Most "free Google Ads audits" are a sales call with a screen-share. Someone scrolls your account for twenty minutes, points at your 4.2× ROAS, says they can get it to 6×, and asks for your card. I've sat through a few of those pretending to be a prospect. Not one of them opened the search terms report — which is the first place the money is actually hiding.
An ecommerce Google Ads audit isn't a vibe check on your ROAS. It's a teardown of where your reported numbers and your bank account stop agreeing. In 8 out of 10 Shopify accounts I open, that gap is 20–40% of ad spend. Not bidding mistakes. Structural waste that the dashboard is built to hide.
This is the audit I actually run. You can run most of it yourself this afternoon.
What an ecommerce Google Ads audit checks that a generic PPC audit doesn't
A generic PPC audit looks at the things every account has: quality score, ad copy, negative keywords, bid strategy. Useful, but it treats a Shopify store like a lead-gen account, and the leaks in ecommerce live somewhere else entirely.
An ecommerce-specific audit goes after the parts that touch your P&L:
- The product feed and Merchant Center — because most of your spend runs through Shopping and Performance Max, and a feed problem is a revenue problem, not a "tagging" problem.
- Attribution against real Shopify revenue — because platform ROAS double-counts, and you need to know what hit the bank, not what Google claimed.
- Contribution margin — because COGS, shipping, returns, and discounts decide whether a "profitable" campaign is actually losing you money.
If your audit didn't mention your feed, your margins, or the gap between Google-reported revenue and Shopify-reported revenue, it wasn't an ecommerce audit. It was a checklist someone ran on autopilot.
The four leaks I find in 8 out of 10 accounts
These show up everywhere, across niches, across spend levels. Find them in your own account before you spend another dollar scaling.
1. Branded search inflation
Pull your search terms report and segment it: queries that contain your brand name versus queries that don't. Then look at how much spend and revenue sit on the branded side.
Here's the problem. Someone typing your brand name into Google was going to buy from you anyway. They saw you on Meta, heard you on a podcast, got an email. Google Ads catches them at the bottom of the funnel and claims the sale. Strip those queries out and your blended ROAS collapses — I routinely see accounts reporting 4.2× drop to 0.9× the moment branded demand comes out.
That's not a reason to panic. It's a reason to stop reporting one number. Isolate branded search into its own campaign so you can see it. Then make an honest call about how much of it you should even be paying for, depending on how aggressively competitors are bidding on your name.
What it usually costs: $3,000–6,000 a month of "revenue" that was already yours.
2. PMax cannibalization
Performance Max is a black box, and left alone it will do the laziest profitable thing available: bid on your own brand and your own retargeting pools, then report a beautiful ROAS while doing almost no new-customer acquisition.
Check two things. First, is a brand exclusion list applied to your PMax campaigns? If you can't confirm it is, assume it isn't — it's off by default and Google won't volunteer it. Second, look at your new-customer rate inside PMax. If the campaign is "crushing it" on ROAS but your blended CAC isn't moving and new-customer acquisition is flat, PMax is eating demand you already created and taking the credit.
The fix is unglamorous: apply the brand exclusion list, set the new-customer-acquisition goal with a real value attached, and stop letting one campaign grade its own homework.
What it usually costs: another $3,000–4,000 a month of misattributed spend, plus the opportunity cost of prospecting that never happens.
3. Retargeting that counts people already in the cart
If your retargeting is showing 8× ROAS, that number is mostly fiction. Most of those buyers had your product in their cart before they ever saw the ad. You paid to remind someone to finish checking out, then booked the entire purchase as ad-driven revenue.
Two checks. What percentage of total spend is going to retargeting and audience-targeted campaigns? And are purchaser exclusions applied everywhere, so you're not paying to advertise to people who already bought? If retargeting is more than 10–15% of spend, you're over-indexed on the cheapest-looking, least incremental traffic in the account.
The honest test is a holdout: stop showing retargeting ads to a slice of your audience and see whether they convert anyway. They usually do. Cap retargeting, exclude purchasers, and judge it on incrementality instead of last-click.
4. ROAS that pretends margins don't exist
This is the one that quietly kills brands while every dashboard stays green.
ROAS is revenue over ad spend. It knows nothing about your COGS, your shipping, your return rate, or the discount code attached to half your orders. A 4× ROAS on a product with a 30% gross margin and a 12% return rate is a very different business than a 4× ROAS on a 70%-margin product — but Google reports them identically, and you scale them identically, and one of them is bleeding.
Run the actual number. Contribution margin after ad spend — call it CM2 — is roughly your revenue minus COGS, minus shipping and fulfillment, minus transaction fees, minus ad spend. Compare CM2 to the story your ROAS is telling. When they disagree, CM2 wins, every time. Then start feeding margin-adjusted conversion values back into the account so Google optimizes toward profit instead of top-line revenue.
If you take one thing from this audit, take this: you cannot scale a number that doesn't include your costs. You can only scale your way into a hole faster.
The rest of the teardown: structure, feed, and tracking
The four leaks are where the dollars are. These three layers are where the leaks come back if you don't fix the plumbing underneath.
Campaign structure. Every campaign should be answerable to one question: what is its job? Branded search isolated. Non-brand search doing the high-margin intent capture. Shopping and PMax with brand exclusions on and a new-customer goal. Prospecting on Meta or non-brand. Retargeting capped and excluding purchasers. If two campaigns are competing for the same query, or you can't say in one sentence what a campaign is for, that's the restructure.
Feed and Merchant Center. This is the most ignored, highest-leverage part of an ecommerce account. Check for disapproved products first — I've opened accounts with 200+ SKUs silently disapproved for months, the founder had no idea, and reapproving them lifted Shopping revenue double digits in a month. Then look at titles (the first 30 characters do the work), GTIN/MPN accuracy, custom labels for tiering products by margin or ROAS, and out-of-stock suppression so you're not paying to advertise things you can't ship.
Tracking and attribution. If your tracking is wrong, every decision above is built on sand. Confirm enhanced conversions for Google and a server-side setup with a healthy match rate. Reconcile Google-reported revenue against Shopify-reported revenue and write down the gap — that gap is the entire reason this audit exists. Decide on one source of truth for CAC and payback, and it should be the one that ties to your bank, not the one inside Ads Manager.
The ecommerce Google Ads audit checklist
Run this top to bottom. If you can't answer a line with evidence from the account, that line is your next job.
Branded vs non-brand
- Search terms segmented into branded and non-branded spend and revenue
- Branded search isolated into its own campaign
- Non-brand ROAS calculated separately from blended ROAS
Performance Max and Shopping 4. Brand exclusion list confirmed applied to PMax 5. New-customer-acquisition goal set with a value 6. New-customer rate reviewed, not just ROAS 7. Shopping feed strategy separated from brand harvesting
Retargeting 8. Retargeting capped at 10–15% of total spend 9. Purchaser exclusions applied across all campaigns 10. Incrementality sense-checked with a holdout
Margin and measurement 11. COGS, shipping, returns, and discounts known per product 12. CM2 calculated and compared to reported ROAS 13. Margin-adjusted conversion values fed back into bidding
Structure 14. Every campaign has one defined job 15. No two campaigns competing for the same query 16. Budgets allocated against margin, not platform ROAS
Feed and Merchant Center 17. Zero disapproved SKUs across active campaigns 18. Product titles optimized in the first 30 characters 19. GTIN/MPN accurate; custom labels for ROAS/margin tiering 20. Out-of-stock products suppressed
Tracking 21. Enhanced conversions and server-side tracking live with a healthy match rate 22. Google-reported revenue reconciled against Shopify, gap documented
When to run this yourself, and when to stop guessing
If you're under roughly $75K a month in ad spend, run this audit yourself. The leaks are findable, the fixes are documented, and you don't have enough conversion volume yet for a paid engagement to compound. Save the money, do the work, ship the fixes.
Above that, the math changes. At $75K+/month, a 30% leak is $20K+ a month walking out the door, and the constraint isn't knowing what's wrong — it's the hours and the second set of expert eyes to find the version of these leaks that's specific to your account. That's what the $499 audit is for: 90 minutes screen-sharing your Google Ads and your Shopify, hard numbers on exactly where your spend is going, and a written plan you keep and run with me, your current agency, or in-house. No retainer attached to it.
Either way, the order of operations is the same. Find the leaks first. Scaling a leaking account just means losing money at volume.
FAQ
How do I know if my ROAS is inflated by customers who would have converted anyway? Segment your search terms into branded and non-branded, and separate retargeting from prospecting. Branded search and retargeting both catch people who already intended to buy. Strip them out, look at your non-brand prospecting ROAS on its own, and run a holdout test on retargeting. The honest number is almost always lower than the blended one — that difference is the inflation.
What's the difference between a Google Ads audit and an ecommerce Google Ads audit? A generic audit checks bids, copy, quality score, and negatives. An ecommerce audit adds the things that decide your profit: the product feed and Merchant Center health, Shopping and Performance Max structure, contribution margin, and the gap between Google-reported revenue and real Shopify revenue. For a store, those ecommerce-specific layers are where the actual money leaks.
How much does an ecommerce Google Ads audit cost? Anywhere from free to four figures, and the price tells you what you're getting. "Free" audits are usually sales calls. A paid audit (mine is $499 per platform) is a working session over your live account and Shopify with a written plan you keep. If you're a smaller account, the checklist above lets you run a real one for nothing but your time.
Can I audit my own Shopify Google Ads account? Yes. The four leaks above are findable from your search terms report, your PMax brand-exclusion settings, your retargeting spend share, and a contribution-margin calculation. The checklist in this post is the same framework I use. The paid version mostly buys you speed and a second expert read at higher spend levels.
How long does a proper audit take? Finding the leaks: an afternoon if you know where to look. Fixing them: the restructure happens fast, but the profit recovery compounds over 4–12 months. Anyone promising a transformed account in 30 days is selling you a screenshot, not a result.
Run the four-leak check on your own account first. If you want a second set of eyes on a $75K+/month account, book the $499 Google Ads audit — live over your account and Shopify, written plan, no retainer.
— Madhukar S.V., Founder, ScaleMarketer
