ROAS 14 was never a baseline

Can you stop treating your best month as your normal month?

A modest-fashion brand in Egypt. Recent months looked like this: 1,200 orders a month, ROAS between 13.5 and 14.5.

This month: everything is dead. ROAS between 6 and 9. Every time he considers scaling, CAC hits him in the face. He tried buy one get the second at 15% or 20% off, and it didn't deliver.

So what's actually happening?

First thing that has to be said: ROAS 14 in Egyptian fashion is exceptional. Not good. Exceptional.

And an exceptional number usually depends on two things: a strong collection and a season. Once the collection burns through the same audience and the season quietens, the number returns to its natural level.

So ROAS 6 to 9 isn't the disaster. The disaster is building spend and expectations on 14 and assuming it holds.

Second, the offer he tried was wrong for the timing. Two pieces at 15% off works on a customer who already wants to buy, by enlarging their basket. It doesn't work in a quiet market, because it doesn't create demand, it discounts demand that isn't there. You're eating margin without moving anything.

Third, scaling hits because the new audience doesn't respond to the same ads the same way. The ads that produced 14 were speaking to people who knew the brand. Expanding means speaking to people who don't know it at all, and that needs a different angle, not the same angle with a bigger budget.

What he should be looking at:

Net new reach, the share of new people Meta reaches each month compared with the previous month. Below 40% with frequency above 3 means the account has stopped expanding. Returning versus new: if sales fall while the share of past customers rises, you're living off the old list. Separating a campaign for new people with new angles and creative they haven't seen, and a retention campaign for existing customers built on a new collection rather than a discount. And the 1,200 customers who bought in previous months, contacted by SMS and WhatsApp about a new collection, before spending one extra unit on ads.

ROAS 6 isn't the disaster. The disaster is building your spend on 14 and assuming it holds.