How to Track Facebook & Instagram Ad ROI: 4 Mistakes Costing You Money

Key Takeaways

• 70% of e-commerce businesses have broken tracking showing zero conversions whilst ads actually generate 3.8 ROAS or higher

• Purchase events on checkout pages create false data making profitable campaigns look like failures worth abandoning

• Pixel-only tracking misses 40% of iPhone sales costing businesses thousands in wasted ad spend monthly

• One decorating retailer increased sales 1,919% after fixing tracking that showed their ads weren't working for six months

• Number mismatches reveal tracking problems when Facebook reports 400 conversions but your store shows 290 actual orders

• Professional tracking delivers 4.5-5.2 ROAS by showing exactly what works so you can invest in profitable campaigns

  
  

The Tracking Problem Costing You Thousands

About 70% of business owners we speak to can't accurately track ROI from Facebook and Instagram ads. They think their ads don't work. Then we check their tracking and the ads were making money the whole time. The numbers were just wrong.

When you're spending £2,000-5,000 monthly on ads, broken tracking destroys decision-making. You kill profitable campaigns because reports say they failed. Or double down on campaigns that look brilliant but aren't making money.

Meta advertising works brilliantly for online shops, but only when you can measure what's happening. Most tracking problems come from four critical mistakes. Let's fix them.

Mistake #1: Your Purchase Tracking Is On The Wrong Page

This breaks tracking for so many businesses. You follow Facebook's setup. They say track purchases. So you add the Purchase event tracking to your checkout page where people enter card details. Makes sense, doesn't it?

Nope. Complete disaster.

Here's what goes wrong: Checkout pages load before anyone's bought anything. Someone types in their card, then changes their mind. Payment gets declined. Your tracking fires anyway. Facebook thinks you made a sale when you didn't.

We've seen businesses running campaigns on completely wrong numbers because of this. Your reports show conversions that never happened.

How to fix it: Purchase events (and custom conversion tracking) only fire on your order confirmation page. The "thanks for your order" page after someone's paid.

Check yours now: Go into Facebook Events Manager, click "Test Events," add your website, then buy something from your own site. Watch events fire in real-time. If Purchase triggers before confirmation, you've got a problem.

One decorating supplies business that's seen a 1,919% increase in weekly sales checks this monthly. Catching problems early means you're not making decisions based on rubbish data.

Mistake #2: You're Only Using The Pixel (Missing Half Your Sales)

Apple changed everything with privacy updates. iOS 14.5 gave iPhone users a choice: let apps track you or don't. About 75% said don't.

If 60% of your mobile visitors use iPhones (standard for UK shops), you're missing conversion data from 45% of traffic. That's not small when you're working out which ads make money.

The problem: Meta Pixel is code that runs in browsers. iPhone users can block it. When they do, Facebook can't see purchases. Your reports say the ad didn't work. Reality? It worked perfectly. You just can't see it.

A watch brand we worked with saw 40% of actual sales through pixel alone. After server-side tracking, that jumped to 85%.

Tracking Method	What It Does	iPhone Visibility
Meta Pixel Only	Browser-based tracking	~40% (blocked by iOS)
Conversions API Only	Server-side tracking	~70% (misses some data)
Both Together	Complete tracking	~85% (recommended)

What server-side tracking does: Sends purchase information straight from your server to Facebook. Bypasses iPhone restrictions. Your website calls Facebook directly: "hey, this person just bought something."

Shopify makes this straightforward through Meta integration. WooCommerce needs a plugin. If you're spending £2,000+ monthly on ads, it's worth sorting.

That watch brand? Revenue up 40% whilst cutting ad spend 66%. Better targeting helped. But a huge part was finally seeing which campaigns made money.

Mistake #3: Your Numbers Don't Match Up (And You're Ignoring It)

Facebook says 400 sales. Google Analytics says 280. Shopify shows 390 actual orders. Which one's right?

Most people pick whichever looks best. Bad idea. Those gaps tell you something's broken.

Why numbers differ: Facebook counts sales within a time after someone clicks or views your ad. Google Analytics gives credit to whatever brought someone to your site before buying (especially if you're using UTM parameters properly). Your platform counts actual orders.

What's normal: 10-15% apart? You're fine. Different systems, different methods.

When you've got problems: Facebook reports 400 but platform shows 290? Something's wrong. Usually duplicate pixels (every purchase counted twice), test orders counted as real sales, or VAT included inconsistently.

A decorating business we've worked with for six years checks all three numbers monthly. Their 5.39% conversion rate? That comes from accurate tracking showing what works.

Mistake #4: Wrong Attribution Windows (Killing Good Campaigns)

Attribution windows are just rules for when Facebook gets credit for a sale.

Someone sees your Instagram ad Monday. Doesn't buy. Wednesday, they remember it, search your brand, buy. Should that count as coming from your ad?

7-day attribution window: yes. 1-day window: no. Change this setting, your ROAS looks completely different.

Why this kills campaigns: Most people use Facebook's default 7-day setting. But if you sell anything over £100 where people take weeks to decide, you're missing loads of conversions outside that window.

A baby carrier brand found customers took 11 days from first ad to buying. With 7-day windows, we missed 30% of conversions. Switched to 28 days, saw the full picture.

The result? They tripled sales and hit 5.2 ROAS. Great products helped. But accurate attribution showed which campaigns actually worked.

How to choose: Look at your customer data. See how long people take from first visit to buying. Set windows to match what your customers actually do.

What Accurate Tracking Actually Gets You

Proper tracking doesn't just give you the right numbers. It completely changes what you can do with your advertising.

You stop second-guessing budget decisions because you know which campaigns make money. You catch problems early when iPhone updates break something or platform changes mess with tracking.

Most importantly, you make decisions based on profit instead of clicks. That watch brand that cut spend by 66% whilst growing revenue 40%? They worked out which campaigns brought in genuinely profitable customers. The decorating business with 5.39% conversion rate? They split tracking between trade and retail customers. The baby carrier brand that tripled sales? Accurate attribution showed which campaigns worked across their actual 11-day buying cycle.

That's what proper tracking delivers. Essential for turning Facebook and Instagram ads from expensive guessing into reliable growth.

When To Sort This Yourself vs Getting Help

Basic tracking setup isn't rocket science. Most business owners can handle it if they're reasonably comfortable with technology. But there's a point where doing it yourself costs you more than getting someone who knows what they're doing.

That point is usually around £2,000-3,000 monthly ad spend.

If you're dropping £5,000 a month on ads and your tracking's 30% out (pretty common with basic setups after all the iPhone privacy changes), you're making decisions based on wrong information. You might be killing campaigns that actually work or pumping money into campaigns that look great but don't make you any money.

You probably need professional help when:

• Facebook and your actual shop orders are more than 15% apart

• You've set up server-side tracking but the Event Match Quality score is below 6.0

• You're spending £2,000+ monthly but honestly can't say which campaigns work

  
  

We're pretty transparent about when you need proper setup versus when you're fine sorting it yourself. Testing ads with £500 a month? Do it yourself, learn as you go. Spending £3,000+ where advertising is a major part of how you grow? Your tracking needs to be spot on.

Those businesses we mentioned seeing 1,919% sales increases, 5.2 ROAS, 40% revenue growth whilst cutting spend? They've got accurate tracking showing exactly what works. Could they have got those results doing it themselves? Probably not, because they wouldn't have had the right data to make the right calls.

Get in touch if you want a straightforward conversation about whether your tracking's accurate enough for what you're spending. We're pretty direct about when you need professional help versus when you're fine sorting it yourself.

Caio for now

Ali Puglianini

FAQs

What's the single most important thing to fix with my tracking?

Make absolutely sure your purchase events only fire on the order confirmation page, not during checkout. This one mistake creates completely wrong conversion data. Use Facebook's test tool to watch events whilst you complete a test purchase. If Purchase fires before confirmation, fix it now.

How do I know if I'm missing iPhone sales?

Check if you've got Conversions API (server-side tracking) set up alongside your Meta Pixel. If you're only using the pixel, you're missing roughly 40% of conversions from iPhone users. Facebook's Events Manager will show you whether server-side tracking is actually working.

Why do Facebook and Shopify show different numbers?

Different ways of counting. Facebook counts sales within your attribution window after someone interacts with your ad. Shopify counts all orders regardless of where they came from. Being 10-15% different is normal. Bigger gaps mean something's broken.

Should I fix this myself or hire someone?

Under £1,000 monthly on ads? Do it yourself with the guides. Between £1,000-2,000? Consider getting help if you're not particularly technical. Over £2,000 monthly? Professional tracking usually pays for itself pretty quickly through better decisions.

What's Event Match Quality?

Score out of 10 showing how well your server-side data matches up with users in Facebook's system. Below 6.0 means Facebook can't reliably connect your sales to the ads people saw. Above 7.0 is good. You improve it by sending complete customer data in the right format.