Not Invented Here: When “We’re Different” Is Just an Excuse to Ignore Outside Work

Beyond UX Design

This episode explores the 'Not Invented Here' cognitive bias, where teams instinctively reject or undervalue ideas and tools simply because they

Key takeaways

• The 'Not An Invented Here' syndrome was identified in a 1982 MIT study by Ralph Katz and Thomas Allen, which found that long-tenured teams eventually stop communicating with outside sources.
• At GE, the rollout of the Cirrus design system failed because local teams prioritized their homegrown components over the centralized system due to origin bias.

Main topics

• Not Invented Here bias
• Cognitive biases in UX design

Notable quotes

The origin of an idea ends up carrying more weight than its quality.

Conclusion

By focusing on the substance of an idea rather than its source, teams

Transcript preview

Speaker 1 (0:00) Somebody outside your team just handed you an idea. Now, maybe it's a component from the platform team, maybe it's a research readout from an agency, or maybe it's a suggestion from an engineer in a design review. It doesn't matter. But somewhere in the back of your head, before you've even really looked at the idea... A little boy says, hmm, yeah, that's just, that's not going to work for us. Not because you found a flaw, you haven't even looked at it yet. It's because of where it came from. And that's called not invented here. It's the tendency to reject or undervalue anything that originated outside of your own team and overvalue the stuff that your team built yourself. It never feels like you're rejecting something for the wrong reason. It always feels like you just have standards or like you know your process or you know your product or you understand your users. It feels just like. Our context is different, and sometimes it is. But a lot of the time, the source of an idea is doing the evaluating for you, and it's doing it in your subconscious. All you're doing is signing off on it. So if your team has ever built something that already existed or quietly shelved the finding because it came from the wrong people, stick around because today we're talking about why your brain treats not ours as not good and what you can finally do about it. Speaker 1 (1:16) What's up, UX fam? How's your mom and them? Welcome to another edition of the Cognition Catalog, where every other week we'll dig into a new cognitive bias and explore how that bias can impact how our team works together and what we can do to ensure we don't get caught up in the pitfalls. Because working with people is hard. and the more insights we have into the human condition make it just a little bit easier. To learn more about Not Invented here, you can find more resources at cognitioncatalog.com. Sign up for the newsletter to get a new bias in your inbox every week, and make sure to subscribe wherever you listen to podcasts so you never have to worry about falling for any of these cognitive biases again. And if you think you're getting something out of the show, then I would really appreciate you leaving a five-star review. That would help me out so much. Shout out to this week's sponsor, Mobbin. If you've been building UIs with AI, then you know the problem. You describe what you want, the agent builds it, and it comes out fine. But it's also pretty generic. The same onboarding, the same card layout that you've seen a hundred times because the model has never actually seen good work. It's just guessing. Now, Mobbin fixes that. Their MCP server connects to over 600,000 shipped UIs straight into Cursor or Cloud, so you can explore real patterns before you build instead of course correcting later. And my whole team uses it, and it saved us a ton of time. Now, visit mobbin.com slash beyond UX for 20 % off Mobbin Pro, and go see what good looks like before you start. And as always, thanks so much to Chris Siriquan, Stacy, Roddy, Megan, Andrew, John, Mark, Kevin, Jason, Michelle, Marty, Tino, Adina, and Zoe for all their support. And for more information on how you can support the show and help more people find out about what we're doing, make sure to check out beyonduxdesign.com slash support. And of course, if you think the shows we're sharing, then for the love of God, tell your mom and them. A few years back when I was at GE, a mandate came down from the top. A central design systems team out in San Ramon in California had spent a long time building one design system for every software team in the company. Now I'm talking not just GE Digital, GE Power, GE Aviation, Oil and Gas, Renewables, the whole lot. And that's no small task. It was called Cirrus because it was light and airy, just like the clouds. And it was actually a pretty good system. It was well documented. It was thoughtfully built. And clearly a lot of work had gone into it. The message from the top was pretty simple. This is the system now. Everybody is going to use it. And at the time, my own team had been building our own thing for years, totally homegrown, and it fit exactly what we needed. And we weren't really unusual. Almost every team at GE had some version of that, their own components, their own patterns, their own way of doing things. So when Cirrus landed, you can pretty much guess what happened. Pretty much everyone said, yeah, no thanks. That's not going to work for us. Pretty much every single team pushed back. And every one of them had some variation of... the same reason it doesn't cover our use case or our users are different or we've already got something that works or we don't have time to switch. We're special. We're a little snowflake.