Yesterday, Joel announced Trello. It’s a project management app.
Watching the demo video, my first thought was: “Well that’s not very minimal.” But upon reflection, I’m going to completely reverse that opinion.
A lot of startups have failed to make a splash in this space. I see a serious attempt about monthly. Compared to those attempts, Trello is higly polished and feature-rich. So is Joel just rolling the dice with a foolish upfront investment in frivolous features and pointless polish?
MVP means you avoid building stuff you don’t need. But this is one of those cases where you need an awful lot (including fluffy stuff like nice PR videos). Product management tools have a couple qualities which make “minimum viable” much larger than one might initially hope.
First, the value proposition takes a long time to pop, by which I mean there’s an extended period when the user has to trust you before they actually feel the value.
Second, they depend on total buy-in. The customer’s entire team has to use it faithfully for a long time. They can’t let their team members choose individual tools and they can’t reasonably trial multiple options simultaneously.
Third, longevity is critical. Moving project data is a nightmare, and nobody wants to risk a months-long workflow switch on a company or product which is still clearly an experiment.
The result of these three qualities is that you need to pitch hard and look like you’re 100% committed to a product which is fully formed from day 1. Something like a product management tool can’t look like you’re testing some hypotheses. It needs to feel permanent and stable.
Ultimately, the first version of a new product management tool demands an awful lot of its new users. The MVP demands trust, commitment, and delayed gratification. To match those demands, you (the tool creator) may need to go ahead and put all your eggs in one basket.
 More like a value popositionâ„¢. Guess it’s time to spearhead a trivially differentiated startup philosophy and make my millions. San Francisco: here I come.
 Pivotaltracker is a counterexample to everything I just said. But looking to the successes suffers from selection bias and ignores the fifty similar companies in the deadpool.