3 ms·
My attempt at a story was taking a stance in the ongoing education reform controversy. It earned me the chance to write the post for GOOD, and got the one blogg
by studiofellow 15y ago
My attempt at a story was taking a stance in the ongoing education reform controversy. It earned me the chance to write the post for GOOD, and got the one blogger to cover my app. That one guest post got me all 10 of my paying users.
Story is huge, but I couldn't make it over the hump. Maybe my story wasn't compelling enough.
Thanks for the kind words, the advice, and the offer. Will send you an email shortly.
- 9oliYQjP 15y agoJust curious. I read Running Lean and always had problems with its advice. I just recently got my hands on Lean Startup and think the advice is a lot better. Running Lean is part of the problem. People can't tell you what they want, you can only observe what they actually do. You should have pivoted Knack so that its customers were parents who wanted to track the progress of their child against other children. That's what parents just love to do. Parents should have been begging their teachers to use the tool, then it would have been adopted and you probably could have charged a lot more for it.
- studiofellow 15y agoRight, but the problem is what do you do if you have no one to observe? That's where Running Lean makes sense to me. Start with highly structured interview scripts and hypotheses, and piece together data by limiting the possible set of answers. A pivot for consumers sounds like a great idea, but we all know consumers are more fickle than professional audiences. Turns out teachers behave a lot like consumers, which I should have figured out much much sooner.
- 9oliYQjP 15y agoSo, I'll just repeat what The Lean Startup would say (or at least my interpretation): a) You still have your hypothesis, but you can't validate them until you have actual data. People's opinions about how they might act to some situation in the future is not data. People actually using your product is though. b) Smoke test demand for the product in the first place. You could have created a sign-up page in a weekend and spent a few weeks marketing that if you wanted to feel comfortable with potential demand. These guys did just that and had thousands of sign-ups before they even begun programming their product (http://vizualize.me/ http://vizualize.me/). Go on Facebook right now and share a link to a webpage that you can track. When you share out the link use the headline "National grade statistics released. Is your child a genius or falling behind the curve?". Tell me how many people clicked on that link. That's your smoke test. My gut tells me parents are not exactly your typical consumers. Parents emigrate to new countries where they have to work menial jobs but so that their kids can attend better public schools than what would be available to them at home. c) Create a "concierge" MVP for your first few customers. It's an MVP version of your software that the first few customers use in which you can assess the usefulness of features before you program them. You could have set your wife up with Google Docs templates and watched how she used a few templates over the course of a couple of months of teaching. You could have scaled this out to a pilot program of a few customers, and setting up the templates could have been a weekend job. Only after getting feedback about likes/dislikes with the Google Docs templates solution would you have started to build your software.