Attribution & Measurement

Salesforce multi-touch attribution: Campaign Influence vs. a warehouse model

Campaign Influence credits revenue to campaigns through campaign members and contact roles. Here is where the native models stop, and how to build an auditable model in the warehouse.

Founder of MetricMaven. Previously managed Google Tag Manager and Google Analytics across 800+ sites at HCA Healthcare, and built attribution and eCommerce reporting at RYZE Superfoods and Mars.

An opportunity can have several campaign interactions behind it. Salesforce Campaign Influence and a warehouse model offer different ways to relate those interactions to the outcome. Neither can recover campaign membership or contact roles that were never recorded.

How Salesforce Campaign Influence works

Customizable Campaign Influence identifies revenue share using standard and custom attribution models. It is available in Salesforce Professional, Enterprise, Performance, Unlimited, and Developer editions, and it adds the CampaignInfluence and CampaignInfluenceModel objects.

Influence models scan active campaigns for members who are also assigned a contact role on an open opportunity. A matching relationship can create a campaign influence record.

Salesforce says every campaign member counts regardless of member status. After an opportunity is closed, won or lost, new influence records are no longer created.

The included Primary Campaign Source model assigns 100% of influence to the campaign in the opportunity's Primary Campaign Source field. Salesforce creates these records through auto-association, so users can't add them manually.

Auto-association can be limited by a time frame and by campaign types, both set in Setup. That setting applies only to the standard models.

The influence models available

ModelHow it assigns creditAvailability
Primary Campaign Source100% to the campaign in the Primary Campaign Source field, created by auto-associationDefault for Customizable Campaign Influence
CustomUsers enter influence percentages on the opportunity, or through the API when the model is lockedNumber of custom models depends on edition, for example 3 on Professional and Enterprise and 5 on Performance and Unlimited
First-Touch100% to the first campaign, based on campaign member Created DateAccount Engagement (formerly Pardot) users only
Even-DistributionSplits credit equally across the campaigns touchedAccount Engagement users only
Last-Touch100% to the last campaign before close, based on campaign member Last Modified DateAccount Engagement users only
Primary Campaign Source and custom models work without Account Engagement. The other three do not.

Where native Campaign Influence falls short

  • Data hygiene decides the result. If the opportunity has no contact role for a person, or the person is not a campaign member, that campaign gets no influence. Missing records fail silently.
  • Leads that never convert don't link to opportunities. When a lead converts, the rep creates a Contact, and optionally an Account and an Opportunity. Anything that happened on the lead before that point depends on the conversion carrying it forward.
  • Campaign Influence only knows campaigns. Website sessions, ad clicks, and organic visits do not appear unless someone turns them into campaign members. Most paid and organic traffic never gets that step.
  • Ad spend lives outside Salesforce. Influence can tell you which campaign was touched. It cannot put that campaign next to what it cost in Google Ads or Meta without a second system.

When to build multi-touch attribution in a warehouse

Consider a warehouse model when the reporting question needs website or ad touches beyond Salesforce campaign records, or when the attribution rules need to be reviewed across systems.

Campaign cards and contact-role folders connect to one opportunity record, while unmatched records sit in a separate tray.
  • You need first-touch, last-touch, or even-distribution credit without an Account Engagement license.
  • Website sessions and paid ad clicks should count alongside campaigns, not only the campaigns a rep entered.
  • Finance or leadership needs a model whose rules someone can read, question, and change.
  • Spend and won revenue need to sit in the same report with the same definitions.

Building Salesforce attribution in BigQuery

The pipeline has five steps: sync the Salesforce objects, keep the membership data complete, apply the attribution logic to won opportunities, add the web and ad touches, and check the output against native reporting.

1. Sync Salesforce into BigQuery

Load these objects into BigQuery on a schedule: Opportunity, OpportunityContactRole, CampaignMember, Campaign, Lead, and Contact. Maven or any other connector can do this. The table names in this guide are illustrative snake_case names such as salesforce_opportunity, so match them to whatever your sync creates.

2. Keep campaign membership and contact roles complete

The model is only as complete as the membership data it reads. Every form, list import, and event that creates a lead should also add that person as a campaign member, and every opportunity should have at least one contact role. Those rules belong in the process, not in a cleanup project each quarter.

UTMs and click IDs are the other gap. Capture them when the lead is created so the web touch can be matched later. The offline conversions guide covers how to capture them at that point.

3. Apply attribution models to won opportunities

The query below walks from each won opportunity to its contact roles, then to those contacts' campaign memberships created on or before the close date. It orders each opportunity's touches by campaign member created date, the same ordering the native First-Touch model uses. It then computes three models from that order: first touch, last touch, and even distribution.

first-touch, last-touch, and even-distribution revenue by campaign
WITH touches AS (
  SELECT
    o.id AS opportunity_id,
    o.amount,
    cm.campaign_id,
    ROW_NUMBER() OVER (PARTITION BY o.id ORDER BY cm.created_date) AS touch_n,
    COUNT(*)     OVER (PARTITION BY o.id)                          AS touches
  FROM salesforce_opportunity AS o
  JOIN salesforce_opportunity_contact_role AS ocr
    ON ocr.opportunity_id = o.id
  JOIN salesforce_campaign_member AS cm
    ON cm.contact_id = ocr.contact_id
   AND DATE(cm.created_date) <= o.close_date
  WHERE o.is_won
)
SELECT
  c.name AS campaign_name,
  c.type AS campaign_type,
  SUM(IF(t.touch_n = 1, t.amount, 0))         AS first_touch_revenue,
  SUM(IF(t.touch_n = t.touches, t.amount, 0)) AS last_touch_revenue,
  SUM(t.amount / t.touches)                   AS even_distribution_revenue
FROM touches AS t
JOIN salesforce_campaign AS c
  ON c.id = t.campaign_id
GROUP BY campaign_name, campaign_type
ORDER BY even_distribution_revenue DESC

Each model column sums to won revenue with at least one matching campaign touch. Inner joins leave out won opportunities with no touches, so that revenue appears as a gap rather than a zero row.

Report that gap instead of hiding it. The table is computed after the close, and the query exposes the contact-role join, close-date cutoff, and ordering rules.

4. Add web and ad touches

To include web and ad activity, union GA4 sessions or ad clicks into the touch list. Join them through UTMs or click IDs captured on a lead or contact, using explicit identity rules.

Carry a campaign name, source, and timestamp for each touch so the ordering is consistent. The SQL above is where the additional source rows can be added.

5. Report and sanity-check

Compare the warehouse output with native Campaign Influence reports for the same period. Explain differences rather than smoothing them over.

Possible causes include missing opportunity contacts, campaign members created after close, web touches that do not match a contact, or time-zone differences. Record the relevant cause alongside the report.

Before you trust the numbers

  1. Every lead-creating form, list import, and event adds the person as a campaign member.
  2. Every opportunity has at least one contact role, and won and closed flags are current.
  3. Duplicate leads and contacts are merged, so one person does not count under two records.
  4. Won revenue with at least one touch plus won revenue with none equals total won revenue, and the gap is documented.
  5. The warehouse output is compared with native Campaign Influence for the same period, and every difference has a stated cause.

For the wider model, read the GA4 multi-touch attribution guide for BigQuery, then the explanation of why Salesforce and GA4 numbers differ. For the underlying models and their trade-offs, the multi-touch attribution technical guide covers them in more depth. Want it done for you? MetricMaven can build the Salesforce model against your objects.

Frequently asked questions

Does Salesforce have multi-touch attribution?

Salesforce has Customizable Campaign Influence, which splits opportunity revenue across campaigns. Primary Campaign Source is the default and gives 100% to one campaign. First-Touch, Even-Distribution, and Last-Touch are limited to Account Engagement users, and custom models use percentages entered on the opportunity.

What is Customizable Campaign Influence?

It is a Salesforce feature that identifies revenue share using standard and custom attribution models. It adds the CampaignInfluence and CampaignInfluenceModel objects and is available in Professional, Enterprise, Performance, Unlimited, and Developer editions.

Why are my Salesforce campaigns getting no influence?

Usually one of three things is true. The person is not a campaign member, the opportunity has no contact role for that person, or the opportunity is already closed, so no new influence records are created. Check campaign membership and contact roles first.

Why doesn't Salesforce campaign influence match GA4?

Campaign Influence only counts campaign members linked to open opportunities. GA4 counts website sessions, events, and conversions under its own attribution rules. Organic visits and ad clicks show up in GA4 but not in Salesforce unless someone turns them into campaign members.

Can Maven help with Salesforce attribution?

It helps with the inputs and the reporting. Maven’s tracking audit checks each client’s Google Tag Manager for broken or stalled tags and events, and tracking health checks run daily. With Salesforce and the ad platforms connected to the client’s workspace, you can ask for spend, conversions and revenue through the MCP and build the report from the answer.

Sources and references

Jamie Isabel

Founder of MetricMaven. Previously managed Google Tag Manager and Google Analytics across 800+ sites at HCA Healthcare, and built attribution and eCommerce reporting at RYZE Superfoods and Mars.

Updated October 16, 2026

Start your 14-day trial

Set up your first client, build their report by asking Claude or ChatGPT, and publish it to their portal.

  • Free for 14 days, no credit card
  • Your first client workspace and portal
  • Every report template, ready to clone
Start a 14-day trial

Nothing is charged unless you subscribe.

Prefer a walkthrough? Book a demo on one of your own clients.