Case study — measuring the right thing

The ads worked. The business closed anyway.

A fashion label, a blended return of 4.8× on advertising, and the cost line nobody was watching.

The client is not named, at their preference and ours. Every figure below is from their own reporting.

What we were hired to do

Grow online sales for a Cairo fashion label. Over the engagement the advertising returned EGP 4.84 for every EGP 1 spent — blended across channels, measured, not modelled. By any normal reading of an agency’s job, that is a success.

The business closed.

What was happening outside the advertising

Over the same period the label tripled production and took retail space in concept stores. Those deals carry rent plus a percentage of sales — commonly 15% to 25% — so the brand pays for the floor twice: once as a fixed monthly cost, and again on every item that leaves it.

Fixed costs36,000 → 48,000 per month, +33%

Advertising spend25,605 → 17,150, −33%

One concession, net−14,844 over four months

Read those three lines together and the outcome is already decided. Fixed cost rose while the thing paying for it was cut. By the following winter the forecast showed a loss no advertising performance could have reversed.

A 4.8× return on a shrinking budget cannot lift a rising floor. Multiplying a smaller number by a good ratio still gives you a smaller number — and the floor does not care about your ratio.Return is a ratio. Fixed cost is a floor.

What we would do differently

We did our job and the business still failed, so the honest question is what we should have done instead of only what we were asked for.

We should have recalculated break-even the week the lease was signed. We had the marketing numbers and not the cost structure, and we did not ask for it.

We should have refused to accept a budget cut without a contribution analysis. Cutting spend while fixed cost rises is not a saving; it is an acceleration.

We should have said out loud that marketing was no longer the constraint. It would have risked the retainer. It would have been the useful thing to say.

What every owner should take from this

Recalculate break-even on the day you sign

A new fixed cost raises the number your marketing has to clear. The day of the lease, not the quarter after.

A concession is a variable cost in disguise

Rent plus a percentage means you pay for the same shelf twice. Judge each location on contribution after both, not on its sales figure.

Blended return hides a failing channel

One strong channel can carry a bad one for months inside a single average. Measure contribution per channel and per location.

Break-even is recalculated with the client every time a fixed cost changes. No budget increase is recommended while contribution by channel is unknown. And we say when marketing is not the constraint, even when saying it costs us the engagement.What we changed at SNS

← All insights

Start here

Send us your links. We will tell you what we see.

The free scan is a written outside-in read of your public footprint — site, store, channels, reviews and search visibility — delivered within one working day. No charge, no obligation, and you can book the call at the same time.

Get the free scan The Growth Diagnostic