Home / Marketing Attribution
Marketing Attribution

Marketing Attribution Models Explained

An attribution model is a rule for assigning credit to marketing interactions. Different models answer different questions, so changing the model can change the apparent performance of a campaign.

Models in CallRail Reporting

CallRail currently documents five cost-per-lead attribution models in its reporting: First Touch, Lead Creation, 50/50, W-Shaped, and Qualified.

Choose the Model Around the Decision

Use a model that reflects the question you are trying to answer rather than selecting whichever model produces the most flattering report.

First Touch

First-touch attribution gives credit to the first known interaction. It is useful for understanding discovery, but it can understate later interactions that moved the prospect toward action.

Lead-Creation and Milestone Models

A lead-creation model focuses on the interaction that produced the first recognized lead event. Milestone approaches can split or shift credit based on stages such as first touch, lead creation, and qualification.

CallRail's Current CPL Models

CallRail currently documents First Touch, Lead Creation, 50/50, W-Shaped, and Qualified models for cost-per-lead reporting. These models distribute marketing cost differently across milestones, so the same campaign can look different depending on the selected model.

Do Not Shop for the Prettiest Report

Choose a model before looking for a winner. If the business question is “What introduces new demand?” first touch may be appropriate. If it is “What produces qualified leads?” a qualification-oriented view is more relevant.

Compare Models Side by Side

When a decision is important, view the same period through more than one defensible model. Large differences are a signal that the customer journey contains meaningful interactions at several stages—not necessarily that one report is wrong.

Keep Model Changes Visible

If the organization changes its default attribution model, document the date and rationale. Otherwise historical comparisons can appear to show a performance shift that was actually caused by a reporting-rule change.