Request a callB2B attribution challenges: why paid media looks worse than it is
B2B attribution usually breaks in three places: the campaign source stops at the lead and never reaches the opportunity, ad platforms optimize to form fills instead of revenue, and the attribution model is picked by default rather than agreed. The fix is a closed loop: capture UTMs and click IDs on every form, stamp the source onto the opportunity in the CRM, distribute revenue with multi-touch Campaign Influence, send outcomes back to Google and LinkedIn, and report pipeline-to-spend, CAC and ROAS in one view.
Key takeaways
- Most paid programs stop measuring at the lead, which is exactly where budget decisions need data.
- If the source is not stamped on the opportunity, revenue can never be attributed to a campaign.
- The same deal can give paid media anywhere from 0% to 100% of the credit depending on the model, so agree the model first.
- Sending CRM outcomes back to the ad platforms is what turns measurement into better spend.
Where B2B attribution breaks
Enterprise deals take months, involve six to ten stakeholders and touch many channels before anyone signs. Ad platforms report clicks and form fills; the CRM holds opportunities and revenue. Attribution is the bridge between the two, and in most B2B stacks the bridge has gaps.
In our paid-to-revenue working sessions with enterprise teams, the same breakpoints come up again and again.
- Attribution stops at the lead: reporting ends at the MQL, so nobody can trace a campaign to a closed deal.
- Platforms optimize to form fills: Google and LinkedIn bid toward conversions that may never become pipeline.
- A disconnected stack: ad platforms, GA4, the marketing automation platform and the CRM do not share one source of truth.
- No pipeline-to-spend view: leadership cannot see return, so budget decisions run on instinct instead of evidence.
What each gap costs
| Failure mode | What happens | What it costs |
|---|---|---|
| Source stops at the lead | UTMs are captured on the prospect but never stamped on the opportunity created from it | Pipeline from paid cannot be attributed at all |
| Platforms bid to form fills | Every demo form counts as a conversion whether or not it becomes a deal | Algorithms chase volume; CAC can be overstated by 30% to 50% |
| Click IDs not stored | gclid and li_fat_id are lost after the visit | Revenue can never be sent back to Google or LinkedIn |
| Model chosen by default | Last touch credits the sales meeting | Paid media looks like it produced nothing and gets cut |
The same deal, five different stories
Take one closed-won deal worth $120,000 with six touchpoints over 61 days: a branded Google search, a LinkedIn whitepaper ad, display retargeting, a LinkedIn demo request, a competitor keyword search and finally a sales meeting.
Under last touch, the sales meeting gets all $120,000 and paid media gets nothing. Under first touch, the brand keyword gets everything. Linear spreads credit evenly and ignores which touches mattered. Position-based (40/20/40) gives 40% to the first touch, 40% to the last and shares 20% across the middle, which credits paid media with 60% of this deal. Time decay favors the touches nearest the close.
None of these is the truth; each is a lens. What matters is that the model is chosen on purpose, signed off by leadership before the reports are built, and shown side by side so nobody mistakes one lens for reality. For long, multi-stakeholder enterprise deals we recommend position-based as the default view.
Try it yourself with our free attribution model simulator, which shows the credit each channel gets under all five models.
The fix: a closed loop from click to closed-won
- Capture: a governed UTM taxonomy on every paid URL, plus click IDs (gclid for Google, li_fat_id for LinkedIn), written to first-party cookies by Google Tag Manager.
- Carry: hidden form fields write source and click IDs onto the lead or prospect. First touch is stamped once and never overwritten.
- Stamp: a set-if-empty flow copies the campaign source from the contact onto the opportunity when it is created. This is the step that fails silently in most setups.
- Attribute: Campaign Influence distributes opportunity amount across every campaign that touched the deal, using the model you agreed.
- Feed back and report: offline conversions send opportunity created and closed-won values back to Google and LinkedIn, and a single dashboard joins ad spend to CRM revenue.
The metrics that matter
| Metric | Formula | Benchmark we use |
|---|---|---|
| Pipeline-to-spend | Influenced pipeline / paid spend | 6x or above |
| Cost per SQL | Paid spend / sales-qualified leads | $800 to $2,500 in enterprise B2B |
| MQL to SQL rate | SQLs / MQLs | 20% to 30% for an enterprise ICP |
| ROAS (closed-won) | Closed-won revenue / paid spend | 1.5x minimum, 3x strong |
| Paid CAC | Paid spend on closed-won / new customers | Should shrink quarter over quarter |
In an anonymized worked campaign we present to enterprise buyers, this model turned $600K of media into $7.2M of attributed pipeline (12:1) and $1.9M of closed-won revenue, with CAC down 34% and cost per SQL down 41% against the prior baseline.
How long it takes
If the paid source already reaches the opportunity and Campaign Influence is on, the reporting layer can be built in two to three weeks. If not, a short foundation sprint comes first: enable connected campaigns and influence, build the source-stamping flows and add hidden fields and click-ID capture. Tracking is forward-looking, so the sooner capture is live, the sooner the data exists.
Frequently asked questions
Why is B2B attribution so hard?
Long sales cycles, many stakeholders and a gap between ad platforms (clicks and form fills) and the CRM (opportunities and revenue). Unless the campaign source is carried onto the opportunity, revenue cannot be traced back to spend.
Which attribution model should B2B teams use?
For long enterprise deals, position-based (40/20/40) as the default, with first touch and last touch available side by side so leadership sees the range.
What is offline conversion import?
Sending CRM outcomes such as opportunity created and closed-won, with value, back to Google Ads and LinkedIn using the stored click ID, so the platforms bid toward pipeline instead of form fills.
How quickly can attribution be fixed?
Attribution can be live in about 14 days when the CRM foundations exist; otherwise a foundation sprint comes first. Results are forward-looking from the day capture goes live.
Related service: RevOps and marketing operations. See also pricing and client results.