Marketing attribution assigns credit for pipeline and revenue to the marketing touches that produced it. The model you choose matters less than the data underneath it: consistent UTMs, campaign membership recorded in the CRM, contacts attached to opportunities, and offline touches such as events captured. We build that foundation first, then a documented model, reported next to self-reported attribution so you can see where the two disagree.
What attribution covers
| Area | What it decides |
|---|---|
| Tracking | UTM conventions, source captured on every form, and the offline touches that never hit a web page, such as events and sales outreach. |
| CRM structure | Campaigns and campaign membership, and contacts attached to opportunities, so a touch can be tied to a deal. |
| The model | Which touches get credit and how much, written down so anyone can explain a number. |
| Reporting | Sourced and influenced pipeline, cost to acquire by channel, and one view that marketing and finance both use. |
| Self-reported attribution | What buyers say brought them in, captured on the form and compared with what the software recorded. |
Where attribution breaks
All the credit goes to the last click
Last-touch reporting rewards whatever the buyer clicked just before converting, usually branded search or the demo page, and hides the channels that created the demand in the first place. Budget then drifts toward the channels that harvest demand and away from the ones that make it.
Marketing and sales data never meet
Touches live in the marketing platform and deals live in the CRM, with no common thread between them. If contacts are not attached to opportunities, there is nothing to credit even when the tracking is perfect.
The model is a black box
A tool produces a number, nobody can say how, and the first time it disagrees with a sales leader's instinct it gets dismissed. A model has to be explainable to survive a budget meeting.
Nothing gets cut or scaled
Without a credible read on what works, underperforming channels keep their budget and the winners never get more, because nobody wants to move money on a number they cannot defend.
One deal, five models
Every attribution model is a rule for splitting the credit for the same deal. Switching the rule changes the answer, which is why two tools, or two people, can look at the same quarter and disagree about what worked.
- Paid search adFirst touch100%
- Blog post from a LinkedIn shareAnonymous visit, matched later0%
- Webinar registrationLead created0%
- Case study emailNurture0%
- Demo requestOpportunity created0%
- Paid search adFirst touch0%
- Blog post from a LinkedIn shareAnonymous visit, matched later0%
- Webinar registrationLead created0%
- Case study emailNurture0%
- Demo requestOpportunity created100%
- Paid search adFirst touch20%
- Blog post from a LinkedIn shareAnonymous visit, matched later20%
- Webinar registrationLead created20%
- Case study emailNurture20%
- Demo requestOpportunity created20%
- Paid search adFirst touch40%
- Blog post from a LinkedIn shareAnonymous visit, matched later20%
- Webinar registrationLead created40%
- Case study emailNurture0%
- Demo requestOpportunity created0%
- Paid search adFirst touch30%
- Blog post from a LinkedIn shareAnonymous visit, matched later5%
- Webinar registrationLead created30%
- Case study emailNurture5%
- Demo requestOpportunity created30%
None of these is the truth. First touch tells you what creates demand, last touch what closes the gap to a conversation, and the multi-touch models spread the credit by a rule somebody chose. Pick the model that answers the question you are asking, write the rule down, and keep it stable long enough to compare one quarter with the next.
The data comes first
Most attribution projects fail before a model is chosen. If a touch is not recorded, or is recorded with no way to connect it to a deal, no model can credit it. The foundation is four things:
- UTM conventions everyone follows. Paid social should not also appear as Paid Social and linkedin-ads. The convention on our campaign operations page is the same one.
- Campaigns and campaign membership in the CRM, including offline touches such as events, webinars and sales outreach that never produce a click.
- Contacts attached to opportunities with their roles, so the touches of everyone on the buying committee count toward the deal, not only the person who filled in the form.
- Anonymous visits matched to the contact once they convert, so the first touch is the real first touch rather than the form fill.
This is also the part that survives a change of tools. A clean campaign structure and complete opportunity contacts make native CRM reporting useful and a dedicated tool accurate.
Ask the buyer too
Software attribution only sees what it can track. A podcast, a recommendation from a peer, or a post someone read without clicking never shows up, and in B2B those are often what start the conversation. A free-text field on the demo form asking how they heard about you captures it. We report the two views side by side. Where they agree, you can trust the channel. Where they disagree, you have found either a tracking gap or a channel your software cannot see.
Tools, and when you need one
Many teams can report attribution from what they already own: attribution reporting in HubSpot, or campaign influence in Salesforce, on top of the structure above. A dedicated tool such as Adobe Marketo Measure, formerly Bizible, earns its cost when you have many channels, a long cycle with many touches per deal, or a need for custom models and data the CRM cannot hold. We will tell you which side of that line you are on before you buy anything.
How an engagement runs
We audit the tracking and CRM structure first, because that decides what any model can see. Then we fix the gaps, agree the model and write it down, build the reports, and run them for a full reporting period before anyone moves budget on them. A focused attribution build usually takes four to eight weeks, depending on how much history needs repairing. Reporting beyond marketing, including the forecast, is covered under revenue forecasting.
What it costs
Published 2026 market benchmarks put a focused revenue operations audit at 2,500 to 10,000 dollars, project-based work at 10,000 to 150,000 or more, and entry-level ongoing retainers at 3,000 to 8,000 a month. A focused attribution build sits toward the lower end of the project band.
Ranges as published in MergeYourData's 2026 RevOps pricing benchmarks. Market-wide figures, not our rate card. More detail in our guide to what RevOps consulting costs.
What you keep
- A written attribution method: the model, the window and the rules, so any number can be explained.
- UTM conventions and a campaign structure in your CRM that your team maintains.
- Sourced and influenced pipeline reports and cost by channel, under your admin ownership.
- A self-reported attribution field, and the report that compares it with the software view.
Frequently Asked Questions
Which attribution model should we use?
The one that answers the question you are asking. First touch shows what creates demand, last touch shows what gets people to a conversation, and U-shaped or W-shaped models spread credit across the moments that matter in a B2B cycle. Most teams use a multi-touch model for budget decisions and keep first touch alongside it, because it is the easiest to explain. The model matters less than writing it down and keeping it stable.
Do we need a special tool for attribution?
Not always. HubSpot and Salesforce can report attribution well once tracking, campaigns and opportunity contacts are clean. A dedicated tool such as Adobe Marketo Measure, formerly Bizible, is worth it when you have many channels, long cycles with many touches per deal, or a need for models the CRM cannot hold.
What is the difference between sourced and influenced pipeline?
Sourced pipeline counts opportunities where marketing produced the first touch or the lead. Influenced pipeline counts opportunities where marketing touched anyone on the deal at any point. Influenced is always the larger number, so reporting it as if it were sourced overstates marketing's contribution. We report both and label them.
Why does our attribution disagree with what buyers tell us?
Because software only sees what it can track. Word of mouth, podcasts and posts people read without clicking are invisible to it. Capturing self-reported attribution on the demo form and comparing it with the software view shows where the gaps are.
How long before attribution data is useful?
Pipeline creation shows up quickly once the tracking is fixed. What drives closed revenue takes at least one full sales cycle of clean data. Older records can sometimes be repaired, but only as far as the history allows.
Ready to fix your marketing attribution?
Book a free 30-minute review and we will look at your tracking, your campaign structure and the reports built on them, and tell you what the numbers can and cannot support today.
Take the RevOps Health Check Book a Free Review