 OK, I've just showed you scenario one. We're going to move on to scenario one A right now. Onscreen, I still have scenario one that showed so we can compare the noise values. So here, if you remember, in my previous example, scenario one, we had assumed that I had about 100 conversions per bucket. So what this means is roughly about 100 users per day have converted for campaign one, geography Europe, product category shoes, and about 100 users have converted for campaign one, geography Europe, product category t-shirts, and so on. What if my campaign was a little bit less impactful? So let's assume that instead, I've got about five conversions per bucket and per day. So I'm going to set my average daily attributable conversion count per bucket to five, and I'm going to go ahead and run my simulation again with these new numbers. Those app automatically scrolls down to the newest simulation. So this is my new simulation here on the bottom. And here, this is my previous one. So you can see here that the APE was one at 5% for purchase value, and previously APE was even smaller for purchase counts. And what happens here is we can see that the average percentage are really increased up to 42% for purchase value, and about 3% for purchase count. You can see a bit of our ability between APE and RMSP, but they're kind of in the same order of magnitude. We'll come to that later. So why are these noise values here higher for that new simulations? Well, that makes sense because here, if you see what we've done is we've made our buckets smaller. So the blue part of each of my entry in my summary report is going to be smaller. And remember, the aggregation service and NoiseLab as a simulation tool is drawing noise from the exact same distribution. It's random noise, but it's drawn from the exact same distribution no matter how big your buckets are, no matter how big your blue bars are. So here, in my second simulation, I end up with more noise relative to the signal. So it makes sense that my noise ratios here are higher for that second simulation.