Marketing sourced pipeline is an argument about incentives

Does this ring a bell from your Q3 QBR? Marketing presents a pipeline review slide: “We sourced 47% of pipeline this quarter.” Sales checks its dashboard and, with a raised eyebrow, finds a completely different number for the same pipeline. According to its data, most of it came from outbound and existing relationships. Meanwhile, Finance is sitting there with two confident, contradictory answers about the same set of deals.

Neither team is lying. They are both reading from systems built to prove their own value. The problem is that neither version necessarily represents how those deals actually developed.

The number depends on the rules

Marketing sourced pipeline sounds objective. It is expressed as a percentage, appears in a dashboard and can be compared from one quarter to the next. Underneath that apparently precise number sits a collection of choices:

• What qualifies as a marketing touch?
• Which system records the first interaction?
• How long is the attribution window?
• Does an event badge scan count?
• What happens if an account already engaged with content?
• Is it an existing relationship?
• Are partner interactions included?
• Does the first touch, last touch or every touch receive credit?

Change those rules and the marketing sourced number can swing significantly, with absolutely zero change to the underlying business.

That is why external benchmarks do not solve the problem. Published guidance puts a “healthy” marketing contribution anywhere from 25-45% to 30-50%. Another benchmark guide suggests 50-60% for inbound-led SaaS companies, but as little as 5-35% for enterprise and ABM motions.

These ranges are not necessarily wrong. They are measuring different GTM motions using different definitions. That is exactly the point. A benchmark cannot tell you whether your attribution model reflects reality when the definition of “sourced” changes from one company to another.

When the metric becomes the target

I once had a colleague whose compensation became tied to the number of MQLs created. The MQL threshold was immediately lowered from 50 to 20 points.

Surprise, surprise: the MQL number went through the roof! Spreadsheets turned green. Targets were met. The SDR backlog grew. Ceremonial celebrations lit up the Slack channels and bonuses were paid.

And absolutely nothing changed for the broader revenue team, which was still struggling to reach its minimum quarterly pipeline target.

Nobody needed to manipulate the data. The system simply encouraged people to optimise the metric instead of the business outcome. Marketing sourced pipeline can produce the same behaviour. If marketing is judged on how much pipeline it sources, it will naturally favour definitions that give marketing more credit. If sales is rewarded for self-generated pipeline, it will classify deals in ways that support that story.

The dashboard is not neutral. It reflects the incentives of the people who designed it. Even Forrester says sourcing metrics have become a problem. This is not a niche objection to attribution.

Forrester reported that marketing sourced pipeline and revenue appeared on 47% of B2B marketing leadership dashboards in 2020, making them the most commonly used performance indicators in its study. But Forrester’s recommendation was not to improve those sourcing metrics. It argued that B2B marketers should move away from them. By 2022 the percentage of dashboards containing marketing sourced metrics had fallen to 40%. Today, the metric remains common, but confidence in it is clearly weakening.

That makes sense. Marketing sourced pipeline is good at answering a political question: how much credit can marketing claim? It is much less useful for answering an operational question: what combination of activities helped this deal progress?

Better attribution does not fix competing incentives

The usual response is to buy a more sophisticated attribution platform. First-touch attribution is too simplistic, so we move to last-touch. Last-touch ignores everything that happened earlier, so we introduce multi-touch. Multi-touch spreads the credit across more interactions, so we build custom weightings based on channel, stage and recency.

The model becomes increasingly sophisticated, but the underlying problem remains. If marketing owns the attribution platform, sales can still distrust the answer. If sales controls the CRM definitions, marketing can still argue that important engagement is missing. If Finance receives a quarterly export from both, it still has to decide which version to believe.

A sharper attribution tool can simply give each team a more sophisticated version of its preferred number.

The wider marketing industry shows the same confidence gap. 85% of marketers felt confident in their ability to measure ROI, while only 32% actually measured it holistically across traditional and digital media.

We are much more confident in our dashboards than we should be.

Budgeting practices reveal a similar problem. 41% of marketing leaders set their annual budget using the previous year’s expenses and adjust it during the year if necessary. Only 28% built the budget from scratch around new needs and objectives.

So incomplete attribution does not merely affect quarterly reporting. It can influence where the next round of money goes.

80% of the buying process is invisible

Even the best attribution model only works with interactions it can observe. It can record email clicks, a form submission, an event scan or a sales call. It usually cannot see the internal conversations happening inside the customer’s organisation, the recommendations shared between peers, the research conducted anonymously or the informal consensus forming inside the buying group.

B2B buyers are already around 60% through their journey before contacting a vendor. And 85% have a preferred vendor in mind by the time those conversations begin.

That means a significant part of the decision has already happened before your attribution system gets a clean signal.

Trying to assign precise departmental credit after the fact creates an illusion of certainty. We are distributing 100% of the credit while observing only part of the journey.

Fix: a shared view

The answer is not to stop measuring marketing. It is to stop treating attribution as a departmental scoring system. GTM teams need a continuous, shared record of how accounts and opportunities actually move. That means bringing marketing engagement, SDR outreach, sales conversations, product activity, partner involvement and customer interactions into one account history.

Not four partial exports reconstructed at the end of the quarter.
Not separate dashboards designed to defend separate teams.
Not another model that produces a more complicated answer to the same political question.

A shared account history should help teams understand:

• What changed inside the account?
• Which people became involved?
• What signals appeared before the opportunity was created?
• What channels are sourcing pipeline & revenue?
• Which activities increased engagement or moved the deal forward?
• What deals are moving faster?
• What should the revenue team do next?

Next to cross functional analytics, this is where AI can genuinely improve GTM motions. Not by inventing an even more precise way to divide credit, but by carrying context across systems and reconstructing the sequence of events in a way that everyone can interrogate.

Marketing should be able to see what happened after engagement became a sales conversation. Sales should be able to see what shaped the account before the first meeting. Finance should be able to examine the same history instead of reconciling two competing exports.

The purpose is not to declare a winner. It is to help the revenue team understand and improve what worked.

Until GTM teams share the same view of the deal, sourced-pipeline maths will remain an argument about incentives wearing the costume of measurement.

This blog was created & posted by Jan’s AI content agent. Get in touch if you want to know how it works.

---

Jan van Zwet, AI driven Revenue Marketing
With 15 years of B2B Marketing experience at high growth companies like Amplitude, Cyberhaven and Sigma Computing, Jan helps Revenue teams increase Marketing’s contribution to qualified pipeline & revenue.