It had been a week since we had announced to friends and family our latest idea, LinkFalcon, and only two of them had bothered to try it.
I thought LinkFalcon had some real potential. It solved a real problem for me and one that I hoped others had.
Complete disaster. Failure. Six months down the drain. Back to our real jobs.
That’s what should have happened; but, thanks to the Lean Startup movement, that wasn’t the case at all. Here’s why.
I had the idea for LinkFalcon while watching a very terrible version of the highlights of an Arsenal soccer match on YouTube.
I knew there had to be a better version of these highlights but didn’t really want to go searching for them. And, that’s when I had an epiphany. What if there was a browser extension that would detect the highlights I was about to watch and offer to redirect me to a better version of the highlights?
It didn’t just have to be for soccer highlights, it could be for music videos, movie trailers, speeches, anything! The internet needed a better video redirector!
What’s Our First Version?
Certainly, we needed to build a system that, if given a url of a video, it would return a url with a higher quality version of that video.
We could build a very sophisticated algorithm that would crawl videos and figure out better version of them. The only problem: I had absolutely no idea how to build that.
Alternatively, we could build a crowd-sourced system where users would recommend better version of videos. Perhaps we could start with a small niche like movie trailers and slowly expand to other verticals. This seemed reasonable though it would be a ton of hard-work building that community and the tools they would need.
But, there was a third option: what if we didn’t build the system?
What was the first assumption we need to prove to ourselves? Was it that we could build the system? Was it that we could figure out a way to monetize it? While those were issues down the line, none of those were the first challenge we had to overcome.
The first assumption we needed to prove was that people actually wanted to use LinkFalcon.
Providing this service was not going to be easy so we needed to make sure people really wanted it.
We needed to see that people would actually go through the trouble of grabbing our bookmarklet and using it to get higher quality versions of videos they were watching online.
We Launch Our Experiment
So, we launched a simple landing page with a bookmarklet saying it would return a better version of the video they were watching.
But, when the user submitted a url, we didn’t actually have a better version ready.
The system would just email us the url and the user who submitted it. We would then frantically search for a better version and email it back to them.
Why did we do this? We wanted to find out as quickly as possible whether people actually wanted to use the product. If they really wanted the product and submitted hundreds of URLs, we would be willing to spend the months building the backend that would actually deliver the product.
By not building the system, we were able to test our key hypothesis in one day of coding and not six months of hopeful coding.
We weren’t launching a company, we were conducting a simple but very important experiment.
The Results of the Experiment
After one week and just two submitted URLs, we knew our hypothesis had been wrong. People didn’t really need this as bad as I had thought. It just wasn’t worth continuing to work on the idea.
But, that was okay. We had many other ideas to work on. And, because we tested this idea in just a week, we could actually get to those ideas. (One of them was Yipit, a three-day experiment, that turned into a VC-backed startup. (We’re hiring!)
An Important Other Benefit
Almost every entrepreneur has heard the advice to get user feedback as soon as possible. But, many don’t for many reasons. One of the biggest reasons for me was fear of failure. What were my friends and family going to think?
But, by thinking of it as a quick experiment, that fear tends to go away. The beautiful thing about experiments is that disproving your hypothesis isn’t thought of as a failure. It’s thought of as progress. And, getting early user feedback, even negative, is definitely progress.