What a real account audit checks, and where AI helps

The word audit has lost its value. In plenty of places it means somebody opened the account, took ten screenshots, and found three settings that are not ideal.

A serious review does not look like that, and it does not start in the account.

The four layers that have to be checked

Measurement. This comes first because everything else depends on it. What counts as a conversion, whether anything is double counted, whether there is a value, how long the window is, whether the account number matches reality. If this layer does not hold, no claim can be made about any other number in the account.

Structure. Whether brand and non brand traffic are separated, whether exclusions exist, whether campaigns overlap, whether the spend sits where the revenue is.

Offer and landing. Whether the page delivers what the ad promised. This is the most common silent loss. The campaign is fine, the ad is fine, and the page is about something else.

Business context. What margin you work on, what a customer is worth, who you do not want. Without that, an audit can only tell you the account is tidy, not that it is profitable.

Where AI genuinely speeds things up

Reading through the data and finding patterns. Categorising thousands of search terms, sorting ad copy by theme, collecting feed errors, spotting outliers. That saves hours, sometimes days.

Where it does not

At the judgement. A machine can tell you a campaign has a higher cost per acquisition than the average. It cannot tell you that this is because customers from that segment are worth twice as much. That requires knowing how the business works, and that information does not live in the account.

What you can expect from an audit

Not a list of settings. Three things: where money is being lost, how much is at stake, and in what order to fix it. If there is no order at the end, the document is an inventory, not a plan.