B2B attribution vs marketing attribution
Marketing attribution scores the clicks, sessions, and campaign conversions that live in ads and analytics. B2B attribution has to keep going until the visit, the meeting, and the closed deal sit on the same path.
That difference is why the two reports disagree. Marketing attribution is built around a campaign conversion, often a form fill or a click. B2B attribution is built around a revenue event that usually happens weeks later, in a CRM, after a call that never fired a pixel.
If you only measure the first system, you will fund the channel that caught the last click and starve the work that created the meeting. We wrote about what B2B attribution has to connect. This piece is the comparison: where marketing attribution stops, and what B2B teams still need after that.
What does marketing attribution actually measure?
Marketing attribution answers a campaign question. Did this ad, this search term, or this email produce a click, a session, or a form fill?
In Google Ads, that looks like conversions tied to clicks and, sometimes, view-through. In GA4, that looks like sessions, events, and a model that assigns credit inside the property. Google Search Console tells you which queries got impressions and clicks, not which deals closed.
Those tools are good at what they were built for. They see the visit. They rarely see the meeting that started the opportunity, and they almost never see the closed-won amount. The conversion you optimize toward is the one the platform can count.
CRM attribution is a different slice. HubSpot can show which contact property or campaign a deal is associated with. That association is only as complete as the first form, the UTM that survived, or the field a rep typed in. It does not automatically include the pages read between the first visit and the Calendly booking.
So marketing attribution, in practice, is a score on campaign activity. It is useful. It is incomplete for a B2B sale.
Why do B2B cycles break that model?
A consumer checkout can finish in one session. A B2B deal usually takes more than one buyer, more than one visit, and a meeting that lives off the website.
The visit happens in the browser. The meeting lives in a calendar. The deal lives in the CRM. Spend lives in ad platforms. Those four records do not share an ID unless you stitch them.
Campaign conversion windows also expire before the deal closes. An ad platform will happily credit a demo request to last week's click. Finance will ask which channel produced the customer that closed this quarter. Those two answers are not the same event, so they will not match.
The buying committee makes it worse. One person reads a comparison page. Another books the call. A third signs. Marketing attribution often sees one of those people. B2B attribution has to keep the account's path together, or the report treats three humans as three unrelated sessions.
Time works against the campaign model too. A VP can see an ad in March, share a pricing page in Slack, and only book in May. The click report for March is closed. The pipeline report for May has no memory of the ad. That is a B2B cycle doing what B2B cycles do.
How does last-click compare to a full path?
Last-click is the default because it is easy. One conversion, one channel, one owner. The last ad or search result before the form fill gets the deal.
On a B2B path that looks like this: a VP sees an ad, reads a comparison post, comes back from search two weeks later, books a demo, and the deal closes the next month. Last-click hands the whole thing to search. The ad looks wasted. The post never appears in the CRM. Sales thinks they sourced it because they sent the calendar link.
A full path keeps every step. First visit, pages, meeting, closed deal. You can still run last-click on that path if you want a simple view. You can also run multi-touch on the same events. The model is a lens. The path is the data.
If the underlying events are not stitched, last-click and multi-touch both fail. They just fail with different math. Marketing attribution can score the last click it saw. B2B attribution needs the clicks it did not see in the ad platform, plus the offline step that actually started the opportunity.
View-through belongs on that full path too. Someone can see an ad, never click it, and still show up later from branded search or a colleague's link. If you only credit the click, the impression did not happen. B2B teams that run long cycles lose a lot of that activity.
Where do LLM referrals and CAC show up?
LLM attribution is the newest hole in marketing attribution. Someone asks ChatGPT, Claude, Perplexity, or Gemini for a vendor and arrives as direct traffic. There is no UTM. GA4 will not name the model. The ad platform never got a click ID.
If those visits sit in leftover direct, you will underfund the pages and answers that LLMs cite, and you will overfund the paid channel that harvested the branded search later. B2B attribution has to treat those referrals as their own source, the same way it treats a view-through or an offline conference meeting. We wrote a full piece on how to attribute ChatGPT and other LLM traffic.
CAC is where the comparison becomes a budget decision. Marketing attribution can tell you cost per click or cost per form fill. Real CAC needs closed customers and the spend that sat on their path. If the deal is missing, CAC is a lead cost wearing a customer label.
Channel quality follows the same rule. A channel that produces cheap form fills and no meetings is not cheap. A channel that produces fewer clicks and more closed deals is the one you keep. You only see that when spend, visits, meetings, and revenue share a journey.
Goals make the split obvious. A marketing goal of more demo requests can be hit by last-click paid search. A company goal of efficient closed revenue needs the rest of the path. If those two goals use different event graphs, the weekly meeting turns into a debate about whose dashboard is right.
A check you can run this week
Pick five closed-won deals. For each one, name the first visit, the meeting that started the opportunity, and the deal amount. Then ask whether last-click, the CRM source field, and the ad platform agree.
If they disagree, you are looking at marketing attribution next to a revenue event. The gap is B2B attribution. The events never met.
You do not need a warehouse to close that gap. One pixel on the site, plus the GTM tools you already run, is enough to stitch visits, page views, meetings, and closed deals. You can chat with the data, get an AI recommendation, and receive a daily report in Slack.
Source is built that way. Drop in one pixel and it connects across the tools you already use. Multi-touch, view-through, offline activity, and LLM attribution sit on the same path as a real CAC, a goal, and a read on channel quality.
Customer data is not used to train external models. Source is SOC 2 Type II and ISO 27001 certified.
If you want the visit, the meeting, and the closed deal in one place, start at Source. Attribution is broken. So we rebuilt it.