Inon Cohen RazWhat happens to sales volume, customer acquisition cost and channel mix when the ad budget drops, and why the outcome is not always proportional.
The intuitive assumption is that half the budget brings half the sales. In practice it almost never works that way. Sometimes sales drop by far less than half, and sometimes they collapse. The difference depends on which half is cut.
Not every shekel in the budget works equally hard
Every ad account has layers. The first layer catches people who are already searching for the product or have already visited the site. It is cheap and efficient, but limited in size. The next layers reach wider audiences and cost more per purchase. As the budget grows, a larger share of it goes to the expensive layers.
That is why a cut that starts from the expensive layers hurts sales less than expected. The average acquisition cost even improves, and the reports look better.
The delayed effect
The problem arrives later. The wide audiences are the ones that feed the cheap layer: whoever is exposed to the brand today is the one who will search for it in a month. When you stop reaching them, remarketing and branded search keep working for a while on the existing pool, and then begin to empty. The real decline appears weeks later, and is not always linked to the cut.
How to cut correctly
- Start with campaigns whose acquisition cost is above the allowable cost, not with an even cut across the account.
- Keep a minimum budget for new audiences, even a small one.
- Track the company's total revenue, not only the return reported in the platforms.
- Check again after four to six weeks, when the delayed effect is already visible.
Cutting budget is a legitimate decision, and sometimes the right one. The question is whether it is made according to data, or according to a percentage set in advance.