s an altar boy.
Oh wow. That Paul Graham article is really something. So what Karp did is the norm and not the exception. Wow. I’m sure I will come across a Karp event some day.
Jeez, look out when they hook a financial model. There’s going to be some incredible possibilities. How long and how much money did it take Facebook or Twitter to get to that page views level?
Congratulations on the round, love Tumblr and the team, but you know that already.
yes i do know that 🙂
can I just recomment this comment? 🙂
Sure feel free, Daryn 🙂
Press the like button.
Wow! That rocks. I love Tumblr a lot, too. The older I get, the more I appreciate simplicity and large type 😉
I often think people forget how much people like large type at all ages…
Right on!Is there a shortage of pixels to go around or something?I guess they just like to dangle as many hook in your face as possible. That just sends me off looking for a more relaxing less crammed visual environment like the tumblr dashboard
this is an impressive milestones, Fred. But page views seems to be a dying metrics (although still used by media agency) because in many times it can be gamed (not saying this is the case for tumblr) or not enough to measure the quality of a media platformAny other engagement metrics you could share? (time spent, repeating visitors, active blogs,…)
http://www.quantcast.com/p-…
Pro response. Salivating for a few seconds.
Impressive. It will be interesting to know your views on Posterous. They are much smaller (38M Global PVs – http://www.quantcast.com/p-… but have been growing at a tremendous pace. Plus they are a small team too.
And a direct competitor! When I started my blog I had doubts between both services and finally I chose Posterous, but I’ve never been really sure if it was the right decision because both services seemed very similar to me.
Swap to both. You can post by email, and let each of their networks duke it out.
I thought about it, but I didn’t do it because Google could consider it duplicate content and penalize. However, with so many tools that make it easy publishing everywhere (Posterous lets you autopost to other places) maybe they’ll change that in the future. Anyway, I should probably think more on posting more frequently and less about tools.
That’s the spirit Fernando. Focus on the quality of your content and gettingenough eyes on it to help you make it better (if your goal is to expandvisibility).The question I’m most concerned with is not how many people visit my blog,but are the right people reading, commenting, and making my views morecomplete.
What makes someone the right person- i try to think I can learn from anyone…
The people that share some of my most important world views:-free and open availability of information-free competition among new players and embedded players-coffee is good for mankind ;)While you and I can both learn from anyone, we can learn faster and be more productive working with a subset of everyone.For instance, you could get a government job where things change very slowly, people are generally unmotivated to risk an untested idea or take responsibility for anything. Or you could work with a hyper motivated startup where things are much more intense, risk is where it’s at, and individual heroic efforts make or break a business.Both of these two groups could help us learn, but one may help us learn faster.
Actually, I learn most with people I disagree with and people I look up to. Not that I think I should be constantly frustrated (I think it will wear me out), yet I purposely do read and look at material that is for example, Marxist in tone about various subjects. At the same time, I’ve read Ayn Rand. And I can see myself as I get older sitting down and reading Smith and reading through Marx, Plato, and Aristotle. (for some reason I never read Smith and only read a touch of of Marx and Engels, something about the way my courses were constructed).I can’t promise someone I am definitely right, I’m only one point among many.And while I don’t question risk and some senses intensity, I do question heroism. Responsibility != heroism. Doing the right thing, especially above the call of duty == heroism. Those acts are often small.Partially, this might be because I see risk as part of the inherent structure of the business. Doing the right thing, probably not.
A meta comment about what I said- there is something about this that reminds me of JLM…he is very wise…
Agreed, I enjoy his attention to detail and history.Also the right group is a diverse one. Intelligent people often come to different conclusions, and it’s important to try and understand why (important to me at least)
so great minds DON’T think alike! go figure!
My own narrow views require a healthy dose of well reasoned dissent 😉
well said shana. it seems as though there were too many anti-heros over the past few years…
You, Shana, fit the bill. You are that right person.
*sigh* I don’t believe in right. You have to work at right, it’s a growth field, always.
You’re both have invaluable perspectives. I’ve been enjoying your posts recently Paramendra, keep up the great work.
By posts you mean my comments? Thanks Mark, means a lot to me. You are a Regular. That is a badge of honor here.
No, I mean technbiz.blogspot.com :D, your blog.
Oh, wow, Mark, you honor me. I am trying to see if I can do the blog daily thing that Fred does. In my world it translates into making sure I put out 30 posts a month, preferably more. Because I can’t do the every morning thing that Fred feeds to his hungry masses.
i see them as a fast follower. that doesn’t usually work when there is a strong category creator but they’ve done ok.tumblr is more of a social network than posterous. at least that’s how i think of it. i don’t use posterous other than a few trials so i could be wrong about that.
Posterous and Tumblr are different in pretty fundamental ways. Disagree 100% with the fast follower label here. Obviously a bit of a biased view, but no different from the other side of the table.Our worldviews are profoundly different — in our approach to users, who our users should be, and what they want to be doing.
Are there any other significant countries in the breakdown not listed Fred ? – there’s approx 280m not accounted for in your first chart.Also as Disqus is another portfolio company is it not worth there being some sort of cross promotional activity between the two companies (unless Disqus are going to expand the Disqus profile)
disqus has become the “default” comment system on tumblr because the tumblrteam are not fans of comments and they are not a feature of the servicebut many tumblr themes support the disqus “shortname” and so it’s simple toadd disqusi agree that they could do more together
Any reason they are not fans of comments? As we see here at AVC everyday, there is tremendous value in the comments/discussions.
Just speculating, but I guess they prefer people setting up their own tumblrs to express themselves and reply to others.
That’s a great point. Interesting that leaving out a feature has perhaps helped them register more new users.
When you reblog, you have the option to comment in the body of the post itself.But, hey, I am a huge fan of Disqus. I personally use Disqus more than Tumblr (Karp, blame Wilson), and my blogging platform of choice is Blogger, not Tumblr, not WordPress, not Facebook, not Twitter.I use Tumblr to listen. Blogger is where I do most of my talking.
There has been plenty of discussion about commenting versus reblogging. Reblogging takes both the original content and your commentary and exposes it to a new audience (yours), while a comment is only of benefit to the author and their own community. Both have their place.By the way, Tumblr does now have “replies”, but they can only be left from the dashboard (so you have to already be a tumblr user), aren’t threaded, and aren’t quite a comment.
Reblogging and commenting are similar and different.
yes, they view the reblog as the opportunity to comment
david thinks this community is the exception that proves the rule that comments sucki don’t agree. we are investors in disqus.but he has a point that many comment threads are awful
I wish David saw a tumblog as the place where a group of friends meet up to have quick exchange of thoughts on a topic that one of them has brought up, then comments add a lot of value to the tumblog. Its a short burst asynchronous multimedia chat with your friends. 🙂 A tumblog without a comments section is like having a one way conversation with your friends. 🙂
I agree with David that sites like A VC, cdixon.org, Hacker News, reddit, etc. are the exception to the rule that comments suck. The key requirement for quality discussion is quality community, and no startup is ever going to commoditize community.Disqus is the best commenting system for heavy discussion, but it’s overkill for sites without community. How many low traffic blogs have you been to with Disqus installed but no discussion? I don’t have any numbers on this, but I feel like this describes most blogs I come across. (This hurts the Disqus brand IMO.)I don’t like having comments on my blog because when visitors see that there are “No Comments,” they automatically think that no one cares what I have to say so they shouldn’t either. Maybe they’re right that I have nothing interesting to say, but they come to the table with this preconceived notion — we all do this.I think it’s unfortunate that we’ve come to judge the importance of things by the number of comments, likes, or upvotes when so much good writing, insight, music, etc. is left unappreciated in the shadows.I wonder what we would be reading or listening to if it weren’t for likes, comments, retweets, and every other groupthink social ranking metric?
Re: blogs w/ low comment traffic, we’re working on some things around lighter weight participation and addressing the ‘cold-start’ issue. Would love to hear if you had any specific ideas as well – short of not having commenting at all.
Wow, it’s cool that you guys are trying to address these things.For blogs w/ low comment traffic, instead of showing the big form, OAuth buttons, etc., you could just show a text link or small button that triggers all that to display when clicked. Anything that looks subtle and doesn’t distract from the post would be ideal, I think.(I’m sure you guys have thought about this way more than I have.)
+1Mark Suster is getting great comments and so is Dave Winer, I’m sure Disqus is playing a big part in this as it makes it so simple to comment.Another common theme that plays a huge role is that, like you, those two both read and reply to their comments in a timely mannerIf you really want to blow the comments suck argument you just have to look at the Daddy of them all – Facebook, it’s a major part of their functionality. I post everything from my personal Tumblog to Facebook and my friends comment in Facebook.Everyone likes comments on their posts. it appeals to their ego (except the snarky one’s a la Techcrunch. Mike Arrington so badly needs Disqus and to turn off anonymous commenting) and it’s proof that you are not sounding off into the echo chamber.
Very impressive. Perhaps even more impressive is actually keeping it simple. As in any art form less is more.
While it makes it harder to gather traffic stats, the ability to use your own domain for your tumblr page is pure genius. Instead of looking for just the right username or alias, you can use an existing domain or have the flexibility to find a new domain that you want. Of course, you could always use a domain redirect or iframes, but like everything on tumblr, they make custom domains so easy, you don’t need to explore any alternatives.
Congratulations to Tumblr for their enormous amount of page views and 2nd (?) round of financing. I have recently become acquainted with it and can attest to its quality. Recognizing that everyone has the desire to blog and conceptualizing such a user friendly product certainly deserves praise. Incorporating social networking into it was icing on the cake.
I’m a heavy user of the service (find me at caterpillarcowboy.com) and I think David and Marco have built up an amazing community.But I suspect the case for the very small startup works for Tumblr because they are in an existing market with existing large players (WordPress, Blogger). If this was a new market (e.g. Groupon in group buying) would it make more sense to raise a lot of money, build up a larger team, and grow as quickly as possible? I realize that more engineers doesn’t always translate to more results (Mythical Man-Month), but when done right with lots of small semi-autonomous teams, it can be successful. Should Etsy have stayed < 10 people? Gilt? Zynga?
Fred, the next terms sheet should require a physical sign outside of Tumblr HQ that says: TumblrOver ___ trillion served.
My cholesterol level is having a Pavlovian response.;)
great idea
This is a great example of how previous hurdles for start-ups (cost of infrastructure, lack of s/w tools) have come down significantly, allowing small teams to do big things. It seems that the big, heavily funded start-up is a thing of the past. Whales of the past have been replaced by leaner, faster moving tiger sharks.
The infrastructure costs will come down a lot more. Google is obsessed with infrastructure and it’s years ahead what’s available for everybody. This touches every single aspect of engineering, from concepts to full blown services. Some of these already tickled down to the open source community (rpc closures, protocol buffers, FDO and LIPO compilation for gcc, etc)In the last year, I had to build a scalable approximate nearest neighbor service almost from scratch. While it was fun, it certainly took time away from working on quality on the data. Plus it’s still nowhere close to being optimized to the extent of the established services.
Are pageviews really the key metric here?
No, pageviews aren’t really the key metric, but as Fred said, they do give some perspective.You can see many more impressive metrics on the directly-measured quantcast page:http://www.quantcast.com/p-…Or on this tumblr staff post from last month:http://staff.tumblr.com/pos…2 Million posts per day15,000 new users a day18 new posts & 5 reblogs per second
They are the key metric for a media company, which is what Tumblr is, a media company.
That is over simplification. How do you compare the value of someone watching a show on abc.com to a tumblr page view? Also, the media that tumblr users create is a new kind of media. People still need to put that into contex and need to determine the value of a page view on tumblr.
Start with the fact that I don’t own a television, never have. But I do consume a lot of video online. Sometimes that is clips from TV shows. You could argue time spent on a page is also important, not just page hits. But you don’t get to argue Disney page hits are different from Tumblr page hits. Tumblr competes with Disney for my attention.
Here are some examples on why not all page views are equal:1) lets say you are watching an episode of lost on abc.com. You are a captive audience for 50 mins. You are shown the same 45 sec ad 4 or 5 times during the course of the episode. There is recall of the ad atleast for a little bit. where as if you spend 10 secs on a tumblr page looking at a new pic someone posted, your attention span is completely different. Even two sessions you have on tumblr might be completely different. If you just started following someone on tumblr, you will spend a lot of time on their page, while you will just spend a sec on fred wilson’s page as you are up to date on it and will only look at the latest addition.2) Similarly, i think a pageview on walmart.com is different from a page view on disney.com . While you are at walmart.com, you may buy something right away. So its a different value.We can agree to disagree on this.
I think Fred produced a list (by someone else) that jots down the names of the less than 50 or so web properties that get a billion page views. Those people are not differentiating between the types of page views. That is all I am saying. This discussion resulted from that list and that is why I am sticking to it.
Start with the fact that I don’t own a television, never have. But I do consume a lot of video online. Sometimes that is clips from TV shows. You could argue time spent on a page is also important, not just page hits. But you don’t get to argue Disney page hits are different from Tumblr page hits. Tumblr competes with Disney for my attention.
I think there are more important business levers that Tumblr has been focused on and the growth in page views is an offshoot of that. I think user growth and # of posts are more important focal points that will naturally lead to more organic page views. Grow the # of users and make it easier for those users to become more prolific bloggers.
Agreed. We agree.
I agree that its ambiguous what the metric should be. A post on tumblr could be the equivalent of a “status update” or a pic on facebook. So do you count someone checking their facebook page as having as many page views as there are status updates on the page? :)But in my opinion, for a young company like tumblr its amazing just to be in the same sentence as walmart.com or disney.com for any metric.
no, it is just one metrici think the key metric is the number of engaged usersthat is a bit tricky to measure for tumblr. should that be the number of tumblr users who are active every month? or should it be the number of people who visit a tumblog every month?
I think the best analogy for tumblr is to view each tumblog as a channel and each person who visits a tumblog as a viewer.
Isn’t the key metric, ultimately, profit? Sorry to get a bit “re-work” here but I just guess that, with an extra $5M the Tumblr team could (maybe) become the 30th most visited site in the world.But they already have the traffic (like you say, they’re number 57 now!). So surely, the key metric now must be ensuring the business is sustainable and can make cash?
That’s what I really liked about craigslist – they are a small group of people but have an amazing impact online. That impact lead to them (at the time I read it) producing $2m profit per employee, which I see as a huge win because they can be so sustainable there. This thought goes along with the idea that sustainable impact (value) is the metric though, I suppose.
I’m finding it interesting as a service- it’s nothing like blogging as I would imagine it- it’s much more like “Share this plus a short comment if you feel like it” which seems sort of natural on the internet. It’s not so predisposed to conversation. When I think of blogging- I keep thinking of something long. Many times when I see Posterous I see something much closer to traditional blogging. So I am not sure if Tumblr should be called a blogging software (though you can do that) as much as a sharing software…
I think that that goes for Posterous too. Most users just share some photos they send from their phone or clips they find online.
I need to run with another web crowd then to go see 😉
The dynamics of Tumblr are incomparable. I feel like the relationships I have with the people I follow and who follow me on Tumblr are much ‘deeper’ compared to the ones I have on Twitter. I have ‘met’ so many great folks through Tumblr, being based in Berlin it has opened my network to many people in the US and beyond. The Like, Reblog, Reply, Ask features and integration of Disqus have played a big part too. If I recall it correctly, it was @pegobry who once said (also a Fredland native): “Posterous is an engineered product, Tumblr a designed product”.
He did say that-I’m getting mixed feelings. I’m actually a bit shy, still figuring out what I want, and still figuring out what I want to do with it. I’m also hearing really mixed reports about tumblr and tumblr + disqus, but no one is giving me specifics. A lot of the grumbling is simplicity over control and the unsurity of what is going on.as I said, a lot is going on in Tumblr, it’s not really truly blogging per say. And i think it may cause some frustration to some people that internally there are some social factors to this that is turning the idea of blogging on its head. We’ll see what it is as it’s user develop it over time. (I’m not one of those people who say technology is what it states it is.)
“Posterous is an engineered product, Tumblr a designed product”. Great quote.Agreed, David. Tumblr is a great networking tool. The follow feature is a major differentiator. That feature is why Tumblr is not Blogger/Wordpress. There is an intimacy to following on Twitter that does not translate when you put that same person on your blogroll.You are the guy who reintroduced me to Tumblr. I ended up following everyone on your list of entrepreneurs and VCs. And one of those guys I met a few hours later at the FourSquare party. And I am like, wait, I know you. I just started following you on Tumblr a few hours back. http://technbiz.blogspot.co…
Nice!
When Google makes (or made) search 0.05 seconds faster, their usage went up. I think you can similarly measure Karp’s focus on simplicity. Can be quantified. Everything he has done to make Tumblr simple has paid back in spades.
A very simple phrase (or not):Ceci n’est pas une pipeThe treachery of images is you can’t quantify them totally. Unless you can tell me how you can quantify this?
I agree that companies should be as small and lean as possible. However, I think that to judge if the team has the right size we have to look at their finance, so there can’t be absolute truths.You may do an amazing product that everybody loves with a very small team (well done, others can’t with a huge team!) but not be able to monetize yet, so you are losing money. Then, with the same usage levels, you start bringing in sales and the team grows a lot (sales usually scale bad and you have to add people to grow revenue) but the company becomes profitable.Anyway, big congrats to Tumblr. I know today’s post is about celebrating the product and their reduced team of engineers. Just wanted to add something to the discussion.
That’s amazing PVs and a great success story. They should now focus on on-boarding, CMS & IA improvements that can help SEO. I’ve seen more than a few Tumblogs that were missing a golden opportunity for way more traffic because of the way they were set-up. That search traffic could also help revenue generation through better adsense performance or other ad revenue streams.
i agree that they could do better on seocan you help?
Sure. I’m presently helping one of the largest tumblogs (2.5M PV/mo) with this to improve their eCPM.
I’d be interested in that, too…SEO on tumblr
oh there you are…I knew someone was missing.
Busy doing work!!:-)
Congratulations to Tumblr!Now let me easily port over my Posterous blog and you’d have a new user.
Fred, you could make it 1.5B if you had AVC there! Congrats! I love and use the product! (http://blog.3bigheads.com)
i have fredwilson.vc therei like blogging, tumblogging, and twitteringi get value from each of those activities
Fred, have you taken a look at a quantcast/comscore equivalent analysis of the entire USV portfolio? I realize it wouldn’t have a lot of bearing on any individual company, but it would show some interesting data related to the fact that Fred Wilson might have more vested reach into the internet than almost anyone else out there right now. Who knows what other usage and design lessons are embedded therein?Whoever gets that GM position is going to have a lot of interesting work to do.
“Fred Wilson might have more vested reach into the internet than almost anyone else out there right now.”You are skipping the trees for the forest. I think you made a big point here by connecting the dots. Facebook, Twitter+Tumblr, FourSquare, Zynga etc. The roaring 10s, that’s what I am going to call it. I’d love for someone to go ahead and quatify this claim that I have been hinting at for months. Expect the AVC comments sections to bloat over the years. Disqus, you got some scaling issues coming up.”Whoever gets that GM position is going to have a lot of interesting work to do.”I am hoping it is me! 🙂 http://technbiz.blogspot.co…
I think this is semi-simplistic. On one hand, yes, we’ll work and play in a rich mediated enviroment. Google Chrome OS, when/if that happens, is a hint of life as it may be. On the other, the larger unexplored areas of the internet are areas where we don’t count hits, rather niche engagement of humans, and how our objects interact with each other and us seamlessly.
I’m not quite sure I understand your point. I see a lot of internet traffic (my own usage, that of my friends and peers) becoming increasingly focused on a series of central hubs: facebook, twitter, amazon, etc. That makes sense to me since it mimics the history of business where you generally have a monopoly, then a bunch of smaller players, then integration up to a new form of monopoly, then you rinse and repeat. The Oil and Gas, Telecom, and Auto industries are all examples of that trend. Internet “monopolies” are a different beast, but if you look through the right lens, it lines up.By looking at individual sites, you only see their individual interest. When you have an entity that has made as many good decisions as USV has, then looking at the common DNA threaded throughout those organizations can provide really interesting lessons.I think that analyzing and systemizing that kind of knowledge would be an important part of the GM’s job. Here’s to hoping I can convince Fred and Co. of the same.
I can’t say that the idea of screen as we know it now is totally clear. Also, the idea that “Facebook is a monopoly, Google is a monopoly” issomething that while may be true for now in the states, may not be true forlong- this may be one of the first time in a while that we’ve had such rapidchange because of how fast usage has changed.Our idea of box-computing has changed tremendously in the past 60 yearscompared to say the use of diamonds or oil. If I were to start addingnetworking capabilities to my toothbrush, what would happen to the box- andwho said it will remain “box” at all?
we run those numbers but we don’t compare them to anything
Is that a reflection of low perceived value, or of a lack of time?
do they have chartbeat installed?i have to say – i just installed this on our service and the tool is absolutely AMAZING. go look at Fred wilsons dashboard in real-timehttp://chartbeat.com/demo2/
great find thanks Mark
i love that they use my blog as a demo.
Open question to the community. I am in the process of adding a blog to my website, and am unsure which way to go. I really like the ease of Tumblr but just got done reading an article about how WordPress is much better suited for SEO.Any thoughts as to which is the best way to go?
I run WordPress and think it does a pretty good job with SEO (there are a lot of plugins that help with this too). I think it depends more on what kind of blogging you plan on doing. If you’re going to be doing short media filled bursts, tumblr is probably the easier option, but if you want something more robust and customizable, WP might give you more power.
if you want something like avc, go with wordpressif you want something like fredwilson.vc, go with tumblri use typepad, but i would have gone with wordpress if i had to make that decision all over again
A huge fan of small creative teams that deliver amazing results. How to build one of my own….?
I am slowly beginning to ‘get’ Tumblr – it strikes me as being naturally elegant, and very much arts/creative-friendly.What MySpace could have been?Impressive stats, well done.
Congratulations. Be proud.This “in house” financing was making the news yesterday, I read up on it a few different places.How will Tumblr monetize is a stupid question to ask, as some are asking, and not just out of curiosity, but to suggest, so what, let them have all the page views they want, they can’t monetize. How will Tumblr monetize? That is a question I have asked myself, in my case out of sheer curiosity. I have some ideas, I have read some of Karp’s thoughts on this. I think they are going to monetize in original ways.That thing between the PC and the smartphone is not the iPad, but that thing between Twitter and Blogger, thy name is Tumblr.I am thinking of doing a blog post on Karp. I did one on Tumblr weeks back.
Maybe not a very insightful comment, but: congratulations.
often, that might be the most insightful one of all.
Sincere works, too. (IMHO) Sometimes it’s just nice to see another face and to know people are out there. —Hey, you’re in Israel! Cool.
I just discovered the Tumblr Bookmarklet – it’s awesome – think I’m going to start using it to bookmark comments that I like here and elsewhere.They should promote it more in the dashboard
2/3 of my tumblr posts are from the bookmarklet
It’s easy to sign up for Tumblr… But they’ve also made it very easy to start using, too. They don’t just dump you off at the dashboard after you sign up — they guide you through it.
Congrats to Tumblr, especially David. It’s well deserved. I love the smell of Tumblr in the morning. Smells like victory.
I don’t happen to know the Tumblr guys personally but stoked for them. Tumblr is all the things everyone has said in the comments – elegant and simple and a valuable platform for democratizing publishing of content. Though I know a lot of people use their tumblogs for photos/quotes/snippets of the Internet, I’ve been doing more long form posts on my personal blog which uses the Tumblr platform for close to 3 years. I’d used WordPress and Blogger in the past and just found myself having the same reaction as the Gotham Gal – it’s so dead simple.A small, but interesting data point re: getting to the next billion page views – In the last 6 months I can count at least 5 of my “Normal” (non-techie) friends who have set up personal/family update blogs and are using Tumblr as the platform. These are true long-tail, non-artsy or techie users which I would say made up the bulk of the Tumblr crowd early on. Hard to extrapolate this but at least from my vantage point, tumblr is making some progress in the minds of users who no longer see WordPress or Blogger as the only default ways to share their life. And certainly tumblr is on the right side of a more macro-level shift in the general willingness of people to share their lives online.
I use wordpress for my blog because of the plugin abilities and use tumblr because it’s so easy to post media (usually music for me) and follow conversations. I would probably switch everything over to tumblr if there was 1. there was an easy way to do that and 2. if plug-in apps were easier on tumblr. Is this on the product roadmap for Tumblr? Will WordPress always have this inherent advantage because it’s open source?
Fred: Wanted to qualify your comment here. comScore can and does roll up pages that run on other domains–we just need Tumblr to tag with us like they do with Quantcast. We do this for lots of sites, Glam bring one of the more visible examples that is well over 1 billion pages. Would love to do this for Tumblr. Happy to discuss further.
Fred–Wanted to qualify your comment. comScore can and does roll up roll up all the pages that run on other domains–we just need Tumblr to tag with us. We do this for lots of companies, Glam being a good example where they are well over 1 billion pages. They derive a lot of value from this, so can Tumblr. We’ll reach out to them.
Finally broke down and decided to contribute to getting it to 1 trillion page views today.Only thing I think it’s missing is this: when I post something via e-mail while I’m mobile, Tumblr should reply back with a shortened link so I can tweet the post. (I don’t like auto tweeting. I want to write the tweet.)Otherwise, what a great, simple and easy service. Amazingly, my domain redirect worked almost instantly, too: http://t.aaronklein.com
another thought on comment differentiation -I can’t find the quote right now but somewhere on the posterous Q&A site they discuss their philosophy on commenting, and how disqus is unlikely to be integrated since the native site should make comments as simple and contained as possible.I think tumblr has a leg up here having integrated disqus. i do like posterous a lot, but their commenting is weak – can’t even do nested replies. i ‘had to’ ignore a lot of comments on my posterous blog for this reason since replying under a series of comments would have led to chaos.
Steve, it is the Posterous philosophy that commenting should be built into the blogging platform. They should be enabled by default, and deeply integrated into the platform.Because Posterous has native comments, we can do smart commenting by email, and that’s what allows users to create groups on Posterous, that act as email lists.Imagine if Google Groups or Facebook required 3rd party commenting…
Pretty surreal to become a new user on Tumblr this weekend and then after playing with it tonight to come over to A VC and see this post. Your points hit home. Great aha moment. BTW, off topic, listened to FredWilson.FM for the first time the other night while working. A full two hours. Was blown away.Oh yeah, congratulations to USV and the Tumblr team!
love tumblr. always have.they’ve done to blogging what I’m about to do to e-commerce, namely utilise principles of disruptive innovation to attract both non-consumers and overshot customers. Whether intentionally of not they’ve followed Clay Christensen’s teachings to a T.——New figures on Facebook came out yesterday, 300bn page views per month. Was surprised to see such a wide gap between 57th most trafficked site on 1bn and 1st/2nd most trafficked on 300bn
As someone who has blogs registered on every available platform, I can tell you that Tumblr is the best (and the easiest) from the user perspective. That’s ultimately what matters. Users & how they feel about your service.While I have all these blogs registered, I have never been inspired to write something. By the time I am done figuring out all the UI stuff on WordPress, for example, I don’t have the time to actually go on to writing a post thereafter. With Tumblr, I am actually inspired to write, write, write. Taking a small non-scientific survey of those around me, people seem to agree.Clicks, page views, and eye balls aside, it’s just visually inspiring, easy to use, has great templates, and a FANTASTIC wizard. The wizard is what’s genius here. The Team is too 😉
Tumblr could be an awesome service, but I’ve noticed something weird. I’ve seen a lot of blog posts about Tumblr, but in thousands upon thousands of articles, links, and searches read, I’ve never once seen a link to an actual tumblr blog. I’m not sure what this means, but I don’t think it bodes well, it gives me ‘geocities’, ‘livejournal’ etc flashbacks.
Great news for the amazing crew at Tumblr
Thanks for providing good source of information is really useful for webmasters
Yes and No. :)I agree that its speaks to the power of the long tail. And of course the content is also different, and the usage is different when we compare walmart page views to tumblr page views. It’s like comparing the # of conversations we have with other people with the number of times we shop or the number of products we consider each week in a grocery store.At the same time, Walmart doesn’t make any of its products but sells them. And in some sense the pages on walmart.com are an aggregation of the product descriptions created by the product manufacturers. (I’ll caveat that with the fact that Walmart, as other retailers like safeway, are into private label where they get make higher margins)Similarly the type of content on Disney and the engagement with each page of the disney content is different from the tumblr content.Either way i agree that its not apples to apples comparison. As you suggested a real comparison of performance is with respect to wordpress, even with facebook traffic and twitter. Because tumblr is creating a space in between those three. 🙂
Disagreed. Page views are page views are page views. Tumblr is kicking Disney’s butt.
comscore has wordpress at 980 million monthly pageviews
Take the .com out of Disney.com and I am with you.
This talk will be easier if we talk about Walmart and Tumblr. My new statement: Tumblr is kicking Walmart.com’s butt. There you go!