...
For example,
If preezie is embedded on a single page then this is straight forward:
Split traffic | Page trigger | Segment | Conversion rate | Sessions per user | Bounce rate |
|---|---|---|---|---|---|
50% Test | myhomepage.com | 40% non-preezie users | 3.2% | 2.4 | 26% |
10% preezie users | 4.5% | 3.2 | 0% (preezie counts as a significant event) | ||
50% Control | 50% non-preezie users | 3.1% | 2.6 | 25% |
This means the buckets aren’t an even 50/50 split but more like a 50/10 split, we’re using traffic split to understand the control vs 40% bucket of non-preezie users.
Once you can see these buckets are performing equally then you can start to compare the preezie 10% against the 50% who never saw it and the 40% who did see it but didn’t engage. This will tell you the user level impact of preezie across sessions, e.g. sessions per user increases.
| Info |
|---|
We do this because preezie won’t be a tool required by 100% of users, we want to help users who need it and not impact those who don’t. |