If you make a small change and expect a big payoff, I think that's very difficult. But I've done a few things as a designer that I've followed up on very clearly and found great results. Balancing quick wins with long-term impact is a challenge we all have across the industry. There's nothing kind of particularly unique with that. One of the things I found in the AI and agentic world is that it's much harder now to build a longer-term roadmap for a product than it used to be. Particularly with the technology changing so rapidly, the features that are coming online, or even the shifts in sentiment in the broader population. I find it's more helpful to look at the next three, six, 12 months. Then you can actually fit in some quick wins in there. want to make sure again that you're driving towards that same durable vision that you've got for the product or the same set of user problems you're trying to achieve. Just to make sure you're not getting distracted by any of these quick wins. A quick win should never feel to the user like a quick win. A quick win should feel like value in the product in the moment. Something that I think a lot of teams get wrong about user research is they want to have a clear, definitive answer. That's why sometimes teams like data, quantitative data over qualitative research. And I think that's a mistake. Design is messy. You're never going to know with 100% certainty that you're moving in the right direction. And that's why we have to be continuously checking in with the people that we're designing for. Continuously getting user feedback, whether that's through our feedback collection mechanisms or our research projects, so that we can understand, are we moving in the right direction? And if we find out that we're not, we can course correct and make sure that we're getting the results that we want to achieve. So at GitHub, we try to move fast and protect quality through a few different clear mechanisms, but also culture. We try to dog food a lot. GitHub is built by a company that uses the product to build the product, which means everyone at the company uses the product on a daily basis. We ship things to staff a lot to try our ideas. And that's a way for us to minimize friction and get early insight constantly in the process. Then we of course do that with customers too, or like super fans. But to a large degree, we move fast with dog fooding. And then we put a lot of effort into making sure that our pipeline to deploy something is as smooth as possible, and we can roll things out quickly. But culture-wise, it's, it's, we care about the product, and we build a product with a product. I do a couple of different things to keep iteration quick at Microsoft. I know that's, that's a very important thing across the industry in general. One of the things I tell my teams is to operate as if we're a well-funded startup. We don't have to worry about paying ourselves, but we do have to worry about getting work, about getting it out, about being quick, about being intuitive. It might be a bit of a misconception about Microsoft, you know, being a slow-moving company. It's actually quite the opposite. I like my teams to fail fast, to bring ideas, and just to keep things moving in line with our squads. So if you make a small change and expect a big payoff, it's, I think that's very difficult. But I've done a few things as a designer that I've followed up on very clearly and found great results. One of them was not me maybe as a person, but rather Spotify as a company. And so Spotify is incredibly good at measuring and quantifying the decisions they are doing. And even if that can slow down some type of iteration mechanism, you learn incredibly well. You get good feedback. And one thing that the company did was they had a tab bar on the iOS and Android apps, and it had three icons. If you go back to the very, very first iteration, it had three icons. And slowly that grew and it became more tabs. And then at some point everything was moved into hamburger menu with a lot of things you could pick and choose from. And then at one point, we and they started running tests. Can we reduce the number of items and see if that actually affects retention or usage? And it did. And they found a clear pattern where essentially the more they removed, the more usage they saw. So today, if you look at the app, they went full circle. It's back to almost the original design that the app launched with. And that insight, I think, was driven through like a tremendous will to go through a lot of friction to simplify without introducing regressions and create clarity amongst the users. But it looks funny because it's literally almost the first iteration going back to it. But it made a huge impact.