Most Meta ad accounts don't fail because of one big mistake. They slowly leak money through a dozen small ones: a broken event, overlapping audiences, tired creative, a budget rule nobody remembers setting.
This is the checklist I use when I audit a Facebook and Instagram ads account. It's split into six areas and ordered the way I work through them, because there's no point judging creative if the tracking is wrong.
How to use this checklist
- Look at the last 30 days, compared with the 30 days before.
- Mark each point as OK, Fix or Not sure.
- Fix tracking problems first, then structure, then creative.
1. Tracking and data (points 1–6)
- The right Pixel (dataset) is connected. Check that your ad account uses the same Pixel your website sends events to. Old or duplicate Pixels are surprisingly common.
- Key events are firing. In Events Manager, confirm ViewContent, AddToCart, InitiateCheckout and Purchase (or Lead) are all receiving events.
- The Conversions API is set up. Events should arrive from both the browser and the server. If you're on Shopify, see my guide on setting up the Conversions API on Shopify.
- Deduplication works. Browser and server events need a matching event ID, or purchases may be counted twice.
- Event match quality is healthy. Higher match quality means Meta can connect more conversions to the people who saw your ads.
- Reported results roughly match reality. Compare Meta purchases or leads with your store or CRM for a full week. They won't match exactly, but big unexplained gaps mean something is broken.
2. Account and campaign structure (points 7–11)
- Campaigns have a clear job. You should be able to say what each campaign is for: prospecting, retargeting, testing or scaling.
- There aren't too many small ad sets. Splitting a small budget across many ad sets slows learning. Consolidate where you can.
- Ad sets aren't stuck in "Learning limited". If many are, they likely aren't getting enough conversions. Combine them or optimise for a higher-volume event.
- Naming is consistent. A simple naming convention (objective, audience, date) makes reporting and handovers far easier.
- The optimisation event matches the goal. An e-commerce brand optimising for link clicks or landing page views will usually get cheap traffic, not sales.
3. Audiences (points 12–15)
- Prospecting and retargeting are separated, or at least reported separately, so you know where new customers come from.
- Audiences don't overlap heavily. Several ad sets chasing the same people compete against each other and push up costs.
- Recent purchasers or leads are excluded from prospecting where it makes sense, so you don't pay to reach existing customers.
- Broad and Advantage+ options have been tested. On many accounts today, broad targeting with strong creative performs as well as or better than narrow interests.
4. Creative (points 16–20)
- There's enough creative variety: different hooks, angles and formats (video, static, carousel, UGC).
- Top ads aren't fatigued. Watch rising frequency together with falling CTR on your best ads.
- New creative is tested regularly, not only after performance drops. Every one to two weeks is a good rhythm for most e-commerce accounts.
- Ads are built for the placement. Vertical 9:16 for Stories and Reels, 4:5 for feeds, with important text away from the edges.
- The landing page matches the ad. If the ad promises a specific product or offer, the click should land there, not on the homepage.
5. Budget and bidding (points 21–23)
- You know your break-even ROAS or target CPA. Divide 1 by your gross margin to get break-even ROAS. Without a target, "good" and "bad" are just opinions.
- Budget changes are gradual. Big jumps can restart learning and make results swing. Many advertisers scale in steps of around 20% every few days.
- Spend goes to what works. Check the share of budget going to ad sets that are below target. Money often stays stuck in old campaigns out of habit.
6. Reporting (points 24–25)
- The attribution setting is consistent across reports, and you compare full weeks, so delayed conversions don't make recent days look worse.
- There's a weekly review with actions, not just numbers: what changed, why, and what you'll do next week.
Quick summary table
| Area | Points | Most common problem I find |
|---|---|---|
| Tracking | 1–6 | Conversions API missing or not deduplicated |
| Structure | 7–11 | Too many small ad sets stuck in learning |
| Audiences | 12–15 | Overlapping ad sets competing with each other |
| Creative | 16–20 | Same few ads running for months |
| Budget | 21–23 | No break-even ROAS or target CPA |
| Reporting | 24–25 | Reports without clear next actions |
What to fix first
Work top to bottom. Tracking problems affect every other number in the account, so fix those before changing campaigns. Then simplify structure, then refresh creative. If ROAS has already dropped, my article on why Meta ads ROAS drops goes deeper into causes and fixes.
Want me to run this audit for you?
I'll go through this checklist on your account and send you the three changes most likely to improve results. It's free and takes 15 minutes on a call. You can also read more about my Meta ads service.
Frequently asked questions
How often should I audit my Meta ads account?
A light review every week and a full audit every quarter works for most accounts. Also audit after big changes, like a new website, a new Pixel or a major budget increase.
What should I check first in a Facebook ads audit?
Tracking. If purchase or lead events are missing, duplicated or connected to the wrong Pixel, every other metric in the account is unreliable.
Can I audit my own Meta ads account?
Yes. This checklist covers the main areas. An outside audit helps because a fresh pair of eyes often spots problems the account owner has stopped noticing.
Get a free 15-minute ad audit
I'm Ali Raza, a performance marketer with 6+ years and $10M+ in Meta, Google and TikTok ad spend managed. I'll review your account and show you the biggest opportunities.