> For the complete documentation index, see [llms.txt](https://docs.hockeystack.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.hockeystack.com/marketing-intelligence/reports/multi-touch-attribution/attribution-models.md).
# Different Attribution Models
When building an attribution report, you have the below attribution models to choose from. This guide will illustrate all models using the example: *Google Ads Touchpoint → Facebook Ads Touchpoint → Google Ads Touchpoint → SDR Touchpoint → Closed Won ($10k)*
#### **Linear: The total value is divided and distributed equally among all touches.**
❓Use Case: Best as a starting point when you want a neutral baseline that distribute credit equally across all touchpoints.
* So, Google Ads got credits for $5,000 (0.5 of closed won), Facebook Ads got credits for $2,500 (0.25 of closed won), and SDR got credits for $2,500 (0.25 of closed won)
#### **Uniform: The total value is assigned to each touch.**
❓Use Case: Best for measuring channel reach and influence across deals. Ignores contribution - totals won't add up to revenue, so it's not useful for budget allocation.
* Note: The total attributed value doesn’t add up to the total actual value when using the Uniform attribution model.
#### **First Touch: All value is assigned to the first touch.**
❓User Case: Best for understanding what created awareness. Ignores everything the nurtured and closed the deal.
#### **Last Touch: All value is assigned to the last touch.**
❓Use Case: Best for understanding what closes deals. Ignores everything that built awareness and nurtured intent along the way.
#### **Position-Based: The first and last touches get 40% of the total value each, and the remaining 20% is distributed among the remaining touches.**
❓Use Case: Best for B2B where both pipeline creation and deal closure matter. Ignores the nuance of middle touches - they all get the same small share regardless of actual impact.
* So, Google Ads got credits for $5,000 (0.5 of closed won), Facebook Ads got credits for $1,000 (0.1 of closed won), and SDR got credits for $4,000 (0.4 of closed won)
#### **Time Decay: A**ttributing less value to a touchpoint the further away it’s from the conversion action. Specifically, the value halves every 7 days.
❓Use Case: Best for short sales cycles where recent interactions drive decisions. Ignores early-stage touches that created the awareness making everything else possible.
* For fellow nerds, the formula is v×2−i7v×2−7i (where vv is total value and ii is days from the conversion action).
* Note: The total attributed value doesn’t add up to the total actual value when using the Time Decay attribution model.
#### [Predictive](/marketing-intelligence/reports/multi-touch-attribution/predictive-attribution-model.md): HockeyStack's proprietary machine learning attribution model
### When to use Linear vs Uniform
Linear is a credited attribution model. It’s used to signify a comparison between the level of influence that each touchpoint has on the final conversion action.
Uniform is not a credited attribution model. It’s used to signify that the influence exists, regardless of its level.
Uniform is most commonly used when looking at:
* Deal amount metrics like ACV, where you need to calculate using the full deal amount, rather than the credited amount.
* The most popular touchpoints before a certain conversion action.
#### Different naming conventions
Customers have different ways of thinking of linear and uniform attribution models. Sometimes they use different wording.
**Linear** may also be called: sourced, attributed,
**Uniform** may also be called: touched, influenced.
However, to be sure of what any of the different naming conventions really mean, you need to check the definition of a specific report. We have seen customers calling both first touch and last touch 'sourced'.
### What touchpoints are included in attribution?
“Qualified touchpoints” depend on the report breakdown you select.
* If you only select web source-related breakdowns such as Source, Campaign, or UTM medium, only web sources will be included as attribution touchpoints.
* If you only select page url as the breakdown, only pageviews will be included as attribution touchpoints.
* If you select any action property as the breakdown, actions that have that action property will be included as attribution touchpoints. Ex: If you select linkedin\_campaign\_name, the only attribution touchpoints will be linkedin ad impressions/engagements. But if you select action type, since all actions have an action type, everything ever recorded will be attribution touchpoints.
* If you select a defined property as the breakdown, similar rules to the above apply, based on the mapped properties used in the defined property. However, if the defined property has “Hide values outside mappings” selected, hidden values will not be included as attribution touchpoints.
* If you select multiple breakdowns, the broadest possible touchpoint range, based on the above rules, will be included as attribution touchpoints.
If using a goal, all “qualified touchpoints” before the first instance of goal in a user/company’s journey within the selected date range are used as touchpoints.
As an example, think of a journey that has a Google Ads visit 3 months ago, a conversion 2 months ago, a Facebook Ads visit yesterday, and a conversion today. If the report date range is This Month, the Facebook Ads visit will be used in attribution, with the conversion today. If the report date range is Last 3 Months, the Google Ads visit will be the **only touchpoint** used in attribution, with the conversion 2 months ago. The Facebook Ads visit and the conversion today will be disregarded.
If using a sequence, all “qualified touchpoints” before the first instance of the “breakdown” selected sequence step within the selected date range are used as touchpoints.
All touchpoints before the first Linkedin Ad Impression.
All touchpoints before the first Deal Created after any LinkedIn Ad Impression.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:
```
GET https://docs.hockeystack.com/marketing-intelligence/reports/multi-touch-attribution/attribution-models.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
