Attribution Models Explained Without the Jargon
An attribution model decides which marketing gets credit when a customer touched several before buying. Last click gives it all to the final step. Data-driven attribution, the default in Google Analytics and Google Ads, spreads credit based on patterns in your data. Every platform also reports its own view, so Google and Meta can both claim the same sale. Use attribution to compare channels, and check big decisions against your actual sales.
- Customers take several steps. A search, an ad, a review, a return visit. Attribution decides who gets the credit.
- Two models are left in Google’s tools. GA4 and Google Ads now offer data-driven and last click. The older rule-based models were removed in 2023.
- Platforms mark their own homework. Google, Meta and others each credit themselves, so their totals overlap.
- Your sales are the referee. Compare platform claims with what actually came in.
- Test when it matters. Pausing a channel in one area, or a lift test, shows what it really adds.

The problem attribution tries to solve
Few customers see one ad and buy. A typical path might be a Meta ad seen on Monday, a Google search on Wednesday that leads to your site, a look at your reviews, and a return visit by typing your name on Friday, when they fill in the form.
Which of those caused the inquiry? All of them played a part. An attribution model is a rule, or an algorithm, for dividing the credit so you can compare what each channel contributes and decide where to spend.
The models, plainly
Last click. All the credit goes to the last thing the customer clicked before converting. Simple and easy to understand, but it undervalues anything that introduced the customer earlier, such as social ads or blog content.
First click. All the credit goes to the first interaction. The mirror image of last click, overvaluing discovery and ignoring what closed the deal.
Linear, time decay and position-based. Rule-based ways of sharing credit evenly, toward recent steps, or toward the first and last. Google removed these from Google Analytics 4 and Google Ads in 2023.
Data-driven. Uses your account’s conversion data to estimate how much each touchpoint contributed, compared with paths that didn’t convert. It’s now the default in GA4 and Google Ads. It’s more realistic than fixed rules, but it’s a model you can’t fully inspect, and it only sees the touchpoints Google can track.
Add up every platform’s reported conversions and you’ll often get more than you actually sold.
Why the platforms disagree
Each advertising platform counts conversions its own way. Meta credits a sale to an ad someone saw or clicked within its attribution window. Google Ads credits clicks on its ads. Both can claim the same customer from the example above. Add up every platform’s reported conversions and you’ll often get more than you actually sold.
None of them is lying. They’re each measuring their own contribution, generously. The mistake is adding their numbers together, or trusting any single one as the full story. Meta ads for lead generation covers this from the social side.
How to use attribution sensibly
Pick one source of truth for totals. Your CRM, your sales system or GA4 with key events properly set up. Judge overall performance there, not in each ad platform. The six GA4 reports that matter covers the setup.
Use platform data for decisions inside the platform. Which campaign, ad or keyword is doing best within Google Ads is still a fair question for Google Ads to answer.
Feed real outcomes back. When the platforms know which leads became sales, their attribution and bidding both improve. Offline conversion imports covers how.
Test before big changes. If you’re considering cutting a channel, pause it in one region or for a set period and watch what happens to total sales. Some platforms also offer lift studies that compare people who saw ads with people who didn’t. That’s the closest you’ll get to proof of what a channel adds.
What attribution can’t see
Word of mouth, a recommendation from an AI assistant, a sign on a van, a review read on a phone and then a call made from memory. Much of what drives a customer never shows up as a trackable click. Browser privacy settings hide more. Server-side tracking recovers some of it.
So treat attribution as a useful guide, not an exact accounting. Ask customers how they found you, too. It’s unscientific and often more revealing than any model. Reporting inquiries, not impressions covers putting all of it into a report that reflects what’s really happening.
Whenever our company has an idea or inquiry we hear back instantly. They handle our account 24/7 as if they are our employees.
Clients were growing tired and concerned. Sublyme understood the situation and were available 24/7 to keep them satisfied.
Powerhouse professionals. Conscientious, and they deliver projects on time and accurately executed.
A briefing built the moment you ask for it.
No subscription, no sequence, nothing to unsubscribe from. Ask for an edition and it arrives in minutes, written for you rather than sent to a list.
Frequently Asked Questions
- What is an attribution model?
A rule or algorithm for dividing credit for a conversion among the marketing touchpoints a customer interacted with before converting.
- What attribution models does GA4 offer?
Data-driven attribution, the default, and last click. Google removed first click, linear, time decay and position-based models from GA4 and Google Ads in 2023.
- Why do Google Ads and Meta report more conversions than I had?
Each platform credits conversions to its own ads, so a customer who interacted with both can be counted by each. Their totals overlap.
- Which attribution model should I use?
Data-driven is a sensible default where you have enough data. More important is using one source of truth for totals and checking big decisions against actual sales.
- How can I tell if a channel is really working?
Test it: pause it in one region or for a set period and watch total sales, or use a platform’s lift study. Asking customers how they found you also helps.
Make sure the site is converting the traffic it gets.
The free audit scores a page on speed, accessibility and technical SEO, and comes back with three fixes. No call, no card.