The Most Common Google Ads Mistakes in Ecommerce: 2026 Audit Report (122 Accounts)

The Store Growers Ecommerce Google Ads Audit Report 2026 analyses 122 Google Ads account audits we recorded between May 2024 and September 2026, most of them for ecommerce stores. It counts the problems we found and the highest ROI fixes.

Most of the damage sits in the basics, not in campaign choice.

Key findings:

• 66 of the 122 accounts had a conversion tracking problem.
• 51 had no product segmentation, the most common single problem and the first thing we fixed in 14 audits.
• 64 of the 84 accounts running Performance Max had at least one Performance Max problem.
• In 50 of the 79 audits where the advertiser named their problem up front, the real cause sat somewhere else in the account.

Below are the ten mistakes we found most often, what each one does to an account, and how to fix it.

The findings at a glance

#	Mistake	Accounts	What it does
1	Inaccurate conversion tracking	66 of 122	Google optimises on wrong or missing data
2	No product segmentation	51 of 122	Bestsellers and slow movers share one budget and one target
3	Arbitrary ROAS or CPA targets	38 of 122	Google spends too loosely, or can’t spend at all
4	Brand and non-brand searches mixed	36 of 122	Weak campaigns look profitable
5	Performance Max misconfigured	64 of the 84 running it	Spend goes to places nobody chose
6	Dynamic remarketing missing or misconfigured	50 of the 71 checked	Remarketing shows the wrong products, or none
7	Unmanaged search terms and keywords	75 of 122	Budget goes to searches that never convert
8	Product titles misaligned with search queries	39 of 122	Products don’t show for the searches that sell
9	Budget fragmented across too many campaigns	29 of 122	No campaign gets enough data to learn
10	Auto-applied recommendations switched on	31 of the 88 checked	Google changes targets and settings for you

You’ll note that the order seems kind of off. That’s because the numbers use different bases: some problems we checked in every account, others only where they applied. So the list follows the order in which we’d fix them, starting with the basics.

Mistake 1: Inaccurate conversion tracking (66 of 122 accounts)

This one feels so basic but is so common. Smart bidding learns from the conversions that come into the account. So if you’re sending no or the wrong data, it optimizes for the wrong thing.

In our 2026 Ecommerce Google Ads Audit Report, 66 of the 122 accounts had a conversion tracking problem. In 16 of them, conversion tracking was so broken that the account was flying blind.

The problems that came up most:

• Micro conversions instead of purchases. Pageviews, add to carts or checkouts counted as primary conversions (27 accounts)
• Conversion settings that distort results, such as double or even triple counting, wrong currencies, or attribution settings (26)
• Purchase tracking that was broken or inactive (21)
• Phone, offline or B2B orders missing from Google Ads (10)

In one gifts store, the main conversion action was a site search goal which recorded about 80 conversions, against 16 purchases. In another account, the purchase conversion imported from GA4 had been inactive for over 60 days.

Fixing tracking was the first thing we recommended. Any other optimization comes after.

Why it happens: errors around conversion tracking usually slip in by accident. Things are checked during campaign setup, and aren’t often revisited. When a site theme or tooling changes, tracking breaks quietly. Only when results really start sliding, deeper investigation reveals this issue.

How to fix it:

1. Open up the Goals section of your Google Ads account
2. Make sure you have a single purchase as your only primary conversion. Keep add to carts and checkouts as secondary, so you still see them without bidding on them.
3. Add in an alternative purchase conversion action as secondary so you can keep tabs on the accuracy of your primary one. We often use GA4 purchases for this purpose.
4. Check the settings of the purchase action: count “every”, pass the order value, and use one purchase action as primary, not two.
5. Turn on enhanced conversions. If you sell by phone or to businesses, import those sales as offline conversions.

More in our guide to Google Ads conversion tracking.

Mistake 2: No product segmentation (51 of 122 accounts)

Most stores start with all products in one campaign, and that’s fine at first. The problem starts when it stays that way. Google then decides where the budget goes, and it rarely picks your best products.

51 of the 122 accounts had no product segmentation, the most common single problem in the report.

One home and lighting store ran one all-products campaign per country. 785 products had spent money without a single conversion, and one product took about $3,000 of an $18,000 monthly budget.

In an outdoor apparel store, a catch-all Shopping ad group took most of the impressions, while the ad groups for specific brands returned a ROAS of 11.4 and 8.7.

Why it happens: By default, Performance Max and Shopping include all products from the linked Google Merchant Center account in your product group, so that’s how most campaigns go live. Once the campaign runs, splitting the catalog doesn’t get revisited. When segmentation does happen, mistakes often slip in, so all products to get advertised instead of a tight selection.

How to fix it:

1. Open the Products report in Google Ads, set the last 90 days and sort by cost.
2. Look at which products eat up most budget. Is this in line with what you want to happen?
3. Revisit which products to include or exclude, or even consider revisiting your Shopping structure
4. Use custom labels in your feed so the split stays up to date without manual work. 12 of the accounts had no custom labels or supplemental feeds to control this.

See our guide to Google Shopping custom labels.

Mistake 3: Arbitrary ROAS or CPA targets (38 of 122 accounts)

Every automated bidding campaign runs on a target. In almost a third of the accounts we audited, that number had nothing to do with the business behind it.

38 of the 122 accounts had a ROAS or CPA target with no link to their margins or goals. It went wrong in two directions:

• Too low. One medical supplies store set targets of 50–120% while its campaigns achieved 150–220%. A target that low is easy to hit, so Google can spend more freely and buy sales at a ROAS well below what the store needs.
• Too high. A jewelry store hit its ROAS target but couldn’t grow, and felt it had run out of keywords. The target sat far above what the business needed, so Google couldn’t spend more without missing it.

Why it happens: when launching a new campaign, advertiser often pick a lower than ideal target to get things going, but this doesn’t always get revisited. Or the number is picked from Google’s suggestion. Seasonlity, comptition, margins, prices and goals change after that, and the target stays where it was.

How to fix it:

1. Open the bid strategy settings of your campaigns and note the target of each one.
2. Work out your break-even ROAS: 1 divided by your margin. At a 40% margin, that’s 250%. Does each target relate to that number, and to what the campaign already achieves?
3. If not, move it in steps of about 20%, and give each step one to two weeks before you judge it.

More in our target ROAS guide.

Mistake 4: Brand and non-brand searches mixed (36 of 122 accounts)

When branded searches end up in a generic campaign, that campaign looks far better than it is.

36 of the 122 accounts ran brand searches inside non-brand campaigns.

In one automotive parts store, brand searches brought in about a third of the Performance Max campaign’s sales, even though brand exclusions were switched on.

In a wallet store, most Performance Max conversions came from brand searches, and only 7 of 121 non-brand clicks mentioned the product the store actually sells.

Why it happens: Performance Max picks up brand searches by default, and a blended ROAS looks good enough that nobody splits it out. When brand exclusions are set up, they don’t always catch every brand search, so some still slip through.

How to fix it:

1. Open the search terms report of your campaign. How many conversions come from searches that include your brand?
2. If it exceeds 25% of conversions or conversion value, we usually opt to split out the branded traffic
3. Brand searches in their own Search campaign, and if relevant, dedicated branded Shopping or pMax campaigns
4. In your generic campaign, add your brand as a brand exclusion as well as as a negative keyword list
5. Then evaluate the non-brand results on their own. That’s the number that tells you whether the campaign is working or not.

See our guide to Google Ads brand lists.

Mistake 5: Performance Max misconfigured (64 of the 84 accounts running it)

Performance Max is the campaign Google recommends most, and the one where we found the most problems. 64 of the 84 accounts that ran it had at least one.

The problems that came up most:

• Shopping only: in 19 accounts, Performance Max spend almost all its budget on Shopping Ads, despite having all assets enabled. One home and garden store spent $37,200 of $38,000 on Shopping, despites having built out asset groups with headlines, images and custom videos. This isn’t as much of a problem, as it indicates wasted efforts and wrong expectation.
• Spend leaking: in 10 accounts, Performance Max wasted spend to Gmail and Display. A printing company’s asset group spent $213 of almost $8,000 on Shopping, and most of the rest on Gmail and Display.
• Campaigns competing with each other. In 18 accounts, Performance Max and Shopping or Search chased the same searches. In another 18, several asset groups or campaigns targeted the same products.
• Weak asset groups and audience signals (36 accounts)

Why it happens: if you’re using Google’s guidance, pMax seems like a pretty simple campaign to set up. But to get it to perform well, it needs some additional configuring after the publishing.

How to fix it:

1. Open the channel performance report of your Performance Max campaign. How much goes to Shopping, Search, YouTube and Display? Is that what you want?
2. Check whether your Shopping or Search campaigns target the same products or searches, and whether any product sits in more than one asset group.
3. Depending on what you find, consider a feed-only campaign, Standard Shopping or separate campaigns.

More in our guide to Performance Max campaigns.

Mistake 6: Dynamic remarketing missing or misconfigured (50 of the 71 accounts checked)

Dynamic remarketing shows people the products they looked at on your site, these days this is usually as part of a pMax campaign. But that only works when the product IDs your site sends to Google match the IDs in your feed.

In 50 of the 71 accounts where we checked it, remarketing was switched off or the IDs didn’t match.

In one store, the match rate was 34%, where 80–95% is normal. In a lighting store, the tag didn’t fire at all, so Performance Max didn’t have the remarketing bonus it could have.

You can check the portion of Dynamic Remarketing in the channel performance report by looking at Display placements with Ads using product data.

Why it happens: the remarketing tag gets set up once and rarely checked again. When a theme or app update changes the tag, it stops firing without any warning. And even when the tag works, the site often reports variant IDs that don’t match the IDs in the feed.

How to fix it:

1. In Google Ads, open Tools > Data manager > click the three dots next to your Google Tag and click “View stats”
2. In that graph, add Match rate (Retail)
3. Check the match rate between your site’s product IDs and your Merchant Center feed. Is it above 80%?
4. If it’s low, compare a product ID on the site (including variants) with the same product in Merchant Center.
5. Re-check after every theme, app or platform change.

See our guide to dynamic remarketing.

Mistake 7: Unmanaged search terms and keywords (75 of 122 accounts)

Most accounts we audited spent money on searches that were never going to convert. 75 of the 122 accounts had a keyword or search term problem.

The problems that came up most:

• Unfocused ad groups. Too many unrelated keywords in one ad group (41 accounts)
• Missing negatives. Irrelevant or competitor searches that nobody excluded (34)
• Uncontrolled broad match (24)

A bakery’s Shopping campaign showed for searches like “food” and “cake” and for supermarket names. A wallet store ran 62 broad match keywords in one ad group on $30 a day, and only 8 of them got clicks.

Why it happens: Google suggests broad match and long keyword lists when you create a campaign. Do get good results from Search Ads, you ned to do your homework when it comes to campaign and ad group strcuture, as well as keyword and match type selection. Because once the campaign runs, the search terms report is easy to forget, and irrelevant searches pile up.

How to fix it:

1. Open the search terms report, set the last 30 days and sort by cost.
2. Look at the searches that spend without converting. Would you have picked them?
3. Add the irrelevant ones as negatives, and check that your negative keyword lists are attached to every campaign that needs them.
4. Look at your ad groups and match types. If one ad group holds many unrelated keywords, or broad match runs without enough conversion data, consider splitting or tightening them.

More in our guides to negative keywords and keyword match types.

Mistake 8: Product titles misaligned with search queries (39 of 122 accounts)

In Shopping and Performance Max you don’t choose keywords. Google matches searches to your product titles, so a title without the words people search for won’t show for those searches.

In 39 of the 122 accounts, the titles didn’t match how people search for the product.

The gaps that came up most:

• Missing brand. The brand people search for wasn’t in the title
• Missing use. The main use of the product wasn’t in the title
• Store text first. Store names or taglines took up the first characters

In a lighting store, buyers searched for brand, product type and model number together, and the titles had none of that structure.

Why it happens: Most product feeds pull product names straight from the store. In some cases, these names were written for shoppers already on the site, and don’t hold a lot of keywords power. Things like “Midnight Star Earrings”, instead of “large solid gold earrings”

How to fix it:

1. Open the search terms report for your Shopping and Performance Max campaigns, and note the words buyers use in searches that convert.
2. Compare them with the titles of your bestsellers. Are those words in the title, and near the start?
3. If not, rewrite the titles as brand, product type, the key attribute (size, material, use) and model, in the order people search. A supplemental source (feed) lets you do this without changing the titles in your store.

More in our guide to product title optimization.

Mistake 9: Budget fragmented across too many campaigns (29 of 122 accounts)

Smart bidding needs conversions to learn, and every extra campaign splits those conversions further. So a key decision in account structure is about how to balance specificity with volume.

In 29 of the 122 accounts, the budget was spread over more campaigns, countries or asset groups than the conversions could support.

A home and garden store had around a hundred Shopping campaigns, many getting five or six clicks a month. Beyond the top ten campaigns, each one got only a few hundred clicks a month.

Why it happens: Each new country, category or test gets its own campaign, because that feels like more control. Campaigns get added over the years and rarely merged back.

How to fix it:

1. Open the campaigns view, set the last 30 days and sort by conversions.
2. How many campaigns get fewer than about 30 conversions a month? Do they really need to be separate?
3. Consider merging them, and give each remaining campaign a daily budget of at least one target CPA.
4. If you advertise in several countries, check whether each one has enough volume to run on its own.

See the scaling section of our Google Ads for ecommerce guide.

Mistake 10: Auto-applied recommendations switched on (31 of the 88 accounts checked)

Auto-apply lets Google change your account without asking. It’s one of the nastiest features Google cooked up the last few years, and it can easily wreck a working account.

In 31 of the 88 accounts where we checked it, it was switched on.

In one outdoor apparel store, auto-apply had changed the target ROAS 19 times in one week, two weeks before the audit. The store wanted a ROAS of 400%. The target now sat at 150%.

Why it happens: Google offers auto-apply as a time saver, often during setup. Once it’s on, the changes happen in the background, and the change history is the only place they show up.

How to fix it:

1. Go to Recommendations and open the auto-apply settings. Which types are switched on?
2. Check your change history for recent changes made by Google. Are those changes you would have made?
3. Switch off the types you don’t want, and review recommendations yourself every few weeks.

More in our guide to Google Ads recommendations.

Other, less critical mistakes we found

Two problems came up often but were almost never the main issue:

• Weak ads. 48 accounts had weak responsive search ads: no testing, missing assets or old text ads. Ads were the first fix in only one audit.
• Low conversion rates on the site. 29 accounts had a landing page, trust or pricing problem. We rarely reviewed sites in depth, so this number is probably low.

Fix the ten above first. Better ads won’t help if Google is bidding on the wrong data.

What advertisers thought was wrong vs. what we found

One very interesting meta finding is that in 79 audits, the advertiser told us up front what they thought was wrong. In 50 of those, the real cause sat somewhere else.

What the advertiser said	What we found
“We’re getting click fraud”	Page views counted as conversions, and campaigns on Maximize Clicks (two separate audits)
“ROAS drops whenever we scale”	One all-products Performance Max campaign per country, no segmentation
“Performance is down while spend is up”	Dynamic remarketing matching 0% of products, and Performance Max spending almost only on Shopping
“Costs are up and sales are lagging”	Brand searches leaking into Performance Max, hiding weak non-brand results
“We’ve run out of keywords”	A target ROAS far above the business goal, capping growth
“We want more sales for the same spend”	Remarketing tags missing, and bidding that ignores product value
“Sales dropped since the takeover”	Phone and B2B orders never tracked, and titles missing the main search term

This is a big reason in-house accounts stall: whoever runs the account fixes the problem they can see. If you’re weighing running Google Ads in-house against handing it over, here’s how to decide.

What we recommended fixing first

Area of the first fix	Audits	Share
Structure (segmentation, Performance Max setup, brand split, tiering)	35	29%
Tracking	27	22%
Bidding	21	17%
Keywords and search terms	12	10%
Budget allocation	11	9%
Feed	6	5%
Other	10	8%

Fix tracking and structure before you touch bids or ads. Ads came up in 48 audits but were the first fix only once.

Rules of thumb that kept coming up

These are the rules our Google Ads experts repeated across audits, as general advice.

Rule	Audits
Smart bidding needs about 30 conversions per campaign per month	21
Separate brand from non-brand	19
Run brand search on manual CPC	18
Start a Performance Max review with the channel performance report	17
Remarketing product IDs must match Merchant Center (80–100% match rate)	15
Feed titles do the heavy lifting for Shopping and Performance Max	14
Only purchases as primary conversion, everything else secondary	9
Change target ROAS in steps of about 20%, not small daily tweaks	8

How we collected the data

Between May 2024 and September 2026, we analyzed 122 Google Ads account audits recorded by the Store Growers team. All ecommerce, covering many different niches, geographically spread over the UK, the US, Canada, Australia and Europe.

We transcribed every recorded audit and coded it against a fixed list of 46 issue types and 15 check areas. A second reviewer re-checked a random 11% of the audits against the original recordings. 96% of the coded issues held up, and the top priority our expert set matched in every audit we sampled.

Percentages show the share of audits in which our team identified a problem. We didn’t check every area in every audit, so the real numbers are probably higher. On the other hand, these stores asked us for an audit, so they likely had more problems than an average account.

Check your own account

Every mistake above can be checked in an afternoon. Our Google Ads audit checklist for ecommerce walks you through it, point by point.

Store Growers reviews and manages Google Ads for ecommerce stores spending roughly $10–50K a month. The first step is a free account review. Tell us what you think is wrong and give us read-only access, and we’ll show you where the account loses money. We’re not a fit for stores just getting started or for non-ecommerce businesses.

Most of these mistakes aren’t new. What we can’t tell yet is whether they’ll matter less as Google automates more of the account. If Performance Max and AI Max make more of the decisions, does a broken tracking setup do more damage, or less? We’ll know more when we compare next year’s audits with these.