How to Run a Facebook Ads Audit on a Shopify Store
A Facebook ads audit sounds like a job for an agency with a long checklist. For a Shopify store, most of it comes down to a few questions about money.
It is tempting to audit Facebook ads by sorting on ROAS and pausing whatever looks worst. That misses where the money leaks. A campaign can sit above break-even while some of its ads spend for weeks without a sale. Another looks weak only because Meta is still learning after Tuesday's edit.
Order matters as much as the checks themselves, so we follow one example store from export to ranked fixes.
What Does a Facebook Ads Audit Check?
A Facebook ads audit is a review of your Meta ad account against your own break-even numbers. You export 30 days of data and set your break-even ROAS. Then you grade 10 areas and rank the fixes by the spend each one touches.
The areas are tracking, structure, the retargeting split, budget spread, platform mix, creative depth, formats, fatigue, waste and ROAS. Grade each one Pass, Watch or Fix. Thresholds below are our working rules unless we name Meta as the source.
A free Facebook ads audit needs nothing more than that export and a spreadsheet. An Instagram ads audit runs the same way, from the same Meta Ads Manager account.
Export the Last 30 Days at the Ad Level
In Ads Manager, open the Ads tab, because an ad-level export still carries every campaign and ad set name. Set the date range to the last 30 days. Clear any Active filter too, so paused ads that spent still show.
Add spend, impressions, reach, frequency, link CTR, purchases, purchase value and purchase ROAS as columns. Then choose Breakdown, By time and Week, so frequency, CTR and ROAS show week by week. Export the table as a CSV file.
Set Your Break-Even ROAS First
Break-even ROAS is 1 divided by your gross margin after product cost, shipping and fees. Take a store with a 40% margin and $60 per order. Its break-even ROAS is 1 / 0.40 = 2.5, and below that, first orders lose money.
Divide the $60 order by 2.5 for a break-even cost per purchase of $24. Every check below leans on those two numbers, so pin your margin down with our break-even ROAS calculator.
Tracking Checks in a Meta Ads Audit
Check tracking first, since every later grade depends on it. Compare Meta's purchase count for the 30 days with your Shopify orders for the same dates. Some gap is normal, since Meta only counts sales it ties to an ad. That means a sale inside its attribution window, such as 7 days after a click.
A count far above your orders, or far below them when Meta drives most of your traffic, means fixing tracking first. Old pixel code left in a theme is one cause, because it puts more than one pixel on the store. If the Conversions API runs beside the pixel, deduplication has to match both copies of each sale, or Meta counts it twice.
Leave New and Edited Ads Out of the Grades
Results swing while an ad set is in the learning phase. It usually ends after about 50 results in the week after the last significant edit. Changing the audience, creative or optimization event counts, and so can a big budget change.
So anything launched or edited in the last 7 days gets a note, not a grade. Ads Manager's activity history shows who changed what and when.
Judge each campaign on its own objective too. An engagement or giveaway campaign is not built for purchases, so it never counts as waste.
Facebook Ads Audit Checklist for Structure and Budget
Use this checklist to review how your Facebook ad campaigns are organized, how your budget is distributed, and where you can make improvements.
Campaign Count and the Retargeting Split
Up to about $200 per day, one prospecting campaign plus one retargeting campaign is enough for most stores.
Myth: More campaigns give Meta more chances to find buyers.
Reality: Each extra campaign splits the budget and learns more slowly, while merging similar ad sets pools their learnings.
Give retargeting roughly 20 to 25% of the account's Meta spend. For the example store's $100 a day, that is $20 to $25. Under about $5 per day, too few new visitors flow in to keep it fed. A store spending $20 per day would have just $4 at a 20% share, so it should run prospecting only.
Campaign names do not always show what retargets. Once you define audience segments, Meta's report splits results into new, engaged and existing customers. Check too that retargeting excludes recent buyers, a step our guide to Facebook retargeting ads for Shopify sets out.
Ad Set Budgets That Are Too Thin to Read
An ad set that spends less in a week than one target cost per purchase cannot give a readable result. The example store's floor is $24 a week, so merge anything below it into a stronger ad set.
Meta's own bar sits far higher. At $24 per purchase, the roughly 50 results that usually end learning cost $1,200 a week per ad set.
The account generated 138 purchases in 30 days, averaging about 32 per week. That's below Meta's recommended 50 weekly results for a single ad set. Dividing the budget across five ad sets makes it even harder for each one to gather enough data. Meta recommends combining ad sets to help resolve the Learning limited status.
Meta Only or Meta Plus Google
Under about $35 per day in total, pick one platform, since Meta and Google together at that budget starve both. Above that line, the Google account needs its own pass, which our Google Ads audit checklist covers.
How to Check Creative Depth, Formats and Fatigue
Review your ad creatives to see if you have enough variety, use the right formats, and spot signs of ad fatigue that could hurt performance.
How Many Ads Each Ad Set Needs
Each prospecting ad set needs 3 to 6 ads with spend. One ad tests nothing, and ten or more leaves most of them without spend.
Product count changes the mix. With 7 or more products, run one dynamic catalog ad plus about 3 graphic ads. With 6 or fewer, run about 4 graphic ads and no catalog ad.
Feed and Vertical Versions of Every Image
Every image needs a square or 4:5 version for Feed and a 9:16 version for Stories and Reels. Meta's own image templates pair the same ratios with the same placements. Rebuild the design for the tall frame rather than padding the square with empty space.
Frequency and Falling CTR
Frequency is impressions divided by reach, the average number of times each person saw your ad. Flag a prospecting ad set when weekly frequency passes about 3 while link CTR falls. A 30-day figure says little about any single week, hence the weekly breakdown.
Meta also flags fatigue in the Delivery column. An ad shows Creative fatigue once its cost per result reaches twice that of your past ads. The fix is a materially different ad, as our guide to creative fatigue on Meta explains.
How to Find Wasted Ad Spend and Weak ROAS
An ad counts as wasted spend when it spends twice your target cost per purchase or more with zero purchases. At $24 per purchase, that line is $48. In the example store's account, five ads crossed it and spent $310 between them without a sale.
Then set each campaign's ROAS against break-even. A steady campaign above 1.3 times break-even, here 3.25, is a scale candidate. One below 2.5 for the 30 days and the last 2 weeks is a fix.
Read campaigns one at a time, because the account total can hide a loss. Say the store's prospecting spent $1,950 at a 2.1 ROAS and retargeting $1,050 at 4.0. The account shows 2.77, above break-even, while prospecting loses money on its first orders.
How to Rank the Fixes From a Facebook Ad Account Audit
Rank every Fix and Watch by the spend at stake, meaning the 30-day spend on the campaigns or ads it touches. That stops a $130 problem from jumping ahead of the store's $1,950 one. Write each fix with its number, its steps in Ads Manager and the metric that should move.
Do not add budget to a campaign below break-even until it is fixed. For the example store, shift money from retargeting only once prospecting clears 2.5.
A Facebook Ads Audit Template You Can Copy
The scorecard below uses the example store's numbers. Copy it into a sheet and it doubles as a Meta ads audit checklist each month.
| Area | Grade | Example Finding | Spend at Stake |
|---|---|---|---|
| ROAS vs break-even | Fix | Prospecting at 2.1 vs 2.5, also below in the last 2 weeks | $1,950 |
| Creative depth | Fix | Main prospecting ad set runs 1 ad with spend | $1,110 |
| Retargeting split | Watch | Retargeting takes 35% of sales spend | $1,050 |
| Wasted ad spend | Fix | 5 ads, $310 spent, 0 purchases | $310 |
| Budget spread | Watch | 2 ad sets spend under $24 a week | $130 |
| Tracking | Pass | Meta shows 138 purchases, Shopify 160 orders | None |
Pixel and Conversions API health, audience settings, placements and landing pages never show in an export. A Facebook ads audit tool that only reads the export cannot grade them, so check those by hand.
How Agency AI Runs These Checks for You
Our free Meta Ads Account Audit skill runs the same checks in Claude or ChatGPT in about 5 minutes. It reads the export you upload inside your own AI chat, never logs in to your ad account and changes nothing. Nothing is sent to Agency AI.
Agency AI, our Shopify app for Meta and Google ads, keeps checking after that. Our 90+ rule engine reviews the last 7 days of account data and measures each recommendation against your break-even ROAS. Nothing changes until you click Apply.
When the audit finds thin creative, Create Ad builds a Meta ad in about 2 to 3 minutes. Pick Dynamic, Image or Video, choose New Customers or Remarketing, and use Generate with AI across 12 copywriting angles.
Through our MCP connector, Claude or ChatGPT can read live Meta, Google and Shopify data. Any write it proposes runs only after you confirm it.
Final Thoughts
A Facebook ads audit works when the checks run in order. Set break-even first, confirm tracking, then grade structure, creative and spend against that number.
Start with the fix that has the most spend at stake. Between audits, the weekly approval loop in our plan to automate Facebook ads for Shopify catches the same problems sooner.
Frequently Asked Questions
Sources
Meta facts come from Meta's Business Help Center and Transparency Center, and the pixel note from Shopify's Help Center. All were checked October 28, 2026.
Our working rules and product features come from Agency AI, and the example store is hypothetical.
Free to install. Launch campaigns and pay nothing for your first 30 days.
