B2B sales attribution is how you decide which team created each opportunity in your pipeline. At The GTM Engineering Company, we sort every deal into one of four categories: pure sales, pure marketing, sales-sourced with marketing influence, and marketing-sourced with sales influence. Each record gets tagged when it enters the CRM and tagged again when it becomes an opportunity, timestamped at the first meeting booked. Both teams then read the same numbers.
This guide walks through why sales attribution turns into an argument, the two pipeline definitions you need, and how we build the model in HubSpot or Salesforce. It ends with where multi-touch attribution models fit once the basics hold up.
Key Takeaways (TL;DR)
Most attribution arguments trace back to how SDRs, AEs and marketers get paid. Fix the comp plan before you rebuild the report.
In most deals we work on, one of four categories fits: pure sales, pure marketing, sales-sourced with marketing influence, or marketing-sourced with sales influence.
Tag twice. Once when the contact or company is created, and again at opportunity creation, timestamped at the first meeting booked.
HubSpot’s deal create and revenue attribution reports need Marketing Hub Enterprise. The four-category model runs on Professional with deal properties and workflows.
Add a multi-touch model only after 3 to 6 months of clean four-category data.
Sales Attribution: at a Glance
Question | Short answer |
|---|---|
What it measures | Which team created each opportunity, and which team touched the account before that |
Why it causes fights | Commission plans and budget defense, more often than missing data |
The model we use | Four categories: pure sales, pure marketing, sales-sourced with marketing influence, marketing-sourced with sales influence |
When to tag | At record creation, then at opportunity creation, timestamped at the first meeting booked |
Marketing-sourced benchmark | 25% to 45% of pipeline, median 35% (Dreamdata 2026, via GrowthSpree) |
Marketing-influenced benchmark | 60% to 85% of pipeline, median 72% (same source) |
HubSpot tier | Contact create attribution on Marketing Hub Professional; deal create and revenue attribution on Enterprise only |
When to add multi-touch | After 3 to 6 months of clean four-category data |
Why B2B sales attribution has never been solved
Sales attribution is one of the oldest open problems on a B2B revenue team. We see two reasons it stays open.
First, a lot of what moves a buyer never reaches the CRM. A founder hears your name on a podcast, or a peer recommends you in a private Slack group, and neither leaves a UTM. Buyers also do much of their research alone: in Gartner’s 2025 B2B buyer survey, 61% of B2B buyers said they prefer a rep-free buying experience.
Second, a buying decision usually has several causes, and they feed each other. Forrester’s State of Business Buying 2024 found that an average of 13 people take part in a purchase decision, and 89% of purchases involve two or more departments. Picture a VP of Sales who ignores three cold emails, sees a retargeting ad, reads a case study her CFO forwarded, and then books a demo through your website. Which of those created the deal? Each one made the next one more likely to work.
So no model will capture 100% of what influenced a purchase. The goal we set with clients is smaller and more useful: a split that sales and marketing both accept as fair, built only on data the CRM can capture. Once both teams trust the split, they stop auditing each other’s numbers and get back to the pipeline.
Why attribution fights are political
From what we see at client companies, most attribution fights are political. Department heads want to walk into the leadership meeting with the bigger pipeline number, because that number protects their headcount and their budget. The argument about data is usually standing in for that.
In most companies we work with, the fight starts with commissions. Look at how each person gets paid and the positions make sense:
SDRs are usually paid on qualified meetings, so they want every meeting they touched to count as theirs.
Some plans pay AEs a different rate on deals they sourced themselves, which gives them a reason to claim self-sourced pipeline.
Marketing wants credit for every lead it brings in through paid, brand, content and LinkedIn, because it defends that spend every quarter. Attributed pipeline is the ROI evidence finance asks for, and it is also how the team gets recognized for the work.
Each person in that list is responding rationally to the plan they signed.
Marketing’s pressure shows up in the research too. In the Gartner 2025 CMO Spend Survey, budgets held flat at 7.7% of company revenue and 59% of CMOs said they lack the budget to execute their strategy. A 2024 Gartner survey of senior marketing leaders found only 52% could prove marketing’s value and receive credit for it, and CMOs named the CFO (40%) as the executive most doubtful of that value.
That is why a new report rarely ends the argument. If commissions still reward whoever wins the attribution call, people will keep fighting over the call. Change the incentive first, then build the report. The company hits or misses its number as one revenue team, and the comp plan should say so before any dashboard does. Our guide on how to structure revenue operations covers who should own that shared number.
Marketing-sourced vs marketing-influenced pipeline
Pipeline attribution needs two definitions that most teams blur.
Marketing-sourced pipeline is every opportunity that marketing created, whatever sales outreach came before it. In our model, that covers pure marketing deals and deals that are marketing-sourced with sales influence.
Marketing-influenced pipeline is every opportunity where marketing touched the account during the buying cycle, whichever team created it. It includes everything marketing sourced, plus sales-sourced deals with marketing influence.
Influenced is always the bigger number. GrowthSpree’s 2026 benchmarks, citing Dreamdata data, put marketing-sourced pipeline at 25% to 45% of total pipeline (median 35%) and marketing-influenced at 60% to 85% (median 72%).
Deal size moves the sourced share a lot. In the same GrowthSpree data, marketing sources 55% to 75% of pipeline below $10,000 ACV and 15% to 28% at $200,000 ACV and above. That matches what we see: the bigger the contract, the more often an SDR or AE opened the conversation.
One caveat before you compare your own numbers. GrowthSpree defines marketing-sourced by the first known touchpoint. We define it by the action that created the opportunity. An account that got three cold emails and then filled out a demo form after an ad counts as marketing-sourced in our model and as sales-sourced under a first-touch rule.
Treat the benchmarks as a sanity check. If marketing sources 80% of pipeline on a $150,000 ACV, look at the tagging before you celebrate.
The four-category sales attribution model
The model we have seen work best is a hybrid. It records how each record entered the CRM, then checks again what created the opportunity, so the sales attribution answer comes from two tags instead of one. It came out of client work, and in most deals one of these four labels fits.
Category | What happened | Example |
|---|---|---|
Pure sales | An SDR or AE hunted the account, booked the meeting and qualified it, with no marketing touch on record | A cold email sequence leads to a booked demo |
Pure marketing | The lead filled out a form, replied to a marketing email or clicked an ad, and nothing else happened | Someone requests a demo after clicking a paid ad |
Sales-sourced with marketing influence | The CRM shows that sales booked the meeting, and it also shows marketing running ads, email or LinkedIn campaigns to the account | An SDR books a call with an account that has been seeing your LinkedIn ads for a month |
Marketing-sourced with sales influence | Sales sent cold emails and made cold calls to several people at the account but did not book; the lead later came in through an ad or an inbound workflow | An account gets three cold emails, then fills out a form after an ad |
Two of the four categories give both teams credit, and that is deliberate. The SDR who books a meeting with an account marketing has been warming up keeps the sourced credit, and marketing keeps the influence. The same holds in reverse when cold outreach softens an account that later converts through a form.
With four options, the conversation moves away from who brought the lead. Teams start working on closing it, and on bringing in better leads next quarter.
The two tagging moments
The model depends on deal level attribution captured at two points.
Record creation. When a contact or company enters the CRM, store how it got there (form, list import, sequence, enrichment or manual entry), which team created it, and which user or workflow did it.
Opportunity creation. When the record becomes an opportunity, store which action led to it, which team took that action and who on that team. Timestamp this at the first meeting booked.
Why the first meeting booked? Deal records get created by workflows, imported in bulk or backdated during cleanups, so their create date is unreliable. A booked meeting is a person agreeing to a time, and both teams already treat it as the start of a real opportunity.
How to set up multi-touch attribution in HubSpot
Start with what HubSpot gives you at each tier, because the tier decides what you build. According to HubSpot’s attribution documentation, contact create attribution is available on Marketing Hub Professional and Enterprise. Deal create attribution and revenue attribution are Marketing Hub Enterprise only, and those are the reports that carry the W-shaped and full-path models.
On Professional, then, native multi-touch revenue attribution is not available. The four-category model is how we handle that. It is a deal-level model built from properties, workflows and standard reports, and it runs on Professional.
Add two properties to contacts and companies, creating team and creation method, and fill them with a workflow when the record is created. HubSpot’s own Original Source property already records the channel, so leave it alone.
Write one UTM naming convention for source, medium and campaign, and use it on every paid, email, LinkedIn and event link. Keep the convention in a shared sheet that marketing owns.
Make sure sequences, calls and emails log to the same contact and company records. If your sequencer sits outside HubSpot, check that it writes activities and replies back.
When the first meeting is booked, have a workflow stamp the action that led to it, the team and the person.
Store one of the four categories in a deal property with a date stamp. Set the workflow to fill the property only when it is empty, so a later automation cannot overwrite it.
Report pipeline, win rate and sales cycle length by category, using the same sales pipeline stages both teams already work from.
None of this needs custom code. The hard part is the data under step 1. If company records are missing domains or industries, the creation tags come out patchy. Teams that connect HubSpot to Clay for enrichment usually fix that gap before they switch on the workflows.
HubSpot attribution reporting by tier
HubSpot report | Subscription needed | Models available |
|---|---|---|
Contact create attribution | Marketing Hub (or Content Hub) Professional or Enterprise | First interaction, last interaction, linear, U-shaped, J-shaped, inverse J-shaped, time decay |
Deal create attribution | Marketing Hub Enterprise only | Adds W-shaped |
Revenue attribution | Marketing Hub Enterprise only | Adds W-shaped and full path |
Source: HubSpot Knowledge Base.
For a Series A to C company, the move to Enterprise is a real budget decision. We recommend running the four-category model on Professional first. Upgrade when you have a question only revenue attribution can answer, such as which campaigns show up most often in closed-won deals.
Setting up deal-level attribution in Salesforce
The logic carries over to Salesforce. Only the parts have different names.
Create a custom picklist field on the Opportunity object with the four categories, plus a date field for when the value was set. Then build a record-triggered Flow that fills the category when the first meeting is booked, based on which team logged the qualifying activity, and only if the field is blank.
Before the Flow assigns a sales-sourced value, check Campaign Influence. If a campaign reached the account’s contacts before the meeting, the deal is sales-sourced with marketing influence. One catch: Salesforce only counts a contact’s campaign involvement when that contact is added as a Contact Role on the opportunity. If reps skip contact roles, the influence check comes back empty and every deal looks like pure sales.
If you are still choosing a CRM, our HubSpot vs Salesforce comparison covers how much customization each kind of motion needs.
Where B2B marketing attribution models fit
Multi-touch attribution models answer a different question. The four categories tell you which team created a deal. Multi-touch models spread credit across channels and campaigns, which is what marketing needs for budget calls. These are the attribution models HubSpot offers, with the weights from its documentation.
Model | What it credits | Where it fits |
|---|---|---|
First interaction | 100% to the first touch | Finding which channels bring in new names |
Last interaction | 100% to the last touch before conversion | Short cycles with one or two touches |
Linear | Equal credit to every interaction | Mid-length cycles where no single touch stands out |
U-shaped | 40% to the first interaction, 40% to lead conversion, 20% split across the rest | Teams judged on lead generation |
W-shaped | 30% to the first interaction, 30% to contact creation, 30% to the last interaction that created the deal, 10% across the rest | B2B cycles with a clear lead-to-opportunity handoff |
Full path | 22.5% each to first interaction, lead creation, deal creation and last interaction, 10% across the rest | Long enterprise cycles with many stakeholders |
Time decay | More credit to recent touches, on a 7-day half-life | Short, campaign-led cycles |
Our verdict: add a multi-touch model on top of the four categories once you have 3 to 6 months of clean data. Run one on messy tagging and you get a more detailed version of the same wrong answer.
Common sales attribution mistakes
Most broken attribution setups we audit fail in the same places.
Lifecycle stages that move backward. When a contact drops from SQL back to lead, the date it first hit a milestone can disappear. Understory recommends custom date properties that never get overwritten, and we build the same thing.
A workflow built for an unrelated job resets a source property. Nobody gets an alert, and months of tags change quietly.
UTMs written two ways, such as “linkedin” and “LinkedIn_Ads”, which then report as separate sources.
Treating tracked conversions as complete. Understory reports that HubSpot typically records 20% to 40% fewer conversions than the ad networks report for the same campaigns, because of consent blocking and browser privacy features.
Paying commissions on a model nobody trusts. That only moves the fight from the deal to the report.
Most of these come down to CRM data hygiene. One result in our Spara HubSpot case study was 100% of CRM data cleaned and deduplicated, and that is the kind of base attribution tags need to hold up.
Everything You Need to Know About Sales Attribution
Topic | What to know |
|---|---|
Definition | Sales attribution decides which team created each opportunity and which team influenced it |
Why it is hard | Dark social and word of mouth never reach the CRM, and most deals have several connected causes |
Buying groups | 13 people per purchase on average; 89% of purchases span two or more departments (Forrester 2024) |
Root of the fights | Commission plans for SDRs and AEs, and marketing’s need to defend budget |
Sourced vs influenced | Sourced means marketing created the opportunity; influenced means marketing touched the account at any point |
Benchmarks | Marketing-sourced 25% to 45% (median 35%); influenced 60% to 85% (median 72%), Dreamdata 2026 via GrowthSpree |
Deal size effect | Sourced share runs 55% to 75% below $10,000 ACV and 15% to 28% at $200,000 and above |
The four categories | Pure sales, pure marketing, sales-sourced with marketing influence, marketing-sourced with sales influence |
Tagging moments | Record creation, then opportunity creation, timestamped at the first meeting booked |
HubSpot | Four-category model runs on Professional; deal create and revenue attribution need Marketing Hub Enterprise |
Salesforce | Custom opportunity field, record-triggered Flow, Campaign Influence check with contact roles |
Multi-touch models | U-shaped is 40/40/20; W-shaped and full path sit in Enterprise reports; add after 3 to 6 months of clean data |
Biggest data risks | Lifecycle reversals, overwritten source fields, inconsistent UTMs, 20% to 40% of conversions lost to consent blocking |
Build a sales attribution model both teams accept
The GTM Engineering Company builds attribution inside your CRM, in weekly working sessions with your team, so you see every property and workflow as it goes in. Every workflow ships with a written SOP and a Loom walkthrough, which means your RevOps lead can audit and change the model after the engagement ends. We are members of Clay’s Expert Program, and we connect enrichment, scoring and attribution into one system in the CRM you already pay for.
This fits VC-backed B2B SaaS teams from post-seed to Series C, on HubSpot or Salesforce, where sales and marketing argue about pipeline credit every quarter. Engagements run 3 or 6 months at $5,000 to $7,000 per month, with a CRM audit included. By day 30 you have a prioritized fix list, a working enrichment table and at least one live signal flowing into your CRM. Our breakdown of fractional RevOps cost shows how that compares with a full-time hire.
If your next board deck needs a pipeline split that both teams sign off on, start the tagging this month so it has a full quarter of data behind it.
FAQs About Sales Attribution
What is B2B sales attribution?
B2B sales attribution assigns each opportunity to the team that created it and records which other team influenced it. The GTM Engineering Company uses four categories: pure sales, pure marketing, sales-sourced with marketing influence, and marketing-sourced with sales influence. Records are tagged when they enter the CRM and again at opportunity creation, timestamped at the first meeting booked.
How is marketing-sourced pipeline different from marketing-influenced?
Marketing-sourced pipeline differs from marketing-influenced pipeline in that sourced counts only opportunities marketing created, while influenced counts every opportunity marketing touched, whichever team created it. Influenced is always the larger number. GrowthSpree’s 2026 benchmarks, based on Dreamdata data, put marketing-sourced at 25% to 45% of pipeline (median 35%) and marketing-influenced at 60% to 85% (median 72%).
How do I set up multi-touch attribution in HubSpot?
To set up multi-touch attribution in HubSpot, you need Marketing Hub Enterprise for deal create and revenue attribution reports. Marketing Hub Professional includes contact create attribution only, with models such as linear and U-shaped. On Professional, build a four-category deal model instead: tag how each record entered the CRM, standardize UTMs, log sales activity on the same records, tag again at opportunity creation and store the category in a deal property.
Do I need HubSpot Enterprise for attribution reporting?
You need HubSpot Marketing Hub Enterprise for attribution reporting on deals and revenue, including the W-shaped and full-path models. Contact create attribution works on Marketing Hub Professional. A four-category deal-level model also runs on Professional, using custom deal properties, workflows and standard reports. Most Series A to C teams can start there and upgrade after 3 to 6 months of clean data.
What is sales attribution?
Sales attribution is the process of deciding which team or touchpoint gets credit for an opportunity or a closed deal. In B2B it usually works at two levels: deal-level attribution says whether sales or marketing created the opportunity, and multi-touch attribution splits credit across campaigns. GrowthSpree’s 2026 benchmarks show marketing sourcing a median 35% of B2B pipeline and influencing a median 72%.
What is an attribution model used for in marketing?
An attribution model is used in marketing to decide how much credit each touchpoint gets for a conversion, so budget can move toward the channels that create pipeline. Each model splits credit differently. HubSpot’s U-shaped model gives 40% to the first interaction, 40% to lead conversion and 20% to everything in between, while first interaction gives 100% to the first touch. Pick the model after deciding whether you need channel budget answers or team credit answers.
What data does sales attribution rely on?
Sales attribution relies on CRM data: how each contact and company was created, UTM parameters on marketing touches, logged sales activity such as calls and sequences, and the opportunity record with its create date. If any of those is missing or overwritten, the model credits the wrong team. Tracked data is also incomplete by default, since Understory reports that HubSpot typically records 20% to 40% fewer conversions than the ad networks report because of consent blocking and browser privacy features.
Is sales attribution worth building if it can never be fully accurate?
Sales attribution is worth building even though it can never be fully accurate, because the goal is a split both teams accept, and that needs far less data than a perfect one. Podcast mentions and private referrals will never show up in a CRM. A four-category model only needs two tags per deal, one at record creation and one at opportunity creation. Once it runs on 3 to 6 months of clean data, you can add a multi-touch model on top.
