Solved: Facebook Ads Metrics Don't Match Backend Data

Hi there,

Thanks for reaching out!

Honestly, don't feel bad about being confused by the numbers. What you're seeing is incredibly common, and something we see in about 90% of the ad accounts we audit for the first time. The ad platforms are notorious for their... let's call it 'creative' accounting. It's built to make the platform look good, not necessarily to give you the stone-cold truth.

The good news is that this mess of data is actually a blessing in disguise. It's a massive red flag pointing to some deeper issues with your tracking and probably your campaign setup. Fixing this isn't just about getting the numbers to line up; it's about building a foundation you can actually trust to scale your business. I'm happy to give you some initial thoughts and a bit of guidance on how to untangle this mess and start making decisions based on reality, not vanity metrics.

We'll need to look at why your numbers are a complete mess...

Right, let's get straight to it. The reason your numbers don't line up is because they are measuring different things in different ways, and one of them is almost certainly broken. It's not a simple apples-to-oranges comparison; it's more like apples-to-invoices-for-oranges-that-never-arrived. You cannot make good decisions with bad data, and right now, your data is very, very bad.

Let's break down the specific issues you mentioned. This is where the detective work begins.

"Purchases" (Platform) shows 1 vs "Results" (Thank-You Page) shows 0

This is the classic platform attribution puzzle. The "Purchases" metric inside your Ads Manager is governed by what's called an 'attribution window'. By default, Meta (and others) will use something like a "7-day click, 1-day view" window. This means the platform will take credit for a purchase if someone:

• -> Clicked your ad anytime in the last 7 days and then purchased (even if they came back later through a Google search or by typing your URL directly).
• -> Simply saw your ad (without clicking) in the last 24 hours and then purchased.

You can see how this gets out of hand. The platform is incentivised to claim responsibility for as many conversions as possible. Your thank-you page, on the other hand, only registers a "Result" when someone actually lands on it right after a transaction. It's a much more literal, and therefore more reliable, measurement. In this case, the platform probably saw a user who viewed your ad yesterday, then came back to your site today via another channel and bought something. Meta says "That was me!", while your website correctly says "That ad didn't directly cause this sale."

This is why we tell our clients to treat platform-reported purchases with extreme suspicion. They are a directional guide at best, and dangerously misleading at worst. The only number you should trust for actual revenue is the one in your payment processor's dashboard or your eCommerce backend (like Shopify or WooCommerce). That's real money in the bank. Everything else is just noise.

Landing Page Views (2) vs Website Checkouts (4)

Okay, this one is different. This isn't an attribution quirk; this is a clear-cut technical failure. It is physically impossible for four people to initiate a checkout if only two people ever viewed the landing page. This is the smoking gun that tells us your tracking setup is fundamentally broken. There are a few likely culprits for this kind of nonsence:

• -> The Pixel is Firing Multiple Times: The most common issue is that the 'InitiateCheckout' event code is being triggered more than once. For example, a user clicks "Checkout," and the event fires. Then they change the quantity or refresh the page, and the event fires again. We've seen setups where it fires every single time the checkout page loads, leading to wildly inflated numbers.
• -> The Pixel is in the Wrong Place: Someone might have accidentally placed the InitiateCheckout event code in the header or footer of your entire website, or on a page that isn't even part of the checkout flow. This is less common but we've seen stranger things.
• -> Google Tag Manager Mishaps: If you're using GTM, the trigger for the event could be misconfigured. It might be set to fire on "All Page Views" instead of a specific button click or thank-you page load. GTM is powerful, but it's also a very easy place to make costly mistakes if you don't know exactly what you're doing.

This is the most urgent problem to fix. You can't optimise a campaign if a core funnel metric is completely fabricated. It's like trying to navigate a ship with a compass that spins randomly. To help you figure out where the problem lies, here's a simple diagnostic process you can follow.

I'd say you need to establish a single source of truth...

So, which metric should you trust? The brutally honest answer is: none of them, individually. Not at first. Your job isn't to get the numbers in Meta Ads Manager to perfectly match the numbers in Google Analytics. They never will. Your job is to understand what each tool is good for and build a system where you use the *right tool for the right job*.

You need to decide on your 'single source of truth' for business performance. For 99% of businesses, this should be your eCommerce platform's backend (e.g., Shopify Analytics) or a well-configured Google Analytics 4 property. Why? Because these track actual transactions and user behaviour on your site, independent of how the user got there or what ads they might have seen last Tuesday. It's the closest you can get to objective reality.

Think of it like this:

• -> Your Website Backend (Shopify/GA4) is your accountant. It tells you the final, non-negotiable truth about how much money you made.
• -> Your Ad Platform (Meta/Google) is your salesperson. They're great at telling you how busy they are (impressions, clicks) and will always try to take credit for every sale, but you still need the accountant to confirm the actual numbers.

You use the salesperson's data to judge their performance (e.g., is this ad creative getting a good Click-Through Rate?), but you only use the accountant's data to judge the health of the business. This means you need to get comfortable with calculating your key metrics yourself, rather than relying on the ad platform's potentially inflated numbers.

You probably should focus on the metric that actually matters: ROAS...

This brings us to the most important shift in mindset you need to make. You're asking "Which metric should I trust?" when the more powerful question is "Which metric actually grows my business?". The answer isn't clicks, checkouts, or even the number of purchases. The answer is Return On Ad Spend (ROAS).

ROAS tells you, for every pound you put into the advertising machine, how many pounds you get back out. It's the ultimate measure of profitability. A campaign with 100 cheap purchases that only results in £50 of revenue is a failure. A campaign with just 5 expensive purchases that results in £500 of revenue is a success. ROAS is what separates the two.

You calculate it with a simple formula: Total Revenue from Ads / Total Ad Spend. As we established above, you take the 'Ad Spend' from the ad platform and the 'Total Revenue' from your trusted source of truth (your website backend). Let's see how this works in practice.

But just knowing your ROAS isn't enough. Is a 4x ROAS good? For some businesses, it's incredible. For others, with tiny margins, it means they're losing money on every sale. To understand what ROAS you should be aiming for, you need to know your Customer Lifetime Value (LTV). This tells you the total profit a single customer will bring to your business over their entire relationship with you.

The real question isn't "How low can my Cost Per Purchase go?" but "How high a Cost Per Purchase can I afford to acquire a truly great customer?" LTV gives you that answer. If you know a customer is worth £1,000 in profit over their lifetime, paying £100 to acquire them suddenly looks like an absolute bargain.

You'll need a better campaign structure to get reliable data...

Once you've fixed your pixel and started focusing on true ROAS, the final piece of the puzzle is your campaign structure. Bad data often comes from a messy, illogical campaign setup. When you lump all your audiences into one or two ad sets, you get a blended, confusing average. You have no idea what's actually working.

The way we structure accounts for our clients, especially for eCommerce, is by splitting campaigns by their position in the marketing funnel: Top of Funnel (ToFu), Middle of Funnel (MoFu), and Bottom of Funnel (BoFu). This isn't just jargon; it's a logical way to separate your audiences and analyse their performance cleanly.

• -> ToFu (Top of Funnel): This is your cold audience. People who have never heard of you before. Here, you test broad interests and Lookalike audiences to find new customers.
• -> MoFu (Middle of Funnel): These are people who have engaged with you but haven't taken a key action yet. Think website visitors, video viewers, or social media followers. The goal here is to bring them back and push them towards the checkout.
• -> BoFu (Bottom of Funnel): This is your hottest audience. People who have added a product to their cart or initiated checkout but didn't complete the purchase. These ads are highly targeted and aim to recover what would otherwise be a lost sale.

By separating them, you can allocate your budget more intelligently and get crystal clear data. You'll know exactly what your prospecting efforts are costing you versus your retargeting efforts. A typical starting structure might look something like this:

When you structure your account this way, the data becomes meaningful. You can clearly see if your problem is attracting new customers (ToFu is failing) or converting interested prospects (BoFu is failing). This is how you move from being confused by numbers to confidently directing your ad spend where it will have the most impact.

This is the main advice I have for you:

I know this is a lot to take in, so let's boil it down. If you do nothing else, focus on these core actions. I've detailed my main recommendations for you below:

I hope this detailed breakdown has been helpful. As you can see, the problem you're facing isn't just a simple reporting glitch. It's a symptom of a deeper issue with the technical setup and strategic foundation of your advertising efforts. Fixing the pixel is the first, essential step, but the real growth comes from building a robust system for measurement and a logical structure for your campaigns.

This is, quite frankly, where professional help can make a monumental difference. An expert can diagnose and fix these technical issues in a fraction of the time it would take to learn from scratch, preventing weeks of wasted ad spend on bad data. More than that, they can build the strategic framework—the funnels, the audience testing, the creative rotation—that allows you to scale confidently, knowing that your decisions are based on solid ground.

The work we do for our clients is exactly this: we come in, clean up the technical mess, establish a reliable measurement system, and then build and manage high-performance campaigns based on real, trustworthy data. If you'd like to see what that process would look like for your business, we offer a completely free, no-obligation initial consultation where we can review your account together and outline a clear path forward.

Hope this helps!

Regards,

Team @ Lukas Holschuh