How many clicks does it take to judge an ad?

There is no universal number, because it depends on your usual conversion rate. A workable reference is different: the ad should have spent well beyond your normal cost per lead without producing one, and only then does zero stop being chance.

Can I compare two ads by conversion rate?

Only when comparable volumes sit behind the percentages. A gap between 3% and 5% over a few dozen clicks means nothing: at that volume both figures could come from the same audience behaviour.

What should I do while the data is still thin?

Neither pause nor scale. Remove the reasons data accumulates slowly: an audience narrowed too far, competition inside the account, a budget split across many nearly identical ad sets.

When you have enough data to switch an ad off

Zero conversions is not a reason to pause. The working reference: spend so far against your usual cost per lead.

ADS Beast editorial teamPublished September 3, 2026Updated September 14, 20263 min read

An ad has run a few dozen clicks without a single enquiry and your hand reaches for the pause button. Sometimes that is right. Sometimes you are switching off an ad that never got the chance to show what it does. The difference comes from volume, not from the calendar or from a feeling.

In short

Zero conversions is not a conclusion by itself: what matters is the money behind it.
The reference point for pausing is spend well past your usual cost per lead.
Percentages without volume mean nothing, especially over a few dozen clicks.
Thin data is more often a structural problem than a creative one.
Record the decision together with the data it was made on.

Why time is a poor criterion

"It has run for a week and nothing" sounds convincing, but a week can mean a thousand clicks or twenty. The calendar measures your patience, not the behaviour of the audience. Count what actually happened: impressions, clicks, spend, conversions.

Когда выключать объявление. Сравните траты с ценой лида: Выключайте, если потрачено втрое больше обычной цены лида без результата; Проверьте объём кликов: Проценты без объёма ничего не значат при паре десятков кликов; Ищите структурную причину: Узкая аудитория, дубли кампаний, дробление бюджета, низ — Порядок решения по данным, а не по календарю

A working reference

an ad can be switched off once it has spent well beyond your usual cost per lead and produced none. If a typical enquiry costs you a certain amount and this ad has spent three times that with nothing to show, bad luck stops being a plausible story.

The rule is deliberately blunt. It protects against two mistakes at once, pausing on instinct and holding a hopeless line for months.

Comparing two ads

This is where the mistake shows up most. Two ads, one has a higher conversion rate, decision made. But when a few dozen clicks sit behind those percentages, the gap is chance: the same people in a different order would have produced the opposite picture.

Before comparing rates, look at the denominator. When it is small on both sides, the answer is not that the first ad is better. The answer is that there is no data yet.

When data will not accumulate

Slow accumulation usually points at structure rather than at the creative:

The audience is narrowed so far that there are almost no impressions.
The same query runs in several campaigns, and they split impressions between them.
The budget is spread across many nearly identical ad sets, each collecting separately.
The bid keeps the ad near the bottom of the auction, where clicks are scarce by definition.

Fix those before discussing ad copy.

What to do next

Before switching a line off, write down two numbers: what it has spent and what an enquiry normally costs you. If the first is well past the second, the decision stands. If it is not, you do not have a conclusion yet, and the line deserves another look later.

All blog articles