Finding the loss
The campaign is break-even. Something inside it is not.
Averages hide the answer. A campaign sitting at zero is almost always a handful of placements making money and a handful burning it, and the only question that matters is which is which.
Native, push and pop networks sell you thousands of placements at once — zones, widgets, sites, apps. The network has no reason to tell you which of them is worthless to you, and the campaign-level number averages them into one meaningless figure.
The fix is grouping, not more traffic. Take the campaign, group by the sub parameter carrying the zone, and sort by profit. The shape is nearly always the same: a few rows making most of the money, a long middle doing nothing, and a few rows losing steadily.
The trap is cutting too early. Twenty clicks and no sale is not evidence; on a five per cent conversion rate that happens by chance more often than people expect. The useful rule is to wait until a placement has spent about twice your payout with nothing to show, which is roughly when bad luck stops being the likely explanation.
Then go one level deeper. A zone that looks bad overall is sometimes fine on mobile and hopeless on desktop, or fine in one country. Cutting the whole zone throws away the half that worked.
How it works
The mechanism, not the marketing.
- 01
Group by the zone parameter
Usually sub1 or sub2, depending on the network’s macro. Sort by profit, not by clicks.
- 02
Apply a floor
Ignore anything under about twice your payout in spend. Below that you are reading noise.
- 03
Go one level deeper
Inside the bad zone, group by device, country or connection type before cutting the whole thing.
- 04
Block and watch
Add it to the deny list or block it in the network, then check tomorrow that spend moved rather than vanished.
What to check tonight
You can do all of this without us.
None of it needs an account anywhere. If it turns out your tracking is fine, you have lost twenty minutes and learned something.
Sort yesterday by profit, ascending
Not by clicks and not by spend. The row at the top of that list is the one to look at first, every time.
Check the zone parameter is arriving
If your zone column is empty or full of raw macro text, the macro is wrong in the campaign URL and no report will save you until it is fixed.
Look at the lander click rate for the worst rows
Spend with no lander clicks at all is usually automation. Lander clicks but no sales is a landing page or offer mismatch. They need different fixes.
Check the hours before you cut
A zone can be profitable for sixteen hours and ruinous for eight. Day-parting a zone is often better than losing it.
How UTMCAP does it
What you would get here.
Five levels of grouping
Zone inside country inside campaign, and two more if you need them, in one report.
Forty fields to group by
Including all ten sub parameters, connection type, mobile operator, device brand and hour of day.
"Too early to tell" marked
Rows under twenty clicks are labelled rather than presented as findings.
Needs attention, every morning
The dashboard names what is losing money across campaigns, landers, offers and sources, with the reason.
The Copilot does the arithmetic
It flags a zone that has spent twice your payout with no conversion, and says how likely bad luck is.
Block it in the network
On paid plans, blocking a zone in PropellerAds, Taboola, MGID and seven others is one click, with an undo.
Questions
The ones people ask next.
Read next
- How many clicks before I can judge a placement?
- There is no single number, but spend is a better guide than clicks: once a placement has spent roughly twice your payout with no conversion, bad luck is an unlikely explanation.
- Should I block the zone or lower the bid?
- Lower the bid when it converts but not profitably, and block it when it does not convert at all. Blocking something that works at a lower price is how campaigns slowly shrink.
- What if the network does not pass a zone parameter?
- Then you can only optimise at campaign level with that network, which is worth knowing before you spend much. Our platform pages list the macros each one sends.
Point a domain at us and run a click through it.
The free plan takes a hundred thousand clicks a month and no card. If it does not do what you need, you have lost twenty minutes.