Event Timeline
I understand it's early but given the early numbers you shared, ~3.5k devices per group per day, we could start seeing a difference between the groups.
Also, it's good to have confirmation that the data is usable for its intended purposes. There's still time to tweak it if needed.
Ah, I see. I definitely understand the desire to learn the outcome as soon as possible, but I don't think it's a good idea. We planned the experiment length ahead of time so that we wouldn't form a conclusion without sufficient data. Looking at the data every day would defeat that purpose.
Also, it's good to have confirmation that the data is usable for its intended purposes. There's still time to tweak it if needed.
Hmm...I don't think this would give us any assurance that we didn't already get from my first query. If I was doing a deeper data check now, there would be several other things I would check first.