CTR Calculator clicks per impression

Click-through rate is the simplest ratio in advertising and the most misread. Divide the clicks by the impressions, multiply by a hundred, and you have a number between zero and one hundred that everybody quotes and almost nobody interprets correctly. This page exists because the arithmetic is trivial but the reading is not: the same advertisement can produce 0.1% and 4% on the same day, in two placements that look identical in a report.

What an impression is

An impression is a slot that loaded and was filled. It is not a person, and it is not proof that anybody looked. On a long feed, an advert placed below the third scroll is served, counted, and never seen; the platform still bills the impression and your rate carries the damage. That single fact explains most of the panic behind a “collapsing CTR”. Before changing the creative, check the viewability figure for the placement, or simply compare the same placement with itself from one week to the next.

The second complication is that impressions and clicks are usually counted by different systems. The ad platform counts one click per impression; your analytics counts sessions, including the same person returning through a bookmark an hour later. The numbers will differ by a few per cent and there is nothing to fix. Decide once which source you use for the ratio, and use it for every comparison.

Why the same advert has three different rates

Placement changes the intent of the audience. A search advert is shown to somebody who has already typed the problem into a box, which is why 2% is normal there. A display banner interrupts somebody reading a recipe, which is why 0.35% is normal there. A social feed sits between a friend's holiday photo and a video of a dog, and 1% is a respectable outcome. Reading a social rate against a search benchmark produces a decision that is confidently wrong: the campaign is fine, the comparison was not.

Position inside the placement matters as much as the placement itself. The first slot in a feed and the sixth slot in the same feed are different products, and the platform prices them accordingly. When a campaign's average position shifts because a competitor raised their bids, the rate moves without a single pixel of creative changing.

The trap of optimising the click

CTR is easy to inflate. A red arrow, a number in the headline and a question in the caption will raise it, and the raised rate will feel like progress until you look at the other end of the funnel. Clicks bought with curiosity arrive at a page that cannot satisfy them, and the conversion rate falls faster than the click rate rose. The cost per customer is what pays for the campaign, and it is the only number that cannot be improved by tricking somebody into a click.

A useful habit is to read three numbers together: CTR, conversion rate and cost per conversion. A change that lifts the first and lowers the last two is not an improvement, it is a more expensive way to collect bounce visits.

Testing without fooling yourself

One variable at a time, a thousand impressions per variant before reading the result, and at least three days so the weekday effect evens out. If the two variants differ by less than a fifth of the rate, the difference is noise; run longer rather than declaring a winner. Keep the audience fixed while you test the advert, and keep the advert fixed while you test the audience. Changing both and reading the total is how agencies report wins that do not exist.

Reading a change in either direction

A rate that rises is not automatically good: a narrower audience, a more qualified click, or a smaller placement can all raise it while revenue falls. A rate that falls is not automatically bad either — a broader keyword bringing cheaper, less interested people may still lower the cost per sale. The question to ask is never “did CTR improve?” but “did the campaign get closer to paying for itself?”. Used that way, a two-decimal ratio from two everyday numbers becomes a genuinely useful instrument rather than a scoreboard.