Glossary · Attribution
What is time-decay attribution?
Time-decay attribution is a multi-touch attribution model that gives every touchpoint some credit for a conversion, but more to touches that happened closer in time to the conversion.
Also called: Time decay model
Updated
How is time-decay attribution calculated?
Weight = 2^(−days before conversion ÷ half-life) Credit = conversion value × weight ÷ sum of all weights
The half-life sets how fast credit fades. With a 7-day half-life — the conventional default, and the one Google Analytics used when it offered this model — a touch a week before the purchase carries half the weight of one on the day; two weeks before, a quarter.
Time-decay attribution example
| Touch | Days before purchase | Weight | Share of $100 |
|---|---|---|---|
| Google organic | 14 | 0.25 | $11 |
| Newsletter | 8 | 0.45 | $19 |
| X | 4 | 0.67 | $28 |
| Direct | 0 | 1.00 | $42 |
The weights sum to about 2.38, so each share is its weight divided by 2.38. The touch that started the journey gets the least.
Why time-decay attribution matters
For short buying decisions — a $9 tool, an impulse e-commerce order, a limited-time promotion — the touches right before the purchase really are the most influential, and time decay reflects that while still acknowledging what came before.
When time decay misleads
- Long B2B cycles. It systematically under-credits the content and referrals that started a three-month evaluation.
- Arbitrary half-life. Results change a lot between a 3-day and a 14-day half-life; pick one that matches your typical time to convert.
- Retargeting bias. Ads shown to people already about to buy sit at the high-weight end.
How VisitTrack computes time decay
VisitTrack's Revenue tab includes time decay with a 7-day half-life alongside four other models. The touches are each paying visitor's sessions up to the purchase, weighted by how many days before the payment each session started. The Revenue tab also shows days to convert, which helps you judge whether a 7-day half-life fits your buying cycle. See revenue attribution.
Frequently asked questions
What half-life should I use for time-decay attribution?
Seven days is the conventional default. A shorter half-life suits impulse purchases; if your median time to convert is several weeks, a time-decay model will under-credit discovery and a position-based model may fit better.
Is time decay the same as last-touch attribution?
No. Last-touch gives all credit to the final touch; time decay gives every touch some credit and simply weights recent touches more heavily.
Related terms
- Multi-touch attributionMulti-touch attribution is any attribution approach that divides the credit for a conversion across several of the touchpoints that preceded it, instead of giving it all to the first or last one.
- Linear attributionLinear attribution is a multi-touch attribution model that splits the credit for a conversion equally across every touchpoint in the customer's path.
- Position-based attributionPosition-based attribution, also called U-shaped attribution, is a multi-touch model that gives 40% of the credit to the first touch, 40% to the last touch and splits the remaining 20% evenly across the touches in between.
- Last-touch attributionLast-touch attribution is an attribution model that gives 100% of the credit for a conversion to the source of the last visit before it happened.
- Attribution windowAn attribution window, or lookback window, is the maximum time between an ad interaction or visit and a conversion for that interaction to receive credit for it.
Tools and guides
See which channels actually bring paying customers
VisitTrack is cookie-free analytics with revenue attribution built in. One script tag, no consent banner, live in two minutes. 14 days free, no card required.