Your wall reports impressions, clicks, conversions and revenue. Most publishers look at the last one, which is the only one that cannot tell them what to do next.
Read them in order instead. Each number gates the one after it, so the first bad figure in the chain is the one worth fixing.
Impressions: did anyone find it?
An impression is a wall that was opened. If this number is small relative to your active users, nothing else on this page matters — you have a placement problem, not a monetisation problem.
Walls die in menus. The fix is almost always to move the entry point: into the shop, next to the balance, onto the screen where the user just ran out of currency. A user who is short of coins is the only user who is guaranteed to be interested.
Click-through: is the inventory any good for these users?
Of the users who opened the wall, how many clicked an offer?
A low rate means what they saw was not worth trying. Usually one of three things:
- Thin inventory for their country or device. Check what the wall actually shows for your biggest market rather than for you.
- A payout floor set too high. Filtering out everything under a dollar feels like quality control and often just empties the wall.
- Categories switched off. Surveys are the highest-volume category on most walls. Blocking them because they feel low-value removes most of the inventory for a lot of audiences.
Completion rate: the number that matters
Of the users who clicked, how many finished? This is the figure to optimise.
It is the difference between a wall your users trust and one they try once. A high payout with a two per cent completion rate is a page of disappointments — every one of those clicks was a user who tried, failed, and now believes your wall does not pay.
What moves it:
- Clear requirements on the card. "Reach level 5" beats "Play and earn". A user who knows what is being asked either does it or does not click.
- Sensible difficulty at the top of the wall. The highest-paying offer is rarely the best first offer. A wall sorted purely by payout puts a twenty-dollar deposit offer in front of someone who wanted to spend four minutes.
- Removing the offers nobody finishes. If a specific offer collects clicks and no completions over a decent sample, it is costing you the trust of everyone who tried it.
Revenue per active user: the one to compare with
Total revenue tells you about last month's traffic, not about your wall. Revenue per active user per day is what you compare against your other monetisation — video, subscriptions, purchases — because it is the only figure that accounts for how many people participate at all.
Track it weekly. It moves slowly, and every change you make to the wall shows up in it eventually.
The two numbers people misread
Reversals. A reversal rate of zero is not good news, it means nobody is checking. A few per cent is normal and healthy. A sudden spike is worth investigating — usually one offer, one country, or one very busy user.
Average payout. It rises when a handful of users complete big offers and tells you nothing about the median experience. Look at the distribution, not the mean, or a single deposit offer will convince you your wall is doing better than it is.
Where to look first
Walk the chain in order and stop at the first number that is worse than you expected. Fixing a completion rate when nobody is opening the wall is a week spent on the wrong problem.