Blog

Campaign Attribution in Salesforce, what is it and how you can use it

Cards for Salesforce campaign influence approaches: primary campaign source, even distribution, and customizable models

Campaign attribution in Salesforce runs on Campaign Influence, and the version matters. Campaign Influence 1.0 is the Classic-only legacy that gives one primary campaign 100 percent of the credit. Customizable Campaign Influence, the current default for new orgs and available from Professional edition upward, supports multiple models, with the additional First-Touch, Last-Touch and Even Distribution models requiring Marketing Cloud Account Engagement (formerly Pardot).

The two Campaign Influence versions

Campaign Influence 1.0 is the original, and it lives in Salesforce Classic only. Its rule is blunt: one campaign is designated the primary source via the opportunity’s Primary Campaign Source field, and that campaign receives 100 percent of the attribution. Auto-association rules can add other influential campaigns to the opportunity’s record, when a contact holding a contact role on the opportunity is a member of those campaigns, but the credit itself stays undivided: the primary campaign takes all of it. Of course this is not a great method for a cross-channel approach.

Customizable Campaign Influence is the current version, running in both Classic and Lightning, enabled by default for new orgs, and available in Professional, Enterprise, Performance, Unlimited and Developer editions. It attributes a percentage of success across multiple influential campaigns, with models that can be updated manually or by automation, which is a far better fit for how B2B deals actually happen.

For B2B, where a closed deal usually sits on top of a webinar, three emails and an event, a report insisting one campaign did everything should immediately sound suspicious. It is not lying, it is answering a cruder question than the one you asked.

The models, and what needs Account Engagement

The built-in standard in Customizable Campaign Influence is the Primary Campaign Source Model: 100 percent of influence goes to the campaign named in the opportunity’s Primary Campaign Source field. It supports auto-association, and its percentages cannot be manually edited.

Three more models come with Marketing Cloud Account Engagement, the product formerly known as Pardot, and are restricted to its users. First-Touch gives 100 percent of influence and revenue to the first campaign a prospect touches. Last-Touch gives 100 percent to the last campaign touched before the deal closes. Even Distribution splits credit equally across every campaign the prospect touched. Salesforce also supports custom models, between one and five depending on edition, where users manually edit influence percentages on the opportunity’s Campaign Influence related list.

B2B marketing is way more complex than pre-defined rules. Attribution is often overlooked in B2B CRM systems and the campaign attribution by Salesforce is no different. Especially as B2B businesses rely more on ads, attribution can not be static.

Reading Salesforce attribution

What is the difference between the Campaign Influence reporting and decorative dashboards?

First, remember that influence spans only the campaigns the CRM knows about. A campaign that was never created in Salesforce, or that contacts were never added to, cannot receive credit no matter what it did in the real world. Campaign membership discipline is not admin housekeeping. It is the attribution’s raw material, and the auto-association rules depend on it entirely.

Second, watch the primary-source habit. One-campaign-gets-all models systematically flatter whichever campaign someone remembered to mark as primary, and marking habits vary by rep, team and quarter. A campaign that “outperforms” under the Primary Campaign Source Model may simply be the one that wins the paperwork race. Before celebrating, check whether the pattern survives a look at First-Touch or Even Distribution.

Third, never compare revenue-influenced figures across models without saying which model produced each number. Every model spends the same 100 percent differently, so the totals are answers to different questions. Quoting them side by side as if they were one metric is the classic Salesforce reporting sin, and it fuels most of the arguments between marketing and sales about whose number is right.

Campaign Influence for B2B pipelines

For Salesforce-centric B2B teams, Campaign Influence has one great virtue: the credit is tied directly to opportunities and revenue, in the system where the deals actually live. That makes it the natural home of B2B, lead and revenue attribution: long journeys, few conversions, revenue concentrated in a handful of deals.

A sensible rhythm: revenue influence monthly for the board, campaign performance weekly for the team. Same data, different altitude.

The disciplines that keep it trustworthy are unglamorous. Enforce contact roles on opportunities, because they are the join everything depends on: auto-association connects campaigns to opportunities through contacts holding contact roles, so missing roles silently sever campaigns from the deals they influenced. Pick one model as the primary reporting lens, document the choice, and keep it stable so trends mean something. And use Even Distribution as the standing contrast view: when a campaign’s credit collapses between models, you have learned where in the journey it actually lives.

Where Salesforce’s view ends

Campaign Influence credits recorded campaign touches. Whole categories of influence never get recorded. The brand impressions that made your emails worth opening. Dark social: the link shared in a WhatsApp thread that brought the buying committee to your pricing page. The colleague’s recommendation that started everything. None of these are campaigns in Salesforce, so all of their influence silently transfers to the recorded touches that came later. And rules-based splits cannot measure incrementality: they divide credit for deals that closed, without asking which deals would have closed anyway.

Attribution is a very delicate methodology. That said, connecting offline (events) revenue opportunity and digital ads is not easy. MarSci has developed a data-driven attribution ideal for this scenario. You can connect both online and offline consumer interactions without the need for PIIs. Reach out today and find out more.

Frequently Asked Questions

What is Campaign Influence in Salesforce?

Salesforce’s campaign attribution feature, in two versions. The legacy Campaign Influence 1.0, Classic only, gives one primary campaign 100 percent of the credit via the Primary Campaign Source field. Customizable Campaign Influence, the current default for new orgs and available from Professional edition up, splits percentages across multiple influential campaigns using attribution models.

What attribution models does Salesforce offer?

The built-in Primary Campaign Source Model, which gives 100 percent to the opportunity’s primary campaign, plus First-Touch, Last-Touch and Even Distribution, which require Marketing Cloud Account Engagement (formerly Pardot). Custom models with manually edited percentages are also supported, between one and five depending on edition. All are rules-based, so pick one primary lens and keep it stable.

Does Salesforce attribution require Pardot?

The current product name is Marketing Cloud Account Engagement, formerly Pardot, and it is required only for the First-Touch, Last-Touch and Even Distribution models. The Primary Campaign Source Model and custom models work without it, so every org with Customizable Campaign Influence gets attribution, just with a smaller model menu.

How do I make Salesforce campaign attribution accurate?

Through CRM hygiene, because that is what the attribution is made of. Add contacts to the campaigns that actually touched them, and enforce contact roles on opportunities, since auto-association joins campaigns to deals through those roles. Missing memberships and missing roles silently erase real influence from every report the system produces.

Written by

Chris Pinto

Head of Platform.

You might be interested in …