Attribution model

What is an attribution model?

An attribution model is the set of rules that decides how credit for a conversion, such as a purchase or a form submission, is shared between the marketing touchpoints that came before it. A customer rarely buys after one visit. They may find you through organic search, return from a newsletter, click a retargeting ad and finally type your address directly. The attribution model answers one question: which of those steps deserves the credit?

The answer matters because it drives budget. If you credit only the last click, you will over-invest in channels that close and under-invest in channels that introduce your brand.

The main attribution models

• First touch: all credit goes to the first interaction. It favours awareness channels such as SEO and content.
• Last touch: all credit goes to the final interaction before conversion. It favours bottom-funnel channels such as direct and branded search.
• Linear: credit is split equally across every touchpoint.
• Time decay: touchpoints closer to the conversion receive more credit.
• Position-based (U-shaped): the first and last touches get the most credit, often 40% each, and the middle touches share the rest.
• Data-driven: a statistical or machine learning model assigns credit from the real paths of converting and non-converting users.

Comparing the models

How tracking makes attribution possible

An attribution model is only as good as the data behind it. Touchpoints are recorded through UTM parameters on links, cookies or first-party identifiers, and a conversion pixel or event that fires when the goal is reached. A tracking link looks like this:

https://example.com/offer?utm_source=newsletter&utm_medium=email&utm_campaign=autumn-launch

Consistent UTM naming, a clean data layer and good use of first-party data keep paths intact. Cookie restrictions, ad blockers and consent choices create gaps, so every attribution report is an estimate, not an exact truth.

Best practices and pitfalls

• Define the conversion first, then choose the model. A lead and a sale may need different models.
• Compare several models side by side before moving budget.
• Use a lookback window that matches your sales cycle, for example 30 days for e-commerce and 90 days for B2B.
• Do not read last-click reports alone. They hide the role of search, content and social in creating demand.
• Combine attribution with other signals, such as surveys asking how customers heard about you, to cover blind spots.
• Document your choices so reports stay comparable over time.

Attribution, SEO and the customer journey

Organic channels often start journeys that other channels finish. Under last-click, organic traffic can look weaker than it is, while retargeting looks stronger. Mapping the customer journey and testing several models shows the real contribution of each channel, and supports better decisions on conversion rate optimization.

Attribution models at BeBranded

At BeBranded, we set up attribution before we optimise it: clear conversion events, consistent UTM rules, a clean data layer and a reporting view that compares models, so you can see what SEO and GEO really contribute to revenue. We use analytics tools such as Heap alongside your existing stack. It is part of our SEO and GEO service.