Google Ads
How to Audit Your Google Ads Account: The eCommerce Checklist
Most eCommerce stores have never had a proper Google Ads audit. Here's what one actually covers.
There's a difference between an account that has been optimised and one that has been audited. Optimisation is ongoing — bid adjustments, keyword additions, copy tests, budget reallocation. An audit is a structural review: is the account configured correctly, and is it actually capable of producing reliable results?
The problem is that optimisation on a flawed foundation has a ceiling. You can tune a campaign's bidding for months, but if the conversion tracking is misconfigured or the campaign structure is fragmenting the algorithm's data, performance will plateau regardless. An audit surfaces those foundations problems before anyone wastes time optimising on top of them.
This checklist covers the 10 areas that matter most for an eCommerce account running $5k-$50k per month on Google Ads.
An audit isn't a 90-day performance report. Performance data tells you what happened. An audit examines whether those results are trustworthy, and what structural factors are driving them.
In practice, this means I start from scratch. The existing structure, the previous decisions, the workarounds that accumulated over years — I examine each on their own merits rather than inheriting the assumptions baked into them. An account that has run for two or three years has years of layered decisions on top of each other: some good, some outdated, some quietly wrong. Working through those layers systematically is what separates an audit from a performance review.
What an audit isn't: a long list of 40 things that could theoretically be improved, with no prioritisation and no distinction between what matters and what's cosmetic. That's not an audit — it's a distraction. A useful audit identifies the handful of issues that, if fixed, would produce a measurable change in performance, and ranks them in the order they should be addressed.
This is always the first area to examine, and it's consistently the most broken. The question isn't just whether conversion tracking is set up — it's whether it's measuring what actually matters, without double-counting, with the right attribution window, and with transaction value attached.
Common issues I find across accounts:
If conversion tracking is wrong, every number in the account is unreliable. ROAS is fiction. Smart Bidding is optimising for the wrong signal. Fixing bidding strategy, keywords, or copy before fixing tracking is working in the wrong order.
Campaign structure determines how Smart Bidding receives and uses data. Fragment data too far and the algorithm can't learn. Consolidate poorly and you lose the ability to allocate budget where it matters most.
The key diagnostic: does each campaign have enough conversions per month for Smart Bidding to operate reliably? The widely-cited minimum is 30-50 conversions per campaign per month before the algorithm moves out of the learning phase. The higher end applies if you're running target ROAS — the algorithm needs more signal to reliably hit a specific return constraint versus simply maximising volume.
Example: 60 total monthly conversions spread across different campaign structures.
For most eCommerce accounts spending $5k-$50k per month, too many campaigns is the more common problem. Six Shopping campaigns each generating 8-10 conversions per month isn't a structured account — it's fragmented data that the algorithm can't reliably work with.
Pull the 30-day search term report for your Search campaigns and look at it honestly. How many of those queries are from people genuinely in-market for what you sell?
In almost every account I audit at this spend level, a meaningful share of Search spend is going to queries that will never convert: competitor brand names, informational searches from people researching rather than buying, and queries that look like your keywords but belong to a different audience entirely.
A search term audit answers three questions: which terms should be added as negative keywords, which should be added as tightly-matched positives, and which are already working and worth bidding more aggressively on. Most accounts have done some negative keyword work, but haven't done it systematically.
Performance Max campaigns get credit for results they may not have generated independently. They bid across all Google inventory — Search, Shopping, Display, YouTube, Gmail, and Discover — and they will take credit for converting users who were already heading your way.
During an audit, the key questions are:
Useful check: Run brand vs. non-brand conversion breakdowns separately. If branded conversions are doing the heavy lifting in Performance Max, the ROAS figure is flattering the campaign's actual contribution to growth.
For eCommerce accounts, the product feed is the foundation of Shopping and Performance Max performance. Disapproved products don't serve. Products with thin titles, missing attributes, or incorrect pricing get fewer impressions — quietly, without triggering any alerts.
A feed audit covers: disapproval rate in Merchant Center, attribute completeness across titles, descriptions, images, GTIN, and MPN, and whether product titles reflect how buyers search rather than how the business describes products internally. "Men's Short-Sleeve UV Fishing Shirt" will outperform "Performance Shirt - SKU-00492" in almost every account.
This is where many accounts carry silent problems. The bidding strategy in the account doesn't match the economics of the business — and no one has checked whether the two are aligned.
| Signal | What It Means | Action |
|---|---|---|
| tROAS set above actual ROAS | Smart Bidding is targeting a return it has never achieved — it will under-bid and lose auctions | Lower tROAS to 80-90% of actual achieved ROAS to give headroom to scale |
| Manual CPC on high-volume account | You're managing bids without the data advantage Smart Bidding has at auction level | Migrate to tROAS or Maximise Conversion Value if the account has sufficient conversion history |
| tCPA on variable-AOV products | Optimising for cost-per-conversion when order value varies significantly produces misleading signals | Switch to tROAS with revenue values passing correctly |
| High IS Lost (Rank) + high tROAS | The ROAS target is making Smart Bidding too conservative — it's losing auctions it should win | Lower tROAS to loosen the constraint; raising tROAS will make this worse |
Budget allocation is where the strategic decisions live, and in most accounts it hasn't been reviewed deliberately in months. Budgets get set, performance changes, and the allocation drifts out of sync with what the data is showing.
I map two things: which campaigns are budget-constrained (high Impression Share Lost to Budget with acceptable ROAS) and which campaigns have budget available but aren't performing to justify it.
A campaign with 40% Impression Share Lost to Budget and a strong ROAS is giving you a clear signal: it can capture more high-quality auctions if the constraint is removed. That's a different decision than a campaign with low IS Lost to Budget but weak ROAS — which may need restructuring before more budget helps at all.
For Search campaigns: do you have Responsive Search Ads with enough headline and description variety for Google's system to test? Are assets set up across sitelinks, callouts, structured snippets, and seller ratings where eligible?
Asset quality affects Quality Score, which affects both ad rank and cost-per-click. A Responsive Search Ad with a low Ad Strength rating isn't just an optimisation opportunity — it's actively increasing what you pay per click relative to a competitor with stronger assets in the same auction.
For Performance Max: are you supplying enough creative assets across images, headlines, descriptions, and video for the campaign to test across all placements? Accounts that rely on auto-generated assets from their website are usually getting a worse result than accounts supplying curated materials.
The cheapest click is wasted on a landing page that doesn't convert. An audit looks at whether the destination each ad sends traffic to matches the specific intent behind the search query.
The common failure: a Search campaign with tightly targeted keywords sending all traffic to a category page or homepage, rather than a landing page that speaks directly to what the searcher is looking for. The further the distance between what the ad promises and what the landing page delivers, the worse the conversion rate — regardless of how well the campaign is configured.
Quick test: Take your five highest-spend keywords and open the landing page each one sends traffic to. Would someone who searched that exact query immediately recognise they're in the right place? If not, that's a landing page alignment problem, and fixing it will likely produce a bigger performance lift than any bidding adjustment. The fix is usually a dedicated landing page built around the specific search intent — not a tighter keyword list.
The Auction Insights report tells you which competitors are appearing in the same auctions, how often you're both showing, and where you're winning or losing position. It doesn't reveal their budgets or bidding strategies, but it tells you who you're consistently competing with — which matters when interpreting impression share data.
One useful check: are your top competitors running Shopping campaigns, Search campaigns, or both? If a competitor consistently outranks you in Shopping but doesn't appear in Search (or vice versa), that tells you where the head-to-head competition is concentrated — and where you might find less contested inventory.
Low impression share is not always an account problem. If well-funded competitors are dominating the auctions you're targeting, the question isn't just "how do I increase impression share?" but "is it worth competing head-to-head on these terms, or is there a segment of the market where the competitive pressure is lower?"
A good audit produces a prioritised action list, not a document of everything that could theoretically be improved. The prioritisation follows a principle: don't optimise downstream while something upstream is broken.
The order matters. Improving ad copy when conversion tracking is counting the wrong events won't produce useful data. Adding budget when the campaign structure is fragmenting the algorithm's signal won't produce the return you're expecting. Fix the foundations first.
Running through this checklist yourself will surface the obvious problems. But there's a category of issue that only surfaces when I look at an account I didn't build — the kind of misalignment between bidding strategy and business economics that looks fine on the surface but is quietly suppressing what the account could be doing.
In my experience, the accounts that benefit most from an outside set of eyes share a few things in common:
If two or more of those describe your account, that's the profile I work with most.
A second opinion on the foundations is worth more than most ongoing optimisation work. Getting the structure right once changes what's possible at every level below it.
Want me to run this for you? The Google Ads Gap Analysis covers all 10 areas — I work through the account, identify what's holding it back, and deliver a prioritised action plan with a debrief call included. From $997. Book here →
Common questions about Google Ads audits for eCommerce accounts.
A thorough Google Ads audit for an eCommerce account typically takes 4-8 hours of analysis time, depending on account complexity and how long it has been running. Accounts with years of layered campaign structures, multiple conversion actions, and broad product catalogs take longer to examine properly.
A one-hour review call is not an audit — it's an overview. The deliverable from a real audit should be a written prioritised action plan, not a list of verbal observations.
A health check is typically a high-level review of surface metrics: click-through rate, conversion rate, cost-per-click, account-level ROAS. It tells you how the account is performing against broad benchmarks.
An audit goes deeper. It examines whether conversion tracking is reliable, whether the campaign architecture is appropriate for the account's volume, and whether the bidding strategy aligns with the business economics. Most accounts need a full audit first, and regular health checks after the foundations are confirmed sound.
A full structural audit makes sense when the account changes hands, after a significant restructure, or when performance has been flat or declining for more than two quarters despite ongoing optimisation work.
Accounts running without any structural review for more than 12 months often have problems that have been optimised around rather than fixed. Ongoing monthly performance reviews are a different discipline — they check whether recent changes are working, not whether the foundations are sound.
Start by identifying the primary conversion action you want Smart Bidding to optimise toward — usually a purchase with transaction value. Set this as your only primary conversion in Google Ads and demote secondary actions (add-to-cart, page views) so they don't influence bidding.
Verify the tag is firing correctly using Google Tag Assistant, then check for duplicates — if you're importing from GA4 and also running a native Google Ads purchase tag, you may be counting the same event twice. Fix the foundation before adjusting bids or budgets.
Yes, and the checklist above gives you a solid framework to work through. The limitation of a self-audit is familiarity bias — when you've been inside the same account for a long time, it's easy to normalise problems that a fresh perspective would catch immediately.
When I look at an account I didn't build, the issues that surface most often are the ones the account owner has stopped seeing: bidding strategies misaligned with actual ROAS, fragmented campaign structures below the data threshold Smart Bidding needs, or conversion tracking setups that work technically but produce misleading signals. The checklist is a good starting point regardless of who runs it.
The Google Ads Gap Analysis covers all 10 areas — I work through the account, identify what's holding it back, and deliver a prioritised action plan. From $997, includes a debrief call.
Book a Gap Analysis →