Meta has a hundred ways to spend your budget on things that aren’t customers. Likes. Reach. Engagement. Video views. Meta calls these results. You probably don’t.
Let’s find what does.
Disappears.
Most Meta accounts lose money in the same five places. Same algorithm for everyone, so the failures repeat.
Check yours against the list.
Your campaigns are optimised for the wrong thing.
Meta lets you pick an objective when you create a campaign. Awareness. Engagement. Traffic. Leads. Sales. Most accounts pick whatever sounded right at the time and never change it.
If your objective doesn’t match what you actually want, every other setting in the campaign works against you.
Your targeting is “anyone who breathes.”
Meta has two billion users. Picking “Adults 18 to 65 interested in business” is not targeting. It’s surrender.
Without audiences built from your own customer data, Meta will spend your budget reaching whoever is cheapest. Cheapest is rarely your buyer.
Meta doesn’t know what a sale looks like in your business.
Meta tracks what happens on your site through a small piece of code installed there. If that tracking is missing, broken, or pointing at the wrong actions (and on most accounts it is), Meta is guessing at what works.
The bidding learns from incomplete data. You spend more to deliver less.
Your creative is the problem. And you don’t know it.
On Meta, the creative is the campaign. The image, the hook, the first three seconds of the video. That’s what gets people to stop scrolling.
Most accounts run the same three images for six months, can’t figure out why nothing is working, and blame the algorithm. The algorithm is fine. The ad is not.
You hit Boost. Meta loves it.
The Boost button is the option that appears under your Facebook and Instagram posts. One click and you’re spending money.
The downside: Boost picks the audience, the goal, and the placement for you. None of those are the choices an advertiser would make.
Your account has at least two of these happening. Probably three. The audit shows you exactly which ones apply. You decide what to do about them.
Actually Looks Like.
Eight standing tasks. Plus the work between them, which is most of what running a Meta account actually is.
Creative Pipeline.
Three to five new ad concepts produced or sourced every month. Hooks tested against each other. Winners scaled, losers killed. If your brand system is inconsistent, creative can’t compound recognition across touchpoints. Creative is not a deliverable. It is the engine. We treat it that way.
Account Structure Rebuild.
Campaigns separated by objective. Ad sets separated by audience. Ads separated by hypothesis. Most accounts we inherit have one campaign trying to do five things at once. Meta’s algorithm needs clean signals to learn from. You need reporting that doesn’t lie.
Audience Architecture.
Custom audiences from your website, customer list, and engagement events. Lookalike audiences built on buyers, not browsers. Exclusion stacks so you stop paying to reach people who already bought. Cold, warm, and retargeting audiences each in their own ad set.
Pixel and Conversions API Setup.
Meta Pixel installed and verified. Conversions API running alongside the Pixel so server-side data backs up browser-side data. Standard events tied to your real conversion points. Custom events for the rest. The same rigour we apply to Google Ads conversion tracking applies here. We do not run campaigns against tracking that doesn’t reflect reality.
Campaign Strategy By Funnel Stage.
Cold audiences get awareness and consideration ads. Warm audiences get engagement and product ads. Retargeting audiences get conversion ads. Each stage runs its own budget, its own creative, its own measurement. No single campaign tries to do everything.
Real A/B Testing.
Tests run through Meta’s experiments tool, not duplicate ads thrown into the same ad set hoping for clarity. One variable changed at a time. Sample sizes that produce significance, not vibes. Most account “tests” aren’t tests. They’re hopes.
Revenue-Tied Reporting.
Weekly numbers that tie to your bank account, not screenshots of Ads Manager. ROAS, cost per acquisition, contribution to revenue. If the ads worked, the number says so. If they didn’t, the number says that too. Either way, you read English, not dashboards.
Advantage+ Controls.
Advantage+ Shopping campaigns where they actually outperform manual setup. Manual campaigns where they don’t. Audience suggestions defined, not left to Meta’s defaults. Where Advantage+ helps, we use it. Where it eats budget without proof, we don’t.
And then dailyYour account gets opened every working day. Frequency tracked daily. Creative fatigue spotted before it spreads. CPM and CPP watched against last week’s benchmark. When something drifts, you hear about it the same day. Not at the end of the month.
Hot.
Meta ads work in stages. Every stage needs different creative, different messaging, and different expectations. Most accounts run one campaign for all three and wonder why nothing converts.
Showing your brand to new people.
For when your brand isn’t known. Meta puts your ad in front of strangers who fit your audience. Cheap to run. No promise anyone acts on it. Useful only if you have a way to come back to those people later.
Clicks to your site.
Meta finds people likely to click on a link. The downside: clicks aren’t customers. We use Traffic campaigns selectively, when the goal genuinely is to send people somewhere off-Meta. For reaching the same audience in their inbox, cold email handles that motion.
Likes, comments, shares, follows.
Useful for building social proof or warming up an audience for retargeting. Not useful as a primary KPI for a business that needs revenue. Most accounts overuse this.
In-platform lead forms.
Users tap an ad, a form opens inside Meta with their details pre-filled, they submit without leaving the app. Higher volume than off-platform forms. Lower intent. Works only if your sales team can handle the volume and qualify fast.
Actual purchases.
For ecommerce, the conversion objective tied to your Pixel and catalog. Meta optimises bidding toward the people most likely to buy. This is where most account budgets should sit if revenue is the goal.
One more thingIf a campaign type doesn’t match the audience temperature, we don’t run it. Awareness to a hot audience is wasted reach. Sales to a cold audience is wasted budget. Saving you the spend is part of the work.
Numbers Move.
Meta’s algorithm needs about fifty conversions a week per ad set before it stops guessing. So nothing serious happens overnight. Here is what happens, and roughly when.
We log into Business Manager. Pull six months of spend, audience overlap reports, Pixel events, and creative performance. We tell you, in writing, what’s working, what’s broken, what’s quietly bleeding budget. The document is yours whether you hire us or not.
Account structure rewritten around objectives that match your actual goal. Pixel and Conversions API verified end-to-end. Audiences rebuilt from your customer data. Creative produced or sourced and queued in batches of three to five hooks. Nothing goes live until the foundation holds.
The new structure starts producing clean data. We test new creative against the originals. We expand the audiences that converted. We kill the ones that didn’t. CPMs stabilise. Cost per result drops. Frequency stays in healthy range. The account starts feeling like an account, not a leak.
Winning ad sets get more budget. Losing ones get killed without ceremony. New creative joins the rotation every two weeks. Lookalikes refresh against the latest buyers. Relevance scores climb, CPMs drop, you spend less to get the same result. This is where the compound starts.
Meta starts slower than Google and compounds faster. The first month feels like nothing. The third month feels like everything. Anyone who skips the learning phase is either lying to themselves or to you.
To Revenue.
Real businesses. Real numbers. Before. After. How long it took.
4.1x ROAS in 75 days.
Direct-to-consumer brand. The Pixel was firing on the wrong event, the attribution window was set to one-day-click only, and the creative library was six months stale. We fixed all three. The new ROAS held for the next quarter.
CPL down 52% in 60 days.
B2B services account using in-platform lead forms. Audience targeting tightened, creative refreshed monthly, qualifying questions added inside the form. Same monthly budget. Twice the qualified pipeline.
Cost per purchase down 38% in 90 days.
Ecommerce. Catalog campaigns set up properly for the first time. Audience stack rebuilt from the customer list. Two-week creative refresh cadence. More sales on the same monthly spend.
Nobody promised these numbers before we’d seen the account. Anyone who promises numbers before reading your account is selling you something else.
Hear Us Say.
Five sentences common in agency pitches. None of them will come out of our mouths.
“Let’s just boost this post.”
No. Boost is a button for people who don’t know they’re doing it wrong. It picks the audience, the placement, and the goal for you. None of those are the choices an advertiser would make. If your account needs more visibility, we build a proper campaign. We don’t tip.
“Let’s let Advantage+ figure it out.”
Sometimes. Not by default. Advantage+ is Meta’s automated campaign type. It works when the rest of the account is solid enough to feed it. When it isn’t, it spends your budget in ways nobody can trace. We use it deliberately or not at all.
“Engagement is up 40% this month.”
Likes, comments, reach, video views. Those are what Meta reports to make you feel something. They don’t fund payroll. We report the number on your bank statement. If you want a likes-focused agency, there are dozens. Hire one of them.
“We can guarantee you a 5x ROAS.”
No one can. Anyone promising a specific number before reading your account is making it up. We tell you what’s possible after the audit. With reasons. Not before.
“Our fee is fifteen percent of your spend.”
Not how we work. We charge a fixed monthly fee. The percentage-of-spend model tempts agencies to spend more of your money to earn more of theirs. We’re not on that side of the table.
Your competitors are on Meta right now. Their ads are stopping the scroll. Yours aren’t.
Twenty minutes on a call. We go through your account live. You’ll know in ten whether the work is worth doing. So will we.
Before They Book.
We get asked the same things. So we put them here.
How long until we see Meta numbers move?
Two to three weeks for the obvious wins (turning off broken tracking, killing creative fatigue). Six weeks for the learning phase to settle. Twelve weeks for the compound to start showing in your bank statements. Anyone promising faster is guessing.
What does this cost?
Depends on the spend, the creative load, and how much of the account needs rebuilding. We quote a fixed monthly fee after the audit. The percentage-of-spend model that most Meta agencies use tempts them to spend more of your money to earn more of theirs. We don’t work that way. The fee is the fee.
What if my current agency is fine?
Don’t switch. Seriously. We’ve turned away three accounts this year because the existing agency was doing solid work. If your Meta numbers are moving in the right direction and you can explain why, you don’t need us.
Existing campaigns or start from scratch?
Depends on the Pixel. If the historical conversion data is clean, Meta’s algorithm has already learned from it and the campaigns are worth keeping. If the Pixel has been firing on the wrong events for six months, the algorithm has learned the wrong things and we rebuild. The audit tells you which.
Who actually runs my account?
The person who scopes the work on the audit call is the person inside your Ads Manager every week. Not a junior. Not someone three reporting layers away from where the ads actually run. You can call them directly when something is off.
Do you handle creative production?
Creative is the most important variable on Meta, so yes. If you have an in-house creative team, we brief and direct them. If you don’t, we have a network of editors and photographers we work with. Either way, you’re not running ads we put together in PowerPoint.
What’s the minimum monthly Meta spend?
We have one, and it’s high enough that we can do meaningful work. We’ll tell you in the first call whether your account meets it. If not, we’ll tell you what to do until it does.
iOS broke Meta attribution. How do you handle it?
We don’t pretend it isn’t broken. We use Conversions API to recover most of what the Pixel misses, UTM tracking and GA4 to triangulate, and modelled attribution where the data is too thin to trust. The answer is “close to true,” not “perfectly true.” Anyone claiming perfect attribution is lying.
Will you replace our in-house team or work alongside them?
Either. We work with marketing managers who want their Meta account handled and with founders who don’t have a team yet. You keep ownership of the account, the Pixel, the data, and the creative assets. We bring the discipline and the bandwidth.
Our Pixel data is a mess. Is that a deal-breaker?
No. The first three weeks of working together are basically cleanup: fixing the Pixel, setting up Conversions API, rebuilding audiences from real data, reviewing twelve months of spend. Most accounts have a mess. Yours isn’t special.