GTM Sauce

Ladder · B2C · strength training app

Run a big yearly user survey and use AI to read the answers

7,500+ responses in a year; led directly to the nutrition feature

Workedfeature-development

What they did

Three feedback buckets: (1) acute pain via in-app feedback forms that flow into Slack; (2) ad hoc plus recurring surveys, including an end-of-year survey of 80-90 multiple-choice and open-ended questions taking 45+ minutes, parsed with AI; (3) dogfooding by staff (encouraged, not required). Gives little weight to 5-10 person interviews in a diverse consumer base; takes care to write unbiased, non-leading questions.

What happened

7,500+ responses to the annual survey; clear signal for nutrition.

In their words

The Art of Driving Retention Through Product – Ben Gammon, Ladder

Play the episode

David BarnardWell, the last thing I wanted to touch on was user feedback. And I know that's something that Ladder takes very seriously. So, how do you think about feedback and integrating that into the broader product roadmap and decision making?

Ben Gammon… the first thing is feedback's a gift. And I know a lot of people say that, but it really is, it makes our jobs a lot easier. We get to just really listen to people, what do they actually want? And the hard part is actually staying disciplined and following through with that. There's effectively three ways that I've bucketed how we think about feedback. There's one, there's like the acute pain, and that usually comes through our Slack. There's feedback mechanisms throughout the app that they can just write out like, "Hey, I experienced this issue." It's typically built around that. The second is we do a lot of surveys. There's ad hoc, but there's recurring surveys too. And as an example, we do one at the end of the year and we have an insanely productive community.

Ben GammonThere's 7,500 plus responses this past year. It takes over 45 minutes. Thank God for AI to be able to parse it out and understand what they want. And then the third is dog fooding and it's really important. It's not a requirement if you work at Ladder. But I think in order to really test drive the product, but have that end-to-end mental model of what's happening, it really helps. And also building it out and delivering that empathy for the user because you're experiencing it too.

David BarnardDo you do individual user interviews and how much weight do you put on those? This

Ben GammonIs a spicy take. I think they can be helpful, but I think the problem is, especially in consumer, you have so many people that are very diverse, they're all throughout the world. And if you talk to five or 10 people, they might weigh in on a part that gives you a little bit of insight, but the problem is you can't over-index to that because they still only represent N of five or N of 10. So, these surveys are really effective because we do them so often. There's a huge population set. And I think that's a really, to be honest, just a much better way to kind of work backwards from that large amount of people, as long as you can construct those surveys and ask the questions in the right manner that are not too biased or leading them towards an answer.

David BarnardYeah, that's so powerful. And I mean, again, y'all have such an engaged community. It's probably going to be hard for the average app to get that many responses, especially to that extensive of a survey. But again, finding ways to get that larger breadth of feedback versus kind of over-relying on any individual feedback. And it's so hard, you get one bad review in the app store that harps on one missing product feature and you're like, "Oh, I got …

Get tactics like this every Monday

New tactics from the week's founder interviews, each linked to where it was said.

More from this episode

Tags: user-surveys, user-research, dogfooding