When to scale ad spend, and when to sit on your hands

Scaling spend is where most accounts lose their discipline, because the pressure to grow arrives before the evidence that growth is affordable.

The rules below are the ones we actually run, and none of them are complicated. They are just decided in advance rather than in the middle of a bad week.

Decide the floor before you push

Scaling costs efficiency. Reaching more people means reaching less responsive people, and return falls. That is not a failure, it is arithmetic.

So the decision to make first is not a target, it is a floor: the return below which you stop. On one inherited account we set a floor above 3.5x before increasing budget, roughly double what the account had historically managed, and held it while spend rose. Return came down from the opening month’s peak, exactly as expected, and stayed above the line.

Setting the floor in advance turns an uncomfortable surprise into a planned trade.

Scale against proof, not hope

Every budget increase should be made against results already on the board.

On a wellness account we grew monthly spend 44 times over a year — from ₹8,000 to ₹3,50,000 — and every step up followed a month that had earned it. The account also spent its first two months returning nothing at all while we found the product, content and audience combination. Spending into that period would have been spending into a question rather than an answer.

The corollary is that a good week is not proof. A month is usually the smallest honest unit, and for considered purchases a quarter is better.

Accept a lower multiple for a larger base

This is the trade founders find hardest. An account peaking at 11x on ₹26,000 of monthly spend is not better than the same account at 6x on ₹3,50,000 — it is very much smaller.

Chasing the peak multiple caps the brand at a fraction of its size. The question is not which number is higher; it is which one leaves more contribution after costs.

Spend less in a soft month

Most accounts push through a weak month at full budget on the theory that consistency matters. It buys a poor month at full price.

On a mature kids accessories account, the lightest-spend month of seven returned the highest ROAS of the period, because pulling back when conditions did not support volume was the right call. Budget flexed 47% across those months while the return floor held.

The discipline is: spend is the variable, the return threshold is not.

Judge the increment

When you have scaled, the average is a poor guide to whether it worked. What matters is what the additional spend returned.

On the same inherited account, ₹50,000 of extra quarterly spend returned roughly ₹5,10,000 in additional revenue — over 10x on the incremental rupee. That is the figure that justified continuing, and it is invisible if you only look at the blended average.