Most underperforming Google Ads accounts are not badly written. They are badly divided. Someone built one campaign per service, then split each one by location, then added a brand campaign and a competitor campaign, and ended up with eleven campaigns sharing the budget of two. Growth Bully, a Malta performance marketing agency, sees the same pattern in most accounts handed over for audit: the copy is fine, the targeting is fine, and the structure has quietly made automated bidding impossible.
Structure is not an organisational preference. It decides how much evidence each bidding decision is based on.
What is Google Ads account structure, and why does it decide performance?
Account structure is how you divide spend and conversion data between campaigns and ad groups. It matters because bidding is automated and automation learns per campaign. Every extra campaign splits the same conversion history into smaller pieces, so the structure you choose sets a ceiling on how well the platform can bid, before anyone writes an ad.
That is the whole argument. A clean structure is not tidy, it is well fed. The question is never how many campaigns make sense on a whiteboard, it is how many your conversion volume can support.
How many Google Ads campaigns should you actually run?
As few as your reporting genuinely needs. Take your conversions from the last thirty days, divide by the number of campaigns you are planning, and look at what each one is left with. If a campaign cannot produce conversions steadily each week on its own, it cannot support its own bid target and should be merged into one that can.
Run that division before you build anything. Three checks make the answer obvious.
- Conversions per campaign per month. Count real conversion actions, not clicks and not form views. Thin campaigns stay stuck in learning and spend inefficiently while they sit there.
- Daily budget per campaign. A campaign funded below the cost of a few clicks a day cannot gather evidence, whatever its target is set to.
- Decisions you would actually take. If you would never move budget between two campaigns based on their results, they are reporting rows, not campaigns, and the split costs you data for nothing.
What earns its own campaign?
Only four things. A different budget decision, where you would genuinely fund one and starve the other. A different conversion action, such as a call versus a purchase. A different language or market where the ads must change. And brand traffic, which is cheap, converts differently, and distorts any target it shares.
Everything else belongs in ad groups. The test is whether the split changes what you would do on a Monday morning. If the answer is that it only changes what a report looks like, build a segment, not a campaign.
What does not deserve its own campaign?
Each individual service, each product line, each town, each match type and each device. These are the five most common ways an account gets fragmented, and all five are usually attempts to control something that is better controlled inside one campaign with sensible ad groups and exclusions.
Fragmentation also hides your real problem. When eleven campaigns each have a handful of conversions, no single one has enough data to tell you anything, so decisions get made on noise. Consolidating does not improve any individual campaign, it makes the account readable for the first time.
How should ad groups be organised inside a campaign?
Group by the promise the landing page makes, not by keyword similarity. One ad group per intent, with keywords that could all be answered by the same page and the same headline. If two keywords need different pages or different proof to convert, they belong in different ad groups.
Keep the count low enough that every ad group gets impressions. Dozens of near-identical ad groups built around single keywords made sense when bidding was manual and matching was literal. Today they mostly split data and compete with each other, and the search terms report plus a maintained negative keyword list does the control job far better.
Where does Performance Max sit in the structure?
Treat it as a separate campaign with a clear boundary, because it will take traffic your search campaigns would otherwise have won. Decide in advance what it is allowed to claim, exclude the terms you want handled by search, and never judge it on the same report line as a tightly targeted search campaign.
The common failure is running it alongside a well-built search campaign with no containment, then concluding one of them works and the other does not. Both numbers are contaminated. There is more on where it fits and where it does not in our breakdown of Performance Max.
How do you consolidate without losing what already works?
Carefully, in one direction, and one change at a time. Merging campaigns resets the learning period, so the sequence matters far more than the speed. Move data towards your strongest campaign, keep the old campaigns available for comparison, and change nothing else for a fortnight so the result is attributable.
- Write down the decisions you take monthly. That list is the only legitimate reason to keep campaigns apart.
- Pick the destination campaign first, always the one with the strongest conversion history.
- Move ad groups into it, rather than rebuilding them, so the history travels with them where it can.
- Pause the emptied campaigns instead of deleting them, so the old data stays available for comparison.
- Leave budgets and targets alone for two weeks. Changing three things at once means learning nothing from any of them.
Expect performance to dip while the merged campaign relearns. That dip is the cost of the fragmentation you already had, paid once.
What should you fix before touching structure at all?
Conversion tracking. Restructuring an account that counts the wrong things just reorganises bad evidence, and automated bidding will optimise faithfully towards whatever you told it to value. Confirm what is being counted, that it fires once, and that it still works under consent rules, before you move a single ad group.
That order holds generally. Tracking first, then structure, then budget, then creative. Accounts that follow it rarely need the kind of rescue that starts with a tracking rebuild, and the budget conversation only becomes honest once the structure can actually report, which is also when increasing spend stops being a guess. How the overall split between channels should look is a separate question, covered in budget allocation, and the same discipline sits behind every account we run through our Google Ads work.
If your account has more campaigns than monthly conversions to feed them, the structure is the problem and no amount of bid tuning will fix it. Send us the account and we will show you what consolidating it is worth, before anything is changed.