Google Ads Account Structure Audit: When to Split, When to Merge

Account structure is where good intentions do the most quiet damage.

Someone builds an account carefully. They separate everything: campaign per category, ad group per theme, sometimes ad group per keyword. It looks organised. It looks professional. Every report slices cleanly. And it performs worse than the messy account it replaced, because in the age of automated bidding, that neat separation starved the system of the data it needed to work.

Then, in reaction, someone else consolidates everything into a handful of broad campaigns. Cleaner, fewer moving parts, plenty of data. And now they have lost the control they needed to protect margin, separate brand, or manage different markets.

Both are structural faults. Neither is obviously wrong from the outside. The right structure is not the tidiest one or the most consolidated one. It is the one that gives you control you will actually use, while giving automation enough data to learn from. Auditing structure means finding where an account has drifted to one extreme or the other.

The Old Rules Do Not Apply Any More

For years, the accepted wisdom rewarded granularity. Tight keyword-to-ad-group mapping, single-keyword ad groups, one theme per ad group, so that every ad matched every query as closely as possible. Under manual bidding, that made sense. You controlled the bid on every keyword, so more control was better.

Automated bidding changed the incentive completely.

Smart Bidding learns from conversion data, and it learns per campaign or bid strategy. Split your account into fifty tiny campaigns and you split your conversion data fifty ways, so no single campaign has enough signal for the system to bid well. The granularity that helped under manual bidding actively harms under automated bidding, because it fragments the very data the automation depends on.

So the first thing a structure audit has to establish is whether the account was built for a world that no longer exists. Many were, and they are quietly underperforming as a result.

The Core Question: Control You Will Use

There is a single test that resolves most structural decisions. For any split in the account, ask: what control does this give me that I would actually use?

A campaign split is worth it if it lets you set a different budget, a different bid target, a different geographic focus, or a different commercial priority that you genuinely need to manage separately. It is not worth it if it exists only to make the account look organised, because that tidiness comes at the cost of fragmented data.

Run every existing split through that question. Brand and non-brand deserve separate campaigns, because you genuinely manage them differently. Two campaigns split for no reason other than neatness, sharing the same targets and priorities, should probably be one, because merging them gives the automation more data and costs you no control you were using.

Structure should follow the decisions you make, not the categories on your website.

Over-Segmentation: The Starvation Problem

The most common structural fault in modern accounts is too many campaigns, each too small.

Audit for:

Campaigns with very low conversion volume, below the level a bid strategy needs to learn Many campaigns targeting closely related things that share commercial priority Single-keyword or near-single-keyword ad groups left over from an older approach Whether the segmentation reflects real management decisions or just categorisation Bid strategies that never exit the learning phase because their campaigns are starved

The symptom is automation that behaves erratically, because it is trying to optimise on too little data. The fix is consolidation: merging campaigns that share a commercial purpose so the combined conversion volume gives the bidding something to work with.

The instinct that resists this is understandable. Merging feels like losing control. But control you cannot exercise well, because the automation underneath it has no data, is not control. It is the appearance of it.

Under-Segmentation: The Control Problem

The opposite fault is real too, and consolidation taken too far causes it.

Audit for:

Campaigns mixing products or services with very different margins under one target Brand and non-brand demand blended in the same campaign Different markets, currencies or languages grouped together unnecessarily High-priority and low-priority activity sharing a budget and a bid strategy A structure so consolidated that you cannot control what matters commercially

The symptom here is not erratic automation. It is an account that runs smoothly while doing the wrong thing, scaling low-margin products because they sit in the same campaign as high-margin ones, or letting brand demand flatter a campaign you cannot separately assess.

The fix is targeted segmentation: splitting out the things you genuinely need to control separately, and only those things.

Brand and Non-Brand: The One Split That Is Almost Always Right

If there is a near-universal rule in structure, it is this one.

Brand and non-brand belong in separate campaigns, in almost every account. They behave completely differently. Brand traffic is cheap, converts highly, and largely reflects demand you already own. Non-brand is where genuine acquisition happens, at higher cost and lower conversion rates.

Blend them and several things break. Your blended ROAS looks healthy while non-brand acquisition quietly declines. Your bidding optimises across two utterly different types of traffic as if they were one. You lose the ability to fund acquisition and defend brand as the separate decisions they are.

Audit for:

Whether brand and non-brand are cleanly separated Whether brand terms are leaking into generic or Shopping campaigns Whether Performance Max is absorbing brand demand without exclusions What share of the account's conversions and revenue is actually brand

This split is worth it precisely because it gives you control you will use, on the most important distinction in the account.

Match Types and Ad Group Structure

Structure is not only about campaigns. Within them, the audit continues.

Audit for:

Whether ad groups are so tightly themed that they starve individual bidding of data Whether match type strategy suits automated bidding, where broad match plus strong negatives and Smart Bidding can outperform tightly controlled exact match Whether responsive search ads have enough asset variety to work with Whether the ad group structure reflects how you manage, or how you once categorised Whether legacy single-keyword ad groups are fragmenting performance

Modern structure inside a campaign tends toward fewer, better-fed ad groups than the granular approach of the past, though the right answer still depends on volume and how much genuine control each split provides.

Legacy and Duplication

Accounts accumulate structural debt over time, and an audit should clear it.

Audit for:

Campaigns left running after a restructure, competing with their replacements Duplicate campaigns targeting the same queries and splitting data Paused clutter that confuses reporting and analysis Overlapping audiences across prospecting and remarketing Old experiments never cleaned up

Duplication is doubly damaging: it wastes spend through internal competition and it fragments the conversion data that automation needs. Consolidating or removing it is often a quick, high-value structural fix.

Scoring the Findings

As with any audit, structure findings need ranking, not just listing.

Prioritise the brand split and the starved-campaign consolidation first, because they address the two faults that most directly undermine automated bidding. Save the larger restructures for when their value is clear, because restructuring carries risk.

A Warning About Restructuring

The most important thing to say about structure is also the most counter-intuitive: often, the right answer is to leave it mostly alone.

A full rebuild discards learning history. Every campaign that gets recreated starts its bidding from scratch, and the account can dip while the automation relearns what it already knew. That cost is real, and it means a rebuild has to be justified by a genuine structural fault, not by a preference for how things look.

Most accounts benefit far more from a few targeted structural fixes, splitting brand, consolidating starved campaigns, removing duplication, than from a wholesale reorganisation. The bar for "rebuild it" should be high. The bar for "fix these three specific things" should be low.

Common Structure Mistakes

Building for tidiness rather than control A neat account that starves automation performs worse than a messier one that feeds it.

Applying old granularity rules under Smart Bidding Single-keyword ad groups and hyper-segmentation fragment the data automation needs.

Consolidating so far you lose control Mixing margins, markets or brand and non-brand under one target removes levers you need.

Blending brand and non-brand The single most damaging structural fault, because it corrupts every acquisition number.

Rebuilding by default Restructures discard learning history and should be justified, not assumed.

Leaving legacy and duplicate campaigns running They waste spend and fragment data at the same time.

How AI Can Help

Structural analysis means looking across the whole account at once, at conversion volumes, overlaps, margin distribution and data fragmentation, which is exactly the kind of pattern work that is slow by hand.

AI can process:

Which campaigns fall below the conversion volume automation needs Where brand demand is leaking across campaign types Duplicate and overlapping campaigns competing internally Whether segmentation reflects real management decisions or just categorisation Ranked structural fixes by impact, effort and confidence

GoogleAdsAudits.com reviews account structure, tracking, bidding, feed and channel mix, then produces a prioritised action plan positioning each recommendation by impact, effort and confidence.

The decision about which controls the business actually needs still belongs to the business.

Final Thoughts

Good account structure is not the tidiest arrangement or the most consolidated one. It is the structure that gives you the control you genuinely use while feeding automation enough data to bid well.

A structure audit finds where an account has drifted to one extreme: over-segmented and starving the bidding, or over-consolidated and removing control you need. It almost always separates brand from non-brand, consolidates campaigns that share a purpose, and clears out legacy and duplication. And it resists the temptation to rebuild, because learning history is valuable and a few targeted fixes usually beat a wholesale reorganisation.

Structure should follow the decisions you make, not the categories on your website. Get that right and everything built on top of it, the bidding, the budgets, the reporting, works better for it.

Build a personalised Google Ads action plan that identifies wasted spend, scalable opportunities and the fixes your team should prioritise next.

Frequently Asked Questions

What is the best Google Ads account structure?

The one that gives you control you will actually use while feeding automation enough conversion data to bid well. There is no universal template; the right structure follows the decisions you genuinely need to manage separately, such as brand versus non-brand, or different margins and markets.

Are single-keyword ad groups still a good idea?

Generally no, under automated bidding. They fragment conversion data across too many small units, which starves Smart Bidding of the signal it needs. The granular approach that suited manual bidding tends to harm automated bidding.

Should I split brand and non-brand into separate campaigns?

Almost always yes. They behave completely differently, and blending them lets cheap brand demand flatter your acquisition numbers while corrupting how bidding optimises. It is the one split that is right in nearly every account.

What is over-segmentation?

Too many campaigns, each too small to give the bid strategy enough conversion data to learn. The symptom is erratic automation. The fix is consolidating campaigns that share a commercial purpose.

When should I consolidate campaigns?

When several campaigns share the same commercial priority, targets and management approach, and none has enough conversion volume on its own. Merging them gives automation more data at no real cost to control.

Should I rebuild my account if the structure is wrong?

Rarely as a first step. A rebuild discards learning history and can dip performance while automation relearns. Most accounts benefit more from a few targeted fixes, splitting brand, consolidating starved campaigns, removing duplication, than from a full reorganisation.

How does structure affect Smart Bidding?

Smart Bidding learns per campaign or bid strategy from conversion data. Fragmented structure splits that data thin, so no campaign has enough signal. Sensible consolidation gives the automation the volume it needs to bid accurately.

AUTHOR BIO

Jack Felstead is an award-winning Paid Media, AI and Digital Growth Consultant with 17 years of experience helping businesses improve customer acquisition, advertising efficiency, revenue and profit.