[BobWarfield]

Talking to people who just bought or just quit is great, but if that's all you do, a lot of key insights will be missed:

- Talking to people who didn't buy, but instead bought a competitor.

- Talking to expert users who've been with your product for a while to learn what they're doing with it you never even considered a possibility. Or to learn what they thought they'd be able to do, but could never quite make work well.

- Talking with all sorts of people to refine your elevator pitch and figure out how to secure the earlier parts of the funnel. Just because someone bought doesn't mean they're the only oracle of why people might buy. They could simply be early adopters after the latest shiny thing.

These are all things I've done with success.

#1, talking to competitor's customers, is what led me to invent spreadsheet notebook tabs on Quattro Pro at a point when most everyone thought the world wanted 3D spreadsheets instead of formula linking across sheets. By talking to folks who had bought Lotus 123 and asking why they wanted a 3D spreadsheet, I quickly discovered it had nothing to do with summing along a z-axis and everything to do with grouping multiple spreadsheets together as a single file.

I've gained all sorts of insights from #2 by talking to power users. This is particularly true of products that have programmability through scripting and API's. But it's also true when you hear about someone doing something amazing you'd never dreamed the product could do, and then you hear about how many flaming hoops they have to jump through to get it done. Yet, solving the problem is so valuable that even making it a little easier makes them thank you. Suddenly, you see how to make it a lot easier and a lot more accessible to a broader audience. Depending on what you've discovered, you might even open a whole new sub-market this way.

#3 just comes from the realization that the more you pitch an idea, the more you learn about how to present it. It is extremely helpful for the people calling the design shots to get to go through the full two way interaction of trying to sell the design. Engineers, especially, quickly learn that prospects aren't going to give them a blackboard and 2 hours to prove that the laws of physics insist they must buy.

[jasonfried]

Dan didn't go into detail about this, but the reason to talk to people who just bought or just left is because they are very close to that specific moment. Decisions are still fresh in their mind. The information is cleaner, less embellished (people tend to embellish when recalling something a long time ago because they can't remember actual specifics).

Someone who's been using your product for years can't tell you why they bought. It's been too long. They may think they remember, but the reasons are often so tied to a specific event - often emotional - that it's too far in the past to remember the specific timeline that lead up to the purchase.

Yes, talking to power users can be helpful for other reasons, but it's not helpful if you're trying to find out why people buy or quit.

It all depends on what kind of information you're looking for. You have to know who to talk to and when to talk to them.

[gvb]

Another thing recent customers can give that is very valuable is useful feedback on the product's usability. They are still discovering where buttons are and how to use the product, so they can tell you "I was looking for X, and could not find it" or "X was in page Y, but I was looking for it in page Z."

Experienced users end up trained (in the Pavlov sense) to do some pretty obscure clicking to accomplish a given job, but they know exactly what to click and where, and don't give it a second thought. I think of this as the "Microsoft Windows Syndrome" - Windows has pieces of it stuck in really unexpected places if you stop and think about it, but everybody "just knows" to right-click on e.g. "Start / Computer" to get to certain Windows features, even though most configuration is accessed through "Start / Control Panel".

The unfortunate irony of "Windows Syndrome" is that, if you move an existing item from an unintuitive, but "everybody's been trained" location to the intuitive location, you will break everybody's mental model of the software and they will scream bloody murder. The Office "Ribbon" is an example: when it first came out, the people that operated via memorized click-sequences were lost and very upset.

[manuelflara]

Also, while I see the appeal of talking to people who almost chose you but ended up chosing a competitor instead, actually finding those people and engaging with them can be way more complicated than do it someone who just signed up to your service.