Posted on November 28, 2009
Filed Under Business |
Update: Clarification added at the end of the post
Version 1 of Backupify was built with less than $10,000. A friend of mine had suggested that an automated Flickr backup tool might be useful, and a few others agreed. But rather than raise a bunch of money and waste it on something that we was unproven, we chose to launch with the smallest possible feature set and just see what happened. The results were awesome.
In 6 weeks we signed up over 200 paying customers, were featured on several popular blogs, and had a ton of great feedback and ideas from our users. It worked just like Steve Blank said it would. But as time has progressed and we have continued to sign up users at a faster rate than we anticipated, we have struggled to keep up. Much of the struggle is because of the way the customer development process works, and the dark side that no one talks about.
Customer Development and the Lean Startup offshoots of it are the latest tech startup fads. The ideas have some kernels of truth, but they are taken as gospel by web entrepreneurs instead of taken as tools in the toolkit that are appropriate in some instances and not appropriate in others. Since no one ever talks about the down side of the customer development approach, I thought someone should write about it. And since I haven’t received hate mail in awhile, (and honestly, I kind of miss it) I figured that person should be me.
Let me start by saying that capital chases good opportunities. Good ideas can employ lots of capital in a productive fashion, which is why good ideas attract a lot of money. Capital is required to grow. I’m not aware of any $100 million companies built with a lean startup mindset. Those companies are usually built with a “go big or go home” mindset instead. If Google had been a lean startup they would probably be some minor search player right now instead of one of the top technology companies in the world. So if you can’t attract a lot of capital to your idea, one possible reason is that your idea sucks. Granted, there are other reasons (location, lack of connections, etc) that come into play, but if you can’t attract a decent amount of capital to your idea, it could be because the ideas isn’t that great.
If you do go down the customer development path, the main problem, as I have experienced it, is that it puts you in a constant short-term thinking mode. Startups are always a bit more short-term focused than other companies, but product development always needs to have a balance between short-term and long-term. We punted on tons of key issues because we wanted to wait until we proved there was a market for our product. Once we realized there was much more of a market than we thought, we suddenly had to allocate a bunch of resources to customer support and bug fixes… things that we wouldn’t have needed at the same level if we had built a more robust and full featured product out of the gate. As a result, our product development slowed to a crawl at a time when we needed to double our pace to catch up with user demand.
Think of it this way… some ideas are just bigger ideas. Charlie wrote that Path101 was a $1 million product that only raised $350K, and that was a problem. We are finding out that Backupify is the same way. It isn’t a cheap an easy thing to build.
Imagine you have an idea for a new type of hotel. Would you build just two rooms and see if people liked the general idea then, if they did build 15 rooms and if those sold build 40 rooms and then eventually 100? If so, you would end up with a building whose core infrastructure was not designed to make it efficient to service and run a 100 room hotel. That is what can happen with a lean startup. You could argue that once you prove a market then you can always go back and rebuild the thing from scratch when you have more resources, but that too isn’t really true.
Investors always ask how much money you have raised and at what valuations, so when you do small raises on low valuations because of the lean startup process, you get stuck in that mindset. Each time you try to break out of it and go for a lot of money at a larger valuation, that round of investors feel like they are getting screwed compared to the “good deal” the earlier investors received.
The result of all this is that if you are on to something big, you risk proving the market so that someone else can come in and take it from you while you struggle with a small product that should have been bigger. Or you shoot yourself in the foot on valuation. Or you struggle for months to catch up on product development and build what you should have built in the first place.
My advice is to treat these fads like all other fads in business -suck out the kernels of truth and throw the rest away. The broader idea behind customer development is that you should get close to customers early. That should be your goal whether you raise $100 or $100 million. The broader idea behind a lean startup is that you should be capital efficient. And you should, regardless of the size, scope and age of your business.
Don’t be afraid to go big and fail. Lean startups can be attractive because they minimize your chances of failure. They allow you to string along a mediocre business for years while keeping your day job and justifying the whole thing by saying you are lean. But if you are that afraid of failure, you probably shouldn’t be working on a startup anyway. Shooting for a big idea and failing still means you will learn a lot (like how to deal with VCs and how to manage momentum), meet lots of interesting people, and in the end you could end up better off than you would if you had boostrapped a company for years only to go nowhere.
So take my experience and advice for what you paid for it ($0). I’m sure those of you who have never started a company but read startup blogs all day long will feel compelled to email me and tell me how stupid I am. My email address is idontcare-at-coconutheadsets-dot-com. If you do have good counterpoints, questions, or ideas, please leave a comment. They all get published, whether you agree with me or not.
UPDATE: Here is a better way to describe one of the problems we had. Our first 500 customers gave us VERY different feedback than the ones we got beyond that point. Five hundred customers seemed like a good number, but the truth is, they were a skewed sample. Most of our first few hundred customers were hard core tech guys who wanted to use their own S3 accounts and wanted features that later customers didn’t care about. Of course, we listened intently to those first customers. Then we realized our product could have the mass appeal that we thought it may have, but we had made architecture decisions that made some of the mass appeal features more difficult to implement.
UPDATE 2: The other thing I think is misleading is this insistence on surveying all the early users. We did it, and learned a bunch of stuff we already knew. What we really need to know is why the people that don’t sign up don’t sign up. Do the 80% of people coming to the site and not signing up think that we aren’t trustworthy, or do they not understand what we do? Do they think the product is overpriced? Do they want features we don’t have? Surveying my existing user base all the time will never tell me that.