The performance marketing audit checklist we run on every new account

Why audit before you optimise

Almost every ad account we open has leaks — usually the same handful, in the same places. Optimising on top of them is how brands spend six months “scaling” numbers that were never true. An audit is not a report; it is a map of where money is quietly going missing, ordered by how much each fix is worth. Run this checklist quarterly, after any big platform change, and always before you increase budgets — scaling multiplies leaks along with results.

The order below matters as much as the items. Tracking comes first because every later judgement depends on it; measurement comes last because it summarises everything above it. Skip around and you'll optimise creative with corrupted data, or restructure campaigns whose only real problem was a slow landing page.

1. Tracking and attribution: is the data even true?

Nothing else in the audit matters if the numbers feeding it are wrong, so start here and be pedantic — budget an hour for this section alone; it usually pays for the rest of the audit:

• Pixel and Conversions API both firing, deduplicated by event ID — double-counting here inflates every downstream decision.
• Platform purchase values reconcile with your order backend within a sensible tolerance. If Meta says ₹10L and Shopify says ₹7L, find out why before touching budgets.
• COD and RTO adjustment: platform ROAS counts orders that never get delivered. For most Indian D2C brands this is the single biggest fiction in the account — the COD returns and real ROAS math shows how to correct for it.
• UTM discipline: consistent parameters on every ad, so blended reporting is possible at all.
• First-party capture: are you storing first-touch and last-touch attribution on your own leads, or renting all memory from the platforms?

2. Account structure and budget

Once the data is trustworthy, check whether the account's architecture lets the algorithm do its job:

• Campaign sprawl: dozens of overlapping ad sets fragment learning and bid against each other. Fewer, fatter campaigns almost always win in 2026.
• Advantage+ vs manual split: tested deliberately, or inherited from whoever set the account up?
• Budget by funnel stage: how much is genuinely acquiring new customers vs quietly retargeting people who would have bought anyway? Check audience exclusions.
• Zombie spend: paused-in-spirit, live-in-practice campaigns that nobody has looked at in months.
• Frequency: if your core audience is seeing the same ad nine times a week, you are paying for annoyance.

3. Creative: the real lever

Count the number of genuinely distinct hooks live in the account — not resizes, not colour swaps, distinct persuasive angles. Most “fatigued” accounts are running two ideas in fourteen formats. Then check the testing cadence: is there a weekly rhythm of new concepts with a written hypothesis and a recorded result, or does new creative appear when someone remembers? An account that ships creative without accumulating documented learnings isn't testing — it's churning. Finally, look at the mix: UGC, static and video each earn their place by performance, not by fashion.

A useful exercise: screenshot the top ten spending ads and ask what belief each one is trying to change. If three screenshots answer the same way, the account has one ad and nine costumes. And measure your own fatigue window empirically — hooks age out on a cadence specific to your account, not to industry chatter. Creative fixes are also the cheapest to test: a new hook costs a fraction of a restructure and reads out in days.

4. Funnel and landing pages

Clicks are only as valuable as the page they land on. Audit load speed on a mid-range phone over 4G, offer clarity above the fold, whether the product page answers the top three objections, and how many taps stand between intent and payment. Then check the AOV levers — bundles, thresholds, post-purchase offers — because average order value is the quietest way to fix CAC math. In our experience the fastest CAC wins are usually on-site, not in the ad account; the how to reduce CAC playbook goes deeper on this. Time the journey yourself on your own phone this week — founders are reliably the last people to experience their own checkout. Small numbers compound here: a one-second speed gain and one removed checkout field can move conversion more than a month of bid tweaks.

5. Measurement: judge the account like a CFO

The last section decides whether everything above matters. Steer by blended MER and contribution margin after shipping, returns and discounts — not by platform ROAS alone. Check CAC against repeat rate: a ₹700 CAC is fine for a brand with 40% repeat and fatal for one without it. Compare against category reality, not aspiration — our ROAS benchmarks for D2C food and beverage give the honest ranges. The output should fit on one page: five numbers, refreshed weekly, that everyone from founder to media buyer agrees are true. If two of those five can't be produced inside a day, that inability is itself a top-three audit finding. Build the scorecard before arguing about any single number — shared definitions end most founder-agency fights before they start.

Run it yourself — or have us run it on your account

With account access and a focused afternoon, a founder can run this checklist alone — and should, at least once. If you'd rather have the team that has managed ₹150 Cr+ in spend across 160+ D2C brands do it, that is literally what our free Growth Audit is: this checklist, applied to your account, returned as a prioritised fix list. We've written up exactly what to expect so there are no surprises — and no obligation on the other side of it. Either way, insist the output is a ranked fix list with expected impact: an audit that ends in observations rather than an ordered to-do list changes nothing.