The Google Ads account audit: what we look for, and why it matters.
Updated: Aug 19
Most Google Ads accounts are not badly managed. They are quietly out of date.
What this piece covers
Underperformance is rarely one dramatic error. It is drift, budgets migrating towards whichever campaign spends fastest, search queries broadening, tracking that stopped working the day the site was redesigned.
Automated bidding made measurement errors expensive. The data that fills the dashboard is the data the bidding model optimises against, so a broken conversion set misdirects spend as well as reporting.
A serious audit is a sequence, not a list. Eight checks, in order, from account structure through to growth headroom, several are meaningless if the one before them has failed.
Benchmarks diagnose; they do not set targets. Cost per click and conversion rate vary too widely by vertical and market for an imported average to describe your ceiling.
The four markets do not behave alike. Francophone Côte d'Ivoire, multilingual Mauritius, tourism-dominant Seychelles and EU-regulated Réunion each break a different assumption.
The finding is not the value. The plan is. A gap identified in January and actioned in April has already cost three months of spend.
Most Google Ads accounts are not badly managed. They are quietly out of date.
That distinction matters, because it changes what an audit is for. An audit is not an accusation. It is a structured way of establishing how much performance is currently unavailable to an advertiser and what it would take to release it. In our experience across Côte d'Ivoire, Mauritius, Seychelles and Réunion, the gap between an account's current output and its realistic output is almost always larger than its owner expects, and almost never caused by a single dramatic error.
It is caused by drift.
Google Ads Audit: Why accounts drift
A Google Ads account is a living system. Budgets migrate towards whichever campaign spends fastest rather than whichever earns most. Search queries broaden as matching behaviour becomes more interpretive. Conversion tracking that was configured correctly at launch stops working the day the website is redesigned, and nobody is told.
None of this announces itself. The account keeps running. The reports keep populating. Spend continues to clear. The drift stays invisible until somebody deliberately goes looking for it, which, in most organisations, is not part of anyone's job description.
The shift towards automated bidding has made this considerably more expensive. When bids were set by hand, a measurement error produced a misleading report. Today it produces a misleading report and a misdirected algorithm, because the same conversion data that fills the dashboard is the data the bidding model optimises against. An account with broken measurement does not merely mislead its owner. It teaches itself the wrong lesson, every day, at scale, with a budget attached.
The eight things a serious audit examines
A credible audit is not a list of observations. It is a sequence, and the sequence matters as much as the content, because several of these checks are meaningless if the one before them has failed.
One: account settings and structure. Time zone, currency, auto-applied recommendations, account-level exclusions, naming conventions. Then the structural question: are brand, generic and competitor terms separated, or is brand demand quietly subsidising everything else and flattering the reported cost per acquisition?
Two: conversion tracking and measurement. The most expensive place in an account to be wrong, and the hardest place to notice being wrong. Are the primary goals tied to real business value, or is a newsletter sign-up being optimised towards with the same weight as a sale? Is any tag firing twice? Does the account know which enquiries actually became revenue?
Three: campaign settings and targeting. Where budget leaves the market without ever reaching a plausible buyer. The defaults are permissive by design, and permissive defaults are expensive in small, multilingual and island economies where a slightly loose radius reaches an entirely different country.
Four: keywords and search terms. The keyword list states what an advertiser intended to buy. The search terms report states what they actually bought. The distance between the two is the single most reliable predictor of wasted spend in a Search account.
Five: bidding and budget allocation. Smart bidding is usually discussed as a strategy choice. It is more usefully understood as a data threshold. It rarely fails because it was the wrong strategy — it fails because it was given too few conversions, too short a learning period, or a target no history supports.
Six: ads, assets and creative coverage. Creative is the last lever automation cannot supply on an advertiser's behalf. The system will assemble and test combinations faster than any team can — but only from the raw material it is given, and most accounts supply materially less than they are entitled to.
Seven: landing pages and conversion experience. The account can be flawless and still fail here. It is also, in most organisations, the part of the funnel the media team does not control, which is precisely why it goes unexamined.
Eight: growth headroom. The first seven checks establish what an account is doing badly. The eighth establishes what it is not doing at all — and in a well-run account, that is usually the larger number.
The market matters more than the benchmark
Cross-industry benchmarks are a useful diagnostic and a poor target. Cost per click and conversion rate vary so widely by vertical, intent and market that an imported average tells an advertiser very little about their own ceiling.
What does travel well is local context, and the four markets we work across do not resemble one another. Côte d'Ivoire is Francophone, XOF-denominated and heavily agency-intermediated, with a population approaching thirty million. Mauritius is multilingual with an advanced financial services economy and a domestic audience of 1.27 million, a market where audience precision matters more than reach. Seychelles is tourism-dominant at 132,000 people, which means most valuable demand originates outside the country. Réunion is a French overseas department inside the EU regulatory perimeter, where audience data practices acceptable elsewhere in the cluster require a documented GDPR lawful basis.
An audit that applies one set of assumptions across all four will be wrong in at least three of them. Creative written once and translated four times will underperform in at least three of them, too.
The finding is not the point. The plan is.
The value of an audit finding decays quickly. A negative keyword gap identified in January and actioned in April has already cost three months of spend. An audit that produces a document is an expense; an audit that produces a funded, sequenced plan is an investment.
That is why our own commitment is written as an outcome rather than a deliverable:
Within thirty days, advertisers move from audit to action with a clear, data-backed path to scalable growth on Google Ads.
Thirty days is a discipline, not a promise of results. It commits to the speed of diagnosis, decision and deployment — the parts we control. Performance outcomes depend on budget, competition and conversion volume in each market.
Three things make that timeline realistic rather than aspirational. Our Portfolio-Led Market Approach plans accounts across the cluster rather than in isolation, so that seasonality and vertical evidence in one market inform decisions in the next. The 30-Day Conversion Sprint takes an account from audit findings to deployed corrections to a measured growth plan, with a performance baseline captured before anything changes — so the improvement can be proved rather than asserted. And the Africa Centre of Excellence means an advertiser in Seychelles has access to the same depth of technical capability as one in a far larger market.
Start with the guide
We have published the full method as a downloadable guide: eight checks, reference ranges for what a healthy account looks like, and a forty-point scorecard designed to be printed and completed against your own account. It is written to be used by whoever manages your Google Ads in-house team or agency without our involvement.
Would you rather we ran it?
Are you certain your Google Ads account is running as efficiently as it could be? Request a no-cost Google Ads account audit from our Centre of Excellence and receive a practitioner review of your account structure, measurement, targeting, bidding and creative, with a prioritised thirty-day action plan attached to it.
Frequently asked questions
What does a Google Ads account audit cover?
A credible audit runs as a sequence of eight checks: account settings and structure; conversion tracking and measurement; campaign settings and targeting; keywords and search terms; bidding and budget allocation; ads, assets and creative coverage; landing pages and conversion experience; and growth headroom. The order matters, because several checks are meaningless if the one before them has failed.
Why do Google Ads accounts lose performance over time?
Most accounts are not badly managed — they are quietly out of date. Budgets migrate towards whichever campaign spends fastest rather than whichever earns most, search queries broaden as matching behaviour becomes more interpretive, and conversion tracking configured correctly at launch stops working the day the website is redesigned. None of this announces itself, so the drift stays invisible until somebody deliberately looks for it.
Why does conversion tracking matter more under automated bidding?
When bids were set by hand, a measurement error produced a misleading report. Under automated bidding it produces a misleading report and a misdirected algorithm, because the same conversion data that fills the dashboard is the data the bidding model optimises against. An account with broken measurement teaches itself the wrong lesson every day, at scale, with a budget attached.
Are industry benchmarks a reliable target for Google Ads performance?
Cross-industry benchmarks are a useful diagnostic and a poor target. Cost per click and conversion rate vary so widely by vertical, intent and market that an imported average says very little about an advertiser's own ceiling. Local context travels better than a benchmark does.
Does an audit differ across Côte d'Ivoire, Mauritius, Seychelles and Réunion?
Materially. Côte d'Ivoire is Francophone, XOF-denominated and heavily agency-intermediated, with a population approaching thirty million. Mauritius is multilingual with an advanced financial services economy and 1.27 million people, so audience precision matters more than reach. Seychelles is tourism-dominant at 132,000 people, meaning most valuable demand originates outside the country. Réunion sits inside the EU regulatory perimeter, where audience data practices require a documented GDPR lawful basis.
How quickly should an advertiser move from audit findings to action?
The value of an audit finding decays quickly — a negative keyword gap identified in January and actioned in April has already cost three months of spend. Our commitment is that within thirty days advertisers move from audit to action with a clear, data-backed path to scalable growth. Thirty days is a discipline covering diagnosis, decision and deployment, not a promise of performance outcomes, which depend on budget, competition and conversion volume in each market.
365 Digital is an authorised Google Ads sales representative in Côte d'Ivoire, Mauritius, Seychelles and Réunion. Billing for media remains direct with Google. Our advisory, audit and account management services are provided at no additional cost to the media investment. 365 Digital's audit engagement is advisory in scope. Deliverables comprise a documented account review and a prioritised schedule of recommendations. 365 Digital does not access, amend, or manage live campaign settings as part of this engagement. Responsibility for implementation, and for outcomes arising from implementation, rests with the advertiser or its appointed agency.
Comments