Future InTech
Language · English

Measurement and interpretation · FIELD GUIDE

First-click vs last-click attribution: compare the same conversion

Hold conversion definition and date range fixed before interpreting different channel credit.

By Future InTech · Published
Evidence cutoff: 2026-11-03T21:15:50.611897+00:00 · UTC

What the evidence can establish

An attribution model assigns credit; it does not change the historical order of interactions. Umami’s documentation distinguishes first-interaction credit from final-interaction credit and requires a conversion step. Before calling two dashboards inconsistent, check whether they answer the same credit question and use the same conversion definition. This is useful when comparing an AI referral with earlier discovery or a later return visit.

Hypothetical journey: a visitor first arrives through a search campaign, later follows an AI referral, then returns through an email link and submits a request. First-click and last-click models give credit to different endpoints of that sequence. Neither model, on its own, proves that the middle interaction was useless or that any channel caused the conversion.

Fix the conversion definition; Order observed interactions; Compare first and last credit; State the model in the report
Original explanatory diagram based on the dated sources below. Any example values are hypothetical; this is not a screenshot, customer measurement or AI platform result.

The saved evidence used here

This guide interprets dated official documentation. It does not report traffic from a customer or a measured AI answer. Each material claim is mapped to a source below.

c1 · Explanation and verification steps
Umami first-click attribution gives full credit to the first interaction. Source 1
c2 · Explanation and verification steps
Last-click attribution gives full credit to the final interaction before conversion. Source 1
c3 · Explanation and verification steps
A conversion step is required for the attribution insight. Source 1

The diagram explains the investigation or a clearly labeled hypothetical example. No measured counts or platform outcomes are added.

AI answer visibility: Not measured. No original platform answers or verified citation records were collected for this guide. A request count is not an audience, ranking or citation count.

What to do next

  1. Write down the exact conversion step: a specified page view or event. Record the date range, filters and attribution model beside the result.
  2. Construct a permitted test journey with distinguishable source labels and a known sequence; keep it out of customer performance reporting.
  3. Run both model views with the same conversion step and scope. Identify which interaction each model selects before comparing totals.
  4. When reporting channel performance, name the credit model and note missing journey coverage. Keep causal claims separate from model-assigned credit.

How to check the result

The comparison should reproduce the selected first and last interaction for a known journey, use one conversion definition and disclose any tracking gaps. A change in assigned channel credit should not be presented as a change in observed conversions.

Limits of this conclusion

Documentation defines the model; it does not verify your journey coverage. Cross-device gaps or missing referral evidence must remain unresolved unless separately established.

This is a saved, dated edition. Source documentation can change. This edition retains the original retrieval dates and claim mapping; follow the sources before applying version-specific guidance.

Sources and publication checks

Automated checks validate source availability, supported fields, duplicate wording, numeric figures and page configuration. This guide was prepared with AI assistance and checked against saved source passages in the same authoring session. The batch importer checks source fingerprints, quotation locations, wording overlap and required fields. No paid generation API or separate model-review pass was used for this batch. These checks can miss errors and are not expert or human review.

Attribution

Obtained: 2026-11-03T21:15:50.611897+00:00
Location: Models # / b4; Models # / b5; Parameters # / b8

Saved source fingerprint9cdb05c7a30aa1797d988fd8e750b91a0b3794872b4f41720f8cdd57b459ddb2

Download this edition's evidence and action record (JSON)