Tracking you can trust, reporting on your real numbers, and tests that prove what’s working.
Illustration: browser tags and server-side tracking working together. Each sale is matched by event ID so it counts once, and server-side catches the sales ad blockers and privacy settings hide from the browser.
Brightpace sets up and fixes tracking, reports on commercial numbers such as MER, new customers and profit, and runs lift tests to show what your ads really add. If you have no tracking yet, we build it from scratch as part of getting started.
Platform numbers flatter you.
Add up what every ad platform says it drove and you’ll often get more revenue than the business actually made. Each one is marking its own homework, and none of them can tell you which sales would have happened anyway.
That matters, because budget follows the numbers. If retargeting and brand search look brilliant but mostly reach people who were already going to buy, money flows to the wrong place. Measurement fixes the numbers first, then proves what genuinely adds sales. Which number should you trust? See MER vs ROAS vs POAS.
No tracking yet? We build it from scratch. Already have some? We audit and fix it: Google Tag Manager, GA4 key events, the Meta Conversions API, Google enhanced conversions, the Microsoft UET Conversions API, the TikTok Events API and Consent Mode. No double-counting, and real values passed to every platform.
Server-side tracking recovers sales lost to ad blockers and browser limits. Meta reports a 13% better cost per acquisition on average when the Conversions API runs alongside the Pixel, and Google reports median uplifts of about 5% (Search) and 17% (YouTube) with enhanced conversions. Offline conversions feed real outcomes from your CRM (qualified leads, booked calls, closed deals) back to the platforms. See server-side tracking.
A live report built on the numbers your business runs on: new vs returning customers from your own shop or CRM, cost per new customer, MER and profit, shown alongside the platforms’ own figures.
Each sale marked as a new or returning customer from your own records, and a “new customer” flag passed to Google Ads, Meta and GA4 where possible, so campaigns can bid towards new buyers.
Holdout and lift tests (Meta Conversion Lift, Google experiments, geo tests, and brand-search or retargeting tests) to find the sales your ads actually cause, and the incremental ROAS behind them. See incrementality testing.
Marketing mix modelling to estimate each channel’s contribution over time and where extra spend stops paying, with a year or more of consistent data.
By people or by region: a test group that sees your ads, and a control group that doesn’t.
Usually a few weeks, so the difference between the groups is clear rather than random noise.
Extra sales in the test group are the lift your ads caused. Try the maths with our incrementality calculator, or read our guide to lift tests.
What’s broken, what’s double-counting, and what to fix first.
Conversions defined once, with the right values, across every platform.
MER, new customers, cost per new customer and profit, from your own sales data, updated daily.
Which lift tests to run, in what order, and what each will tell you.
Incremental sales and iROAS, and what to do with the budget.
Spend moved toward what genuinely adds sales.
Meta, Google and TikTok each take credit, so the reported total is more than you actually made.
A “7-day click or 1-day view” rule credits sales that would have happened anyway. See attribution vs incrementality.
They reach people who were already going to buy, so they often look far better than they are.
Ad blockers, consent choices and site changes mean some sales never reach the platforms at all.
If we run your ads, tracking fixes and commercial reporting are part of the service, and lift tests are planned in as the account grows. Ask for a quote.
Already have an agency or an in-house team? We can fix tracking, set up reporting or design and read out a lift test on its own, quoted up front. Or start with a paid audit, from £195, which includes a full tracking check.
Measurement runs through everything we do. See how we measure on Meta, Google Ads, Shopping and programmatic. For the set-up details, read our guides to the Meta Conversions API and GA4 key events, or brush up on the terms in our glossary.
Anything else, ask the person who would run your account.
Ask a question →Incrementality is the extra sales or leads your advertising actually causes: the ones that wouldn’t have happened without it. It’s different from the conversions ad platforms report, which include people who would have bought anyway.
Attribution decides which ad gets credit for a sale that happened, based on rules like “7-day click”. Incrementality asks whether the ad caused the sale at all, usually by comparing a group who saw ads with a group who didn’t.
You split your audience, by people or by region, into a test group that sees your ads and a control group that doesn’t. After a set period, the difference in sales between the two groups is the lift your ads caused. Platforms such as Meta and Google offer built-in lift tests, and geo tests can be run across channels.
Enough for the difference between test and control to be clear rather than noise. That depends on how many sales you get and how big the effect is. Platform lift tests also have their own eligibility rules. We’ll tell you honestly whether your account is big enough before you commit.
Often, yes, at least in part. Ad blockers and browser privacy settings stop some browser tracking, so connections like the Meta Conversions API and Google enhanced conversions recover conversions that would otherwise be missed. For most businesses that means proven integrations and templates. If your set-up needs a developer, we’ll tell you up front. On average, Meta reports a 13% improvement in cost per acquisition for advertisers who add the Conversions API to the Pixel, and Google reports median conversion uplifts of about 5% on Search and 17% on YouTube from enhanced conversions. Your own uplift will vary, and Meta shows it in Events Manager. Our server-side tracking page explains the options.
Marketing mix modelling uses statistics on your historical sales and spend to estimate how much each channel contributes over time, without tracking individuals. It suits larger budgets with a year or more of consistent data, and works best alongside lift tests.
Google’s Consent Mode adjusts how tags behave based on each visitor’s cookie choices, and models some of the conversions lost when people decline tracking. It’s part of setting up compliant, accurate measurement in the UK.
From your own records, not the ad platforms. For online shops, your shop platform (such as Shopify) already knows whether each order is a customer’s first, so we use it to count new customers and work out your cost per new customer: total ad spend divided by new customers. Where possible we also pass a “new customer” flag to Google Ads, Meta and GA4, so campaigns can be judged on, and bid towards, finding new buyers. For lead generation, new customers are new clients won, taken from your CRM or lead tracker. We just need read access to your shop or CRM. (GA4’s “new users” counts new visitors, not new buyers, so we don’t report it as new customers.)
Yes. We build it from scratch: Google Tag Manager, GA4 and its key events, conversion tracking in Google Ads and Microsoft Ads, the Meta pixel and Conversions API, and Consent Mode connected to your cookie banner. It’s included when we manage your ads, or quoted up front as a standalone project. We use the platforms’ standard integrations and tell you up front if anything on your site needs a developer.
Yes. Tracking fixes, reporting set-up and incrementality tests can be run as standalone projects, alongside whoever manages your accounts. The aim is a neutral view of what’s working.
Tracking and reporting are included in our media management fee. Standalone projects such as tracking fixes, lift tests or modelling are quoted up front. Our paid audits, from £195, include a full tracking check.
Tell us which platforms you run and how you measure today, and we’ll tell you where the numbers are likely misleading you.
