Could you elaborate a bit more on that? I have a few questions about optimizations, as many other software companies have identified it as a significant headwind. However, for MongoDB, it wasn't as significant a headwind in 2023 when, for instance, Snowflake was significantly impacted by it, along with AWS and others. How easy is it to optimize MongoDB? What were customers doing to reduce their MongoDB spending?

Now, to your point about optimizing MongoDB, it really depends on a number of factors. I've worked in accounts where it could be as easy as turning off certain clusters that are not being used, or turning off development environments, or staging environments that were just on without much usage. Right. So just basic low-hanging fruit that people were not paying too much attention to in prior years, they are now being a lot more careful about things like that. So, there's that layer of low-hanging fruit that you can just cut really easily, and then there are those areas where things could get a little bit more complicated, right? For instance, choosing other types of technologies or layers of technologies that could prevent higher consumption of MongoDB. Adding a caching layer that's not on MongoDB would prevent more utilization at the database layer. That's cheaper than actually buying more MongoDB Atlas or scaling a cluster to multiple nodes. Say you are on 10 nodes, and if you are anticipating high traffic, you might need to add another five nodes. Instead of doing that, you put a caching layer that's cheaper and will attend to the performance that the end users are expecting, and that obviously impacts the MongoDB revenue. So, I think naturally, companies are looking to do things like that. Some of the largest consumers of MongoDB in the West were actually on active plans to reduce their consumption of Atlas when I was there back in '22.