Conventional startup advice is to 'move fast' and 'pump out features.' Are there any startups that created a great new thing that was growing and then failed because they were improving too slowly?
I am aware of multiple companies that died because they were forced to "grow too fast", rather than growing slow.
One was a search engine ahead of google in its category that would have been an ideal google acquisition target (in fact, google is still doing a poor job in their area of search.)
Another was a company that invented a key gaming technology, but was ahead of the market by a couple years.
In both cases the VCs forced them to make compromises to to the product to chase the current fads... rather than invest in the parts of the product that their customers wanted (and would have paid for.)
Both companies had identified markets that would be extremely fast growing and are today worth many billions of dollars, and in both cases, to date no company has really done what they did.... though the market has shifted to get by without them.
VCs focus on growth because it benefits them. Finding the market and addressing it benefits the company. IF the timing is off by a couple years, then its "too late" for the VCs. This was the case even though it was obvious that there was a massive market coming-- in both of the above examples, the hockey stick had started.
VCs would apparently rather have a $30M valuation in year 2 than a $1B valuation in year 6. The VC funds last 5 or 7 years, if I recall.
10 out of 10 VCs would take the six year $1 billion acquisition over the two year $30 million acquisition in fact. Your numbers are way off. It would have to be closer to $500 million in two years, versus $1 billion in six.
When looking at $30 million today with a probability of 1.0 vs $1 billion in six years with a low and unknowable probability, then taking the $30 million today might make sense for the VC, depending on what their books look like.
I don't know if I'd say Ingres is a failure. They weren't as big as Oracle, or even SQL Server, but they've lasted a really long time, their product is still around [1], and they even spawned Postgres.
On a tangent, it's kinda funny how part of the "geek" community will complain about the big players dominating and ruining competition (Microsoft, Oracle, Apple, even Linux sometimes), but on the other hand anybody who doesn't completely dominate their competition is lambasted for being a failure...
This is an interesting article; thanks for the links.
I last used FogBugz in 2008 and it's still my favorite bug tracker. I have used JIRA extensively and while you can configure it and add addons/plugins to it to make it do your bidding, it's just not a nice off-the-shelf solution without configuration (stuff that I usually don't have the access to do, and that I get tired of asking for).
Funny how this stuff works. I wonder how Fogbugz is doing today. I do understand they must have done well with the Trello stuff.
Yep. That's my personal experience as well. Looked at FogBugz several times, but the pricing always put us off. Eventually Atlassian came around and made a good enough product.
Hashable. One of the startups that I believed in the most just crashed and burned. Right out of the gate they got plenty of feedback. Lots of hate. Some amount of love. Much confusion. But the people who got it, really got it (that's my criteria for great new thing). I still long for a contact management tool like that. When I first started using it, I kept giving feedback to the team to the point that they sent me stickers, a button pin, and a tshirt. Yey for #swag. And I kept giving feedback. And then they went dark. No replies. Sporadic tweets. Updates rarely. It was like they just gave up. And then I got the mail that they were closing down. They stopped because they "couldn't make it good enough"
Side note - That was also the first time I understood the risk of putting content into a closed eco system that's using non open standards for data. <sad face>
MapQuest. They achieved the marketing Holy Grail of reaching the Verb Status ("Let me MapQuest it") only to lose it. It was stale in features while Google pushed forward rapidly.
Either they were moving too slow or announced new product too early (killing demand for the model they were actually selling). They went bankrupt before new model came out...
Apparently not entirely true, but that was the takeaway.
it got a name: the "Osborne Effect"
I'd argue that Osborne moved too quickly, he was announcing the future (and they didn't deliver it too slowly, they just announced it too quickly) and subsequently committed suicide (figuratively, of-course.)
Competition and profitability aren't your only reasons to move fast. VCs may judge you on your speed as well.
Our company bootstrapped for the first 2.5 years before raising an angel round. This raises eyebrows. Some investors assume there is a problem if you are looking for a small round after so much time has passed. It's a rare stance to encounter, but something to be aware of.
(Personally, I'm proud of what we were able to accomplish before taking outside investment.)
That depends entirely on your financial structure, your product, and your own lifespan. If you never on-board investors and never depend on a loss-leader strategy for capturing market, you can small-business yourself from you mid-twenties to retirement and beyond. A lot of trade professions do exactly this sort of slow expansion; my neighbors is self-employed plumber and he's been routing tubs for the past two decades.
Jaiku, Pownce, Gowalla, Bing/MSN search, Posterous, Windows Phone
There are tons of them. You'll see a lot of responses that claim all of these companies didn't die due to moving slowly, but for other reasons - that will also always be true though, even in cases where a startup moved too slow. For example, Windows Phone was both backwards, and slow to match the iPhone once it became obvious consumers loved touch.
In many consumer spaces, it's: 'winner take most.' If your competition gets to critical adoption first, most of the time you're toast. So even if Gowalla was growing slowly, they were too far behind. If your competitor takes 60% of the market, you've previously been growing fast but are now growing slowly, and you're now down to 14% of the market - the outcome is most likely that your future growth will be extremely difficult, and eventually you'll burn vast resources just trying to acquire each point of market share thereafter. Microsoft and Google have both delivered rare, extremely expensive examples of this behavior over a long-term basis (in search for Microsoft and social for Google), funded by their prior dominant products.
I still miss Jaiku. I wonder if they would have survived if either they didn't get acquired by Google, or if Google keep supporting them instead of acquihiring.
AWS launched in 2006, while Google App Engine launched in 2008 and Google Compute Engine launched in 2012. But Amazon has always moved significantly faster than Google in the IaaS market, introducing the most comprehensive suite of infrastructure services and relentlessly undercutting on price. AWS has been consistently innovative while Google has long played catch-up at far too slow a pace.
App engine is largely useless. People wanted virtual machines. Google only just worked that out last year. Useless POS could not even allow TCP connections. No long running processes. Dog slow io. So many limitations. Sure it is a different product so I am not comparing apples to apples but Google said 'you want this' and amazon gave us what we wanted.
And Digg lost for several reasons, but one of them wasn't that it moved too slow. More the opposite, really. They released a buggy version too soon, when the users were already angry and considering leaving.
"Failed" sometimes means bankruptcy, but you should also include those companies that survived but remained tiny, while one of their competitors became huge.
Circa 1995, many companies tried online retailing, but none of them could match the speed of Amazon, and so Amazon slowly won out over all of them.
MySpace lost to Facebook, partly by not matching what became popular features on Facebook, such as "The Wall".
And it's not just startups that fail due to moving slowly. The USA auto companies have been losing market share since 1960, partly due to their slow adoption of new techniques.
Toyota in the 1950s was a super nova of new ideas. They invented kanban and "just in time". Oddly, they got some of their best ideas from an American who could not get an audience in the USA:
"Interestingly, important elements of the Japanese production system were enhanced by the work of W. Edwards Deming, an American who developed a system of statistical quality control for the U.S. war effort during World War II. After the war, he was assigned to General McArthur’s staff overseeing civilian transition of Japanese industry, and introduced his system of quality control throughout Japanese industry, but especially in the automobile sector. Largely ignored in the United States after his return, his methods were only adopted by U.S. manufacturers after 1981 when he was recognized by Japan for his contribution to the country’s economic revival."
MySpace lost to Facebook, partly by not matching what became popular features on Facebook, such as "The Wall"
That is a gross over simplification of why MySpace lost to facebook. One of the main reasons I think MySpace lost was that they never seemed to have a grand vision like Zuck and fb seemed to have developed around '06-07. Not to mention, MySpace was an offshoot of a super shady marketing company, not a technology company. That became more and more apparent as MySpace suffered constant performance and security issues.
Facebook was about real life, myspace was about the internet. Facebook was about statuses, groups, uni life, wall to wall convos and organising your life. It's where I heard about climbing club trips, or socials, or that someone in halls was watching top gear on their tv and the price of entry was 1 slice of pizza ordered.
Then of course there was the user experience. Facebook was plain, simple, uniform and quick. Myspace...
-Click on myspace page
-Wait 10 secs for background to load
-Really loud obnoxious music is playing
-Animated page background starts going
-Quizzes load
-A wild photograph appears
-You can see an embedded youtube now - hit pause
-It wasn't the goddamned youtube making the noise, now you've got two things running you idiot
-Try and pause the youtube again
-page freezes
-audio player loads
-manage to stop everything playing
-have epileptic fit from background
-forget why you came here anyway
Crossing the Chasm (Moore) suggests different phases in adoption - when entering the mainstream (after enthusiasts's love of the tech itself, and visionary's grabbing the tech's benefits ahead of everyone else) pragmatists wait until they see who will win. Therefore, in this phase, speed of acquiring marketshare is everything - not tech, not benefits.
Moore generalized this theory from actual startups, so he has lots of examples, but from decades ago (like apple and oracle when they were new). I'd think it would apply to today, just faster - though no one ever seems to cite it.
eg IBM lost RDB to Oracle, even though they invented it. Perhaps all the competitor's to pg's viaweb, who lost to it (note: they employed an expensive PR firm; it wasn't just lisp tech).
There are lots of examples of startups that failed due to moving too slowly. The problem is that you probably won't know about them because they often don't get multiple rounds of financing (and hence don't make the news). Uncountable numbers of startups fail all the time in relative obscurity.
Your question is hard to answer, though, because it depends on where your funding is coming from. Part of your pitch for your business probably included an exit plan (if it didn't, then you should probably figure it out now). You need to get your business to wherever it needs to be to deliver that exit plan. Now you work backwards. What conditions need to be met to make that exit work? For a business that is built mostly on technology, what do you need to have in place to meet those conditions. Then you need to allow time to iterate your development because I can pretty much guarantee you that whatever you think you need now is almost certainly wrong. So you need to deliver early enough to allow yourself time to figure out what is wrong about it and what you need to do to make it right.
I've been in this industry long enough (and been with enough startups) to say that most founders don't understand what they are doing to the extent that they can make a coherent plan. Therefore they simply push to move as fast as possible and "pump out features". This is not necessarily a bad strategy if you are good at reacting to failure and adjusting your plan accordingly. The faster you fail, the more time you will have to discover what will succeed.
This strategy will only work for so long, though, and you need to be able to change your strategy at the correct time. This is very difficult and my experience tells me that most people get it very, very wrong. Sometimes even if the founder can do it, the people they hire to search the solution space (by banging out features one after another) are incapable of building a coherent product. To off-set this problem some startups adopt a demo-only strategy. Their goal is to build a disruptive demo and get bought out by one of the big boys (and probably buried). They have no intention of actually building a viable product. This can actually work well, depending on the prevailing economic environment and how scary you can make your demo.
If you want to build a sustainable business, though, my suggestions is to avoid the "pump out features" track (and the developers who are good at only doing that). Instead build more slowly with very strategic experiments and a very flexible code base (along with programmers who have experience building real products this way). If you can find the programmers who are able to do it, then you should have a higher rate of success. The "if you can find the programmers who are able to do it" part may be limiting, though. You may be forced to go the "bang out features and react to failures" route.
This is something I'm seeing too; having a committee design a product has resulted in feature after feature being put together in an incoherent way. Pump out features can't work for long.
Could not agree more on Lyft. In a winner-take-all market where an all-out land grab is critical to the long-term success of the marketplace, moving quickly and being the first to reach that golden scale of monopolistic ownership of supply and demand is absolutely key.
Not a perfect comparison given Lyft was founded 2 years after Uber and therefore was already at a disadvantage, but had they grown faster, they might have been able to overtake Uber's position.
In any case, the outcome of the Uber / Didi battle in China should be a great data point for this question.
I don't think Lyft has failed by any means, it's a 4 billion dollar company and growing. Also, it seems they're quite quick to innovate. IIRC, they were first to come up with the carpool system that Uber then rapidly appropriated into UberPOOL. It's just hard to overtake a company executing as brilliantly as Uber.
> I can't speak for the other applications, but have you used Hipchat? It's really poor in comparison to Slack.
You're splitting hairs. They are both chat apps for crying out loud.
Slack grew as fast as it did because it got pimped to us by VCs and thought leaders and then wannabe thought leaders and down all the way to the bottom rung of the societal ladder. Great strategy by Slack but there is nothing special about it (the product).
I wouldn't be so sure. Slack was just a better product overall, offering a clear interafce and much better integrations than Hipchat. Hipchat being slow did not impact them as much as having a mediocre product to begin with.
About Yammer, I wouldn't really put it in the same bucket as Slack, as it really tries to be a Facebook for your company, not a group chat.
There is an existing baseline before Slack was released that you can take as a rough baseline.
It's true though - Hipchat was caught napping. Atlassian backburner'd it at exactly the wrong time, and by the time they starting fixing things, it was too late.
This being said, Atlassian isn't really a 'startup' anymore. Hipchat is a side-product of a suite of gorilla products.
Wife pointed out one. Hotmail. The glory days of internet. Sat on its ass. It's actually great now (outlook.com) and I prefer it over gmail now (bring out the flame throwers). But it took a long long time to get there.
Yes and you could almost say Myspace moved too fast, chased after local maxima (rapid A/B testing) and pursued vanity metrics (registered users). In contrast, Facebook stepped back enough to have a real vision, and carefully chose the right metric (active users). DAU and MAU are industry standards today, but back then it was a bold and careful decision by Facebook.
For investors, time value of money is more important than lowering the risk profile of the portfolio. Also growth is easy to sell to other investors including retail on IPO. And it sexes up the whole portfolio with brand recognition. So get big fast is the accepted 'wisdom'.
Really depends if your building a growth engine vs job creation machine vs small cash generator - that will dictate your approach to growing and funding and of course exit, or not.
If the growth rate is slower (or faster) than the optimal growth rate then the management has made a mistake (assuming they have any control over this). When you frame it this way what you are asking is have any startups failed for making mistakes which is does not need to be answered.
What you are really asking is it possible to workout in real-time (no hindsight) what the optimal growth rate should be for each company and if this is always “as fast as possible”. This is very hard question to answer. My feeling is the optimal is more often than not close to the “fast as possible” level, but it will depend on the company and industry.
How can there be any “real-life” examples of failure for this reason unless you first determine what the optimal growth rate of the company should have been? If you can’t answer this question then any discussion about the reason for failure is a “just-so” story made up after the fact.
The more important question to answer is how can you work out what the optimal growth rate should be of any startup.
One was a search engine ahead of google in its category that would have been an ideal google acquisition target (in fact, google is still doing a poor job in their area of search.)
Another was a company that invented a key gaming technology, but was ahead of the market by a couple years.
In both cases the VCs forced them to make compromises to to the product to chase the current fads... rather than invest in the parts of the product that their customers wanted (and would have paid for.)
Both companies had identified markets that would be extremely fast growing and are today worth many billions of dollars, and in both cases, to date no company has really done what they did.... though the market has shifted to get by without them.
VCs focus on growth because it benefits them. Finding the market and addressing it benefits the company. IF the timing is off by a couple years, then its "too late" for the VCs. This was the case even though it was obvious that there was a massive market coming-- in both of the above examples, the hockey stick had started.
VCs would apparently rather have a $30M valuation in year 2 than a $1B valuation in year 6. The VC funds last 5 or 7 years, if I recall.