Headroom, as a number
How much more you can spend before marginal return crosses break-even. Not a feeling, and not a percentage someone picked.
Scale Meter
Average ROAS cannot answer the only question that matters.
“Can I safely spend more?”
A campaign at ROAS 3.0 tells you the average of every euro you have already spent. It says nothing about the next one – and returns fall as spend rises, because you exhaust the cheapest, most interested audience first. Scale Meter measures what the last increment actually earned and compares it against the break-even derived from your own contribution margin.
Open Scale Meter in the demo →Sample data · no signup
What you actually see
Revenue against spend, day by day
The two curves on one axis, so the point where extra spend stops moving revenue is visible rather than inferred.
Your break-even line
Drawn from your contribution margin, not an industry average. At 62% margin it sits at 1.61; at 30% it sits at 3.33.
Saturation before it hurts
Marginal return bends down long before the account looks like it is in trouble. That bend is the signal.
How it gets there
- 1
Measure the increment, not the total
Change in profit divided by change in spend between two levels – the only figure that describes the next euro rather than the last thousand.
- 2
Compare against your own floor
Break-even is one divided by your contribution margin. Every verdict is measured against your number, not a benchmark.
- 3
Wait for the cycle
Budget changes reset delivery optimisation, so the first days after a change are systematically worse than the steady state. We read past the reset.
Where it stops
What Scale Meter will not tell you
- Headroom is a model, not a promise. It describes what recent increments earned, and the audience can change underneath it.
- It cannot separate a saturating channel from a tiring creative. Both look like falling marginal return, and the fix for each is different.
- Below a certain volume the increments are noise. We say so rather than producing a confident number from four conversions.
▸Why does this disagree with my ROAS target?
A ROAS target is an average applied to a decision about the margin. If your average is 3.0 and your last increment returned 1.4, the account looks healthy and the next euro is losing money.
▸How big a budget change should I make?
As a rule of thumb, 20–30% increments. Smaller tends to be lost in noise; larger can reset the learning phase, so you end up measuring the reset instead of the increase.
▸Does it work for a small account?
Up to a point. Marginal return is a statistical read, and below roughly fifty conversions per period you are mostly measuring randomness. We tell you when that is the case.
▸How do I know when to increase my ad budget?
When marginal ROAS – the return on the NEXT euro, not the average of every euro so far – still clears your break-even with room to spare. The meter watches exactly that pair and translates it into a spend headroom figure, so the decision stops being a feeling.
▸Why does average ROAS mislead when scaling?
Because averages carry your history. A campaign whose early cheap wins pulled the average to 3.5 can already be buying its newest customers at 1.2. Scaling on the average means paying yesterday's price expecting yesterday's return – marginal return is the only number the next euro cares about.
The module next to it
All six modules run on sample data in the open demo – no signup. Open Kepra →