Attribution Models

What Are Attribution Models?

When a gym member signs up and makes their first payment, that revenue did not appear out of thin air. The member probably saw multiple ads, visited your website a few times, and interacted with your brand through several channels before they ever walked through the door. An attribution model determines how revenue credit gets distributed across all of those interactions.

Consider this scenario: Sarah clicks a Facebook ad for your gym's "Free Week Trial" on January 5th. She visits your website again on January 8th from a Google search ad. On January 12th, she clicks a retargeting ad on Instagram. On January 15th, she signs up for a $149/month membership. Which ad gets credit for that $149?

The answer depends on which attribution model you choose. Each model offers a different perspective on your marketing performance, and the right choice depends on what question you are trying to answer.

First Touch

First Touch attribution gives 100% of the revenue credit to the very first ad interaction that brought a contact into your world. Every subsequent touchpoint gets zero credit.

Best for: Understanding which campaigns drive initial awareness and generate new leads. If your primary goal is to fill the top of your funnel with fresh prospects, this model shows you exactly which ads are doing that job.

Example: Sarah's journey has three touchpoints: a Facebook ad click (Jan 5), a Google search ad click (Jan 8), and an Instagram retargeting ad click (Jan 12). When she pays $149, First Touch assigns all $149 to the original Facebook ad. The Google and Instagram ads get $0.

First Touch clearly identifies which campaigns bring in new people, which is critical because you cannot close a member you never reached. However, it completely ignores the rest of the journey -- retargeting and nurture sequences get zero credit even if they were essential to closing the deal.

Last Touch

Last Touch attribution gives 100% of the revenue credit to the last ad interaction before the contact converted. Everything that happened earlier in the journey gets zero credit.

Best for: Understanding which campaigns close deals. If you want to know what finally pushed someone from "interested" to "paying member," this model gives you that answer.

Example: Sarah's last touchpoint before paying was the Instagram retargeting ad on January 12th. Last Touch assigns all $149 to that Instagram ad. The original Facebook ad gets $0.

Last Touch is valuable for evaluating retargeting campaigns designed to convert warm leads. But it ignores how the lead was acquired in the first place -- a retargeting ad can only convert someone already in your pipeline. Over-reliance on Last Touch can lead to underinvesting in top-of-funnel campaigns.

Linear

Linear attribution distributes equal credit across all touchpoints in the customer journey. If a contact had five ad interactions before paying, each one gets 20% of the revenue credit.

Best for: A balanced, democratic view of your marketing funnel. It acknowledges that every interaction played some role in the conversion without favoring any particular stage.

Example: Sarah had three touchpoints. Linear splits the $149 equally: the Facebook ad gets $49.67, the Google ad gets $49.67, and the Instagram retargeting ad gets $49.67.

Linear is good for identifying campaigns that consistently appear in successful conversion paths. The downside is that it does not differentiate between an ad that sparked initial interest and one that merely reminded someone who was already going to convert.

Time Decay

Time Decay attribution gives more credit to recent touchpoints and progressively less credit to older ones. It uses exponential decay, so the most recent interaction before conversion gets the highest weight, and the weight decreases as you move further back in time.

Best for: Balancing full-journey visibility with an emphasis on the interactions closest to conversion. Ideal when you believe recent touchpoints had a stronger influence on the buying decision but you do not want to completely ignore earlier interactions.

Example: Sarah's three touchpoints are weighted by recency. The Instagram ad (3 days before payment) might get $75. The Google ad (7 days before) might get $50. The Facebook ad (10 days before) might get $24. More recent always equals more credit.

Time Decay naturally handles long sales cycles well -- an ad click from two months ago gets much less credit than one from last week, which aligns with how influence actually works. The tradeoff is that it can undervalue brand-building campaigns that plant seeds long before conversion.

Choosing the Right Model

There is no single "correct" attribution model. Each one answers a different question:

• "Where do my leads come from?" -- Use First Touch.
• "What closes deals?" -- Use Last Touch.
• "What does the full journey look like?" -- Use Linear.
• "What had the most recent impact?" -- Use Time Decay.

The best approach is to compare models regularly. If a campaign looks great under First Touch but terrible under Last Touch, it means the campaign is good at generating leads but those leads are not converting. If a campaign looks bad under First Touch but great under Last Touch, it is likely a retargeting campaign that closes well but cannot generate new demand on its own.

How Weights Are Computed

Attribution weights always sum to 1.0 (100%) for any given transaction. Revenue is never double-counted across touchpoints. When you view campaign performance in the dashboard, the revenue figures represent each campaign's weighted share based on the selected model.

Weights are computed on-the-fly when you load the dashboard or change filters. They are not stored in the database. Switching between models, adjusting date ranges, or adding new data does not require any reprocessing -- you always see fresh calculations based on the latest data.

When a contact has only one touchpoint, all four models produce the same result: 100% credit to that single interaction. The models only diverge when a contact has multiple touchpoints in their journey.

Related Pages

• Matching Engine -- how Adsu identifies contacts across data sources before attribution can happen
• Meta CAPI Feedback -- how attribution data feeds back into Meta to improve ad targeting
• Dashboard Overview -- where you see attribution data in action