B2B Attribution Playbook: The Lead Source Chain
Attribution breaks somewhere between the ad click and the CRM. This playbook shows the seven links of the lead source chain, the fields to capture, and what to do when nobody fills in the form.
Attribution in B2B marketing was never really about dashboards. It is about whether anyone can say, six months later, which campaign produced the opportunity that just closed.
Most teams cannot. Not because they lack tools, but because the journey between marketing, sales and revenue is broken in a specific place. A B2B decision runs for months rather than days, it moves through a committee rather than a person, and most of the research happens where no tracking code is watching. By the time sales is having a conversation, the UTM has gone, the CRM field is empty, and the ad platform, GA4 and the CRM are each telling a different story.
I wrote The Lead Source Chain for the people who have to answer for that gap: demand generation and performance marketers, marketing ops, RevOps, and the sales and leadership teams who end up arguing about the numbers.
It is not a case for perfect attribution. Perfect attribution was never available. It is a case for building enough connected evidence to make a budget decision you can defend.
Where attribution actually breaks
The chain runs from the ad click, through UTM and click-ID capture, into the landing page and the tracked action, then into the CRM record, the sales status, the opportunity, and finally back out to the ad platforms as an optimisation signal. Seven links.
Most companies do not fail because they are missing every link. They fail because the links are not connected — and the break is almost always in the same place, at the handoff into the CRM. Marketing can prove volume. It cannot prove quality, pipeline or revenue, because the fields that would prove it are owned by someone who was never asked to fill them in.
What the playbook covers
The first half is the plumbing: the seven links of the chain, the minimum set of CRM fields worth capturing at conversion and who owns each one, why UTMs still matter and what they were never able to do, and what privacy-aware measurement genuinely changes.
The second half deals with the part most attribution content avoids — the journeys you cannot see. There are two honest answers to those, and the playbook covers both. You can ask the buyer directly, with one open question on high-intent forms and the same question asked by sales in the first call, then read those answers against your tracking to find the channels your analytics has never heard of. And when neither method can settle it, you stop attributing and start testing: geo holdouts, brand and non-brand splits, channel pauses, mix modelling once the spend justifies it.
It closes with the practical material — a minimum CRM agreement between sales and marketing, the constraints that catch teams out when they start uploading qualified events back to ad platforms, a reusable AI audit prompt, a 30-day implementation plan, and a ten-point checklist you can run through before blaming the platform or the campaign team.
One correction worth making
A lot of attribution advice still assumes the cookieless web arrived. It did not. Google abandoned third-party cookie deprecation in Chrome in July 2024, dropped the replacement choice prompt in April 2025, and retired most Privacy Sandbox APIs from October 2025 onward. Safari and Firefox still block cross-site tracking by default, and consent requirements have not changed at all.
So the pressure on your measurement is real, but it is not the pressure most people are preparing for. It comes from long buying cycles, buying committees, consent gaps and research that happens somewhere you cannot instrument — not from a cookie apocalypse that never happened. Safari alone caps a JavaScript-set cookie at seven days, and drops that to 24 hours the moment a visitor arrives on a link carrying a click ID — which is exactly what every paid click carries.
Download the playbook if you want to understand where attribution breaks, what data should be captured, what sales and marketing need to agree on, and how to build a stronger lead source process in a privacy-limited, AI-influenced marketing environment. No email required.
If you would rather not do it yourself
The checklist at the end of the playbook is the short version of the audit I run for clients. If you work through it and find more than three links broken, that is usually the point at which an outside pass is faster than an internal project.