How Foxwell Members Audit Ads in Real Time (and Top 10 Audit Questions You Can Ask Yourself)
Why It Matters
Forget generic 40-item checklists. The audits that actually move accounts start by picking one of four jobs first: struggling, optimizing, second opinion, or numbers-don't-add-up, since a struggling-account audit and an optimization audit look at the same screens and reach opposite conclusions.
Go wide before going deep: pull 90 days as a line graph to spot where things turned, then drop to daily data, since outliers (a single fake $100k order) can quietly skew an entire period.
Put spend next to total revenue, new customer revenue, and MER. If Meta looks like a hero but the P&L disagrees, the culprit is almost always audience mix or over-reliance on view-through attribution.
Structure problems are math, not vibes: calculate how many conversions per day each ad set needs to exit the learning phase, then check whether the account is fragmented across too many campaigns to ever get there.
Creative diagnosis means sorting every active ad by format, angle, and awareness level, then watching spend velocity (not ROAS) to see what Meta is actually feeding, since that reveals what the algorithm believes before conversion data catches up.
The sharpest insight: an underspending ad set isn't a budget problem, it's Meta telling you where the market really is. Walk the cap up and watch.
The real diagnosis rarely comes from the framework alone. It comes from someone else asking "wait, why is that still on?", since account owners talk themselves out of their own findings within minutes of spotting them.
Includes 10 concrete audit questions readers can run solo in 30 minutes before they need anyone else's help.
Who This Is For
This is essential reading for Foxwell Founders members managing ad accounts at any spend level, whether you're running your own brand or auditing client accounts as an agency owner. If you've ever built a beautiful scaling plan for an account that turned out to have a pricing or fulfillment problem underneath, this reframes what an audit is actually for. It's also a direct pitch for the Live Meta Ad Account Reviews happening inside Founders every other Tuesday, where members get the "someone else asking the right follow-up question" experience these audits depend on.
People send me ad account audits pretty often. Most of them are checklists somebody bought off a Gumroad page in 2022. Those usually include something like 40 line items, 30 of which are settings, and a note at the bottom saying that your naming conventions are inconsistent.
(They probably are. Mine are too.)
But I've never once seen a brand grow because somebody cleaned up their campaign names. The audits that create meaningful change are the ones trying to answer a real question, which for us is always some version of, okay, why is this account not making more money than it is. Everything else is downstream of that.
We've done hundreds of these between the agency side and the membership, and I've sat in on a lot of the live ones, so I want to walk through how we do it, what it sounds like when it's happening with numbers on a screen, and then give you 10 questions you can go ask yourself tonight without anybody's help.
Start with context, not the account
Almost everyone starts by opening Ads Manager. That's where the buttons are, so that's where people go.
Before you do that, figure out which audit you're doing, because there are really only four and they're not the same job. Either:
- the account is struggling and you need it to work;
- the account is fine and you think there's more there;
- things look good and that makes you nervous so you want a second set of eyes; or
- the numbers don't add up and you need to know why.
Our community guru, Courtney Fritts, has talked plenty about this and I think it's the most underrated part of the whole process. A struggling account audit and an optimization audit will look at the exact same screens and come to opposite conclusions, and if you don't know which one you're doing you'll wander.
You also need the stuff that isn't in the platform, like real margin, your break-even point after COGS, returns, discounts, all of it, what happened to the site in the last year, what's launching, etc. An old colleague of mine, Susan Waldes, asks a version of this that I've stolen: if we saw a way to double your volume at your goal number, what would we need to think about – like cash, inventory, sell through, whether you can staff the support tickets.
Ask that question early, otherwise you'll build a beautiful plan to scale a brand that can't ship the orders – which I have done, and it's embarrassing.
Real numbers, real platform breakdowns, no vague "diversify your channels" advice.
Go wide before you go deep
Pull 90 days, sometimes six months. Look at the data as a line graph and not a table, because at this stage you're looking for the shape of the thing and not the actual numbers. Where did it turn? What was normal before that?
Then go to daily. Daily data will save you, because outliers lie constantly. Courtney had an account once where a single fake $100k test order had skewed a whole period and nobody had noticed.
This is also where you find out whether you even have an ads problem. Put ad spend next to total revenue, new customer revenue, and MER. If Meta says you're a hero and the P&L disagrees, the two suspects are almost always audience mix and attribution. Either too much of the budget is finding people who were going to buy anyway, or too much of the credit is view-through. Some view is fine – I run 7 day click and 1 day view on most things – but an account that's living on 1 day view is telling you something.
It goes the other way too, although less often. Platform numbers look weak, but the business is strong. Then I'm looking at whether revenue moves up and down with spend, whether the pixel is actually firing, and, on higher AOV stuff, what 28 day click looks like.
Then structure
Now you can look at the account.
Count your campaigns. Then count how many conversions a day each one is actually getting. From your seat, all those campaigns feel like diversification, because each one had a reason when you built it. New geo, new product line, a seasonal push. From Meta's view it's just fragmentation, and everything is getting thin data and learning slowly.
Do the math on it once and it stops being abstract. Say your CPA is $60 and you want an ad set walking toward the learning threshold, roughly 50 conversions in seven days. That's about seven a day. Seven times $60 is $420 a day into that one ad set. Now go look at how many ad sets you have live and what each one is getting. If you're running a $50k a day structure on a $1,500 daily budget, you found it, and you can stop looking for something cleverer.
Get creative, but make it concrete
This is where most of the money is and where most audits get soft, because "make better ads" isn't a finding.
Take every active ad and sort it three ways: format, angle, awareness level. Then look at where they all landed. If most of them are sitting in one square, you have the same ad wearing different outfits, and Meta is grouping them anyway. In the same vein, a different creator reading the same script is the same ad, as is a 60-second video cut to 50.
Then stop looking at ROAS for a minute and look at spend. Which ads is Meta feeding, and how fast did it start feeding them? Spend velocity tells you what the algorithm believes before your conversion data is anywhere near significant.
Alex Afterman, who runs our live account reviews, tracks basically three things at the creative level: CPA or ROAS, spend, and who the top spenders were each week, which he reports back to clients. He's open about being a soft metrics skeptic. His take is that CTR and CPM are the canary in the coal mine that tells you something is wrong, not a number you optimize toward, and he's never seen the correlation to performance that other people say they see. If you're a smaller spender you lean on those early reads harder because you can't buy your way to a real signal. Just know what you're doing when you do it.
And whatever you find, write down three to five things in order with a date to look at them again. If the audit doesn't end in a ranked list with dates on it, you didn't do an audit, you did some reading.
What it sounds like when it's live
That's the skeleton. The reason members bring accounts to the live reviews every other Tuesday is that the actual diagnosis almost never comes from the framework. It comes from follow-up questions.
A member had a campaign set at $350 a day that had spent $532 total in weeks. He came in worried about his cost cap. Alex's read was that the underspend was the whole finding: Meta is telling you the purchases aren't there at the price you're asking. So walk the cap up in steps and watch where it starts spending. That number, wherever it lands, is Meta telling you where it thinks the market is right now. Nobody gets that off a checklist.
Another guy, Josh, had two ads eating his entire budget every single time he launched anything. The instinct is to blame the creative. Alex went straight to frequency and exclusions instead. Meta found a small pocket of people in week one, they bought, and now that same pool is seeing the ads over and over. He said he's usually not a fan of excluding engagers, but in this case, do it and force the system to go find somebody new. Then they scrolled through the ad sets and found detailed targeting suggestions sitting in some of them and not others, which somebody on the team had put in at some point for reasons nobody could remember. You only find that by scrolling with another person asking what that is.
On a July call an apparel brand couldn't sell a new line and every winner eventually tanked. Alex gave four specific things to try in the account, and then said the part that ended the conversation, which was that if the pricing isn't right nothing you do in here fixes it. People look, think it's fine, and go buy it cheaper somewhere else. The account was showing symptoms. The actual thing was upstream.
Then on our September call a swimwear brand came in with CACs that had roughly doubled. After 20 minutes of people asking questions, the real numbers came out, and it was a conversion rate problem. Not a media buying problem at all.
My favorite line from those calls, and I think about it constantly: you want Meta to tell you that you need to do something rather than volunteering to do it.
Why rebuild the infrastructure from scratch? Wire the Founders knowledge base directly into Claude, Cursor, or Codex with Foxwell MCP.
The 10 questions
Take 30 honest minutes and answer these before you assemble people to help.
Which of the four audits (struggling, optimization, second opinion, or the numbers don't add up) am I doing here? Pick before you open the account.
Do the platform numbers agree with the business? Spend against total revenue, new customer revenue, MER, 90 days. If they disagree, go to audience mix and attribution before you touch anything.
What day did it turn? Reference daily data. Find the day, then find what you changed, what the site changed, and what the business changed within 72 hours of it.
Is my structure concentrating signal or shredding it? Analyze campaigns, ad sets, conversions per day in each. Do the CPA math. If nothing has enough volume to learn from, that's your answer and nothing downstream matters yet.
Is my creative actually different, or is it one ad in 10 outfits? Sort by format, angle, awareness level and look at the pile. I tell brands to keep 20-30% of their creative on absurdly out there concepts at all times, because otherwise you over refine yourself into a corner and have nothing ready when the winner dies.
Where is the spend going and how fast did it get there? Look at top spenders week over week, before ROAS. Then group it by angle and watch the mix move across the year against your marketing calendar. That usually explains more than any single metric.
If it isn't spending, what is Meta telling me? An underspending campaign isn't a bug you fix with a bigger budget. Walk the cap up and find out where the market actually is.
What's my click quality? Outbound clicks to landing page views is the fastest tell in the platform. Healthy is around 90 percent and up. When members see it drop into the 60s, they're buying junk traffic, and CPMs falling is not the good news it looks like. Check placements while you're in there.
Am I buying customers or paying to reach the ones I already have? Audience segment breakdown on spend and on revenue, then go look at every exclusion and not just the 90-day purchasers.
What would have to be true for me to double this tomorrow? Write the numbers you'd need. Contribution margin, new customer CAC, payback period, new versus returning. Then write what would stop you even if all of them were perfect.
The part that's hard to do by yourself
Run those and you'll find something, I promise. But be ready to act on what you find.
What usually happens is you find it and then talk yourself out of it over the next 10 minutes, because you built the thing and you've got a reason for every weird item in there. That campaign is there for a reason. Those two ads are supposed to dominate, they're the good ones. The pricing is a brand decision and not a marketing one, so it's off the table.
That's really all the live version is doing. Somebody says wait, go back, why is that still on, and you don't have a good answer, and now you have your audit.
Inside Founders that happens a few ways. Alex and Mike Lucas trade off the Live Meta Ad Account Reviews every other Tuesday, where members bring an account and share the screen. It's not a beat up; I don't think I've ever heard either of them tell somebody their account was trash. Most people who show up don't even bring an account, they just watch, which is honestly where a lot of the value is. If you can't share a client account live, every member gets a monthly audit instead, where you record 10 minutes walking through it and send it in. And the rest of the time it's Slack, where somebody posts at 9am that traffic fell off a cliff and by lunch six people who run accounts at every budget level have told them whether it's them or whether it's just Tuesday.
Trusted by agency owners, brand founders, and marketers everywhere, our advertising membership is where the top 1% of digital marketers go when they need help. Meta ads, creative strategy, Google ads, CRO, email and SMS, we've got experts on it all. If you want your account in front of the room, come hang out with us.
