Like the last runner in a relay taking the photo with the trophy, branded search tends to arrive at the finish line and act like it ran the whole race.
This is the last-click attribution problem, and B2B marketers need to spend some time thinking about it.
Last-click attribution gives all the credit to the final touchpoint or paid interaction before a conversion. In B2B, that often means branded search or retargeting ads get the win, even when another channel created the interest weeks or months earlier. The click is real, of course. It’s just that the full credit is going to one source in a more complex picture.
This article explores why the fix here is never going to be a flawless dashboard. Then we describe a data-driven attribution setup that connects paid activity to pipeline in a workable way.
What is PPC attribution, and what does last-click actually measure?
PPC attribution is the method used to assign conversion credit across the paid-media interactions that preceded an outcome. An attribution model may use a rule, a set of rules or a data-driven algorithm to decide how that credit is distributed.
- Single-touch models assign all the credit to one interaction
- Last-click rewards the last of the eligible touchpoints before conversion; first-click rewards the first.
- Multi-touch attribution models divide credit across several interactions using a predefined weighting. Time-decay attribution, for example, gives progressively more credit to touchpoints that occur closer to the conversion.
None is “objective”. Each is an attribution model that assigns a weight to the interactions the system managed to record. Change the model, and the apparent contribution of each channel changes.
PPC attribution models at a glance
| Model | How credit is split | What it is good for | Where it misleads B2B |
| Last-click | 100% to the final interaction | Simple reporting where journeys are genuinely short | Rewards demand capture and hides earlier demand creation |
| First-click | 100% to the first interaction | Seeing where measurable journeys begin | Ignores the nurture and later interactions that moved the deal forward |
| Linear | Split evenly across recorded touches | A simple first step beyond single-touch reporting | Treats every recorded interaction as equally important |
| Time-decay | More credit to touches nearer conversion | Examining late-stage activity in shorter buying cycles | Systematically reduces credit for early demand creation |
| Position-based | Most credit to the first and final touches | Comparing entry and conversion-stage interactions | The weighting is a reporting convention, not a finding |
| Data-driven | Modelled from observed conversion paths | Distributing credit using patterns in account data | Cannot value interactions the platform did not observe |
These models remain useful ways to think about attribution, but they are not all currently available as native choices in Google’s platforms. Google Ads no longer supports first-click, linear, time-decay or position-based attribution; current options centre on data-driven attribution and last-click. GA4 likewise compares data-driven and last-click reporting.
Data-driven attribution is not a complete view of the buyer journey either. It uses observed converting and non-converting paths to distribute credit, which makes it more nuanced than a fixed rule—but it still only works with interactions the platform can identify.
Why does last-click break down in B2B?
Last-click attribution breaks down when the buying journey is longer than the system’s view of it. Analytics platforms like GA4 and Google Ads only credit interactions that fall inside a configured lookback window or conversion window. When an early paid touch occurs outside that attribution window, it becomes ineligible for credit even if it helped start the buying process.
It also measures people more easily than buying committees. B2B decisions often involve several stakeholders with different concerns, researching through different devices, browsers and identities. Standard user-level attribution may record those journeys as separate people rather than one account.
It rewards the touch closest to conversion
Last-click also favours demand capture. In many accounts, branded search or retargeting ads sit near the end of the journey, after another channel has created awareness or interest. The final ad interaction receives the credit because it was closest to conversion.
Then there is the dark funnel: the parts of the buyer journey that influence a decision but leave no attributable click. A podcast mention, peer recommendation, private community conversation or organic LinkedIn impression may shape demand without appearing in conventional attribution. The system cannot credit an interaction it never recorded.
It gets worse, because last-click has an actual cost of its own
Last-click attribution often creates a budget spiral.
Picture this: a B2B SaaS company running LinkedIn ads
As a top-of-funnel channel, LinkedIn looks expensive because it rarely receives the final click. So the company cuts the budget. This means cost per lead improves, at least temporarily. The report looks cleaner, so this decision looks like it was the right one to make.
A few months later, branded search volume starts to soften. Retargeting audiences shrink. Pipeline becomes thinner. By then, the connection to the earlier budget cut is easy to miss.
They cut more than they bargained for
This is where channel comparisons become especially misleading. LinkedIn may have created awareness and shaped demand, while Google captured the eventual search. Last-click awards the result to the channel nearest the conversion and makes the earlier investment look inefficient.
The consequence is not just inaccurate reporting. It is a budget model that repeatedly rewards demand capture while starving the channels that help create future pipeline.
//INSERT IMAGE (see design brief on the second tab of this document)
So what can you do instead?
There is no perfect attribution model. But it is possible to create a B2B measurement setup that combines imperfect signals that fail in different directions.
- Use multiple attribution views: Compare first-touch with a model that gives weight to early and late interactions. The gap shows which channels introduce demand and which capture it.
- Model pipeline in the CRM: Measure paid media against qualified opportunities and closed-won revenue, not form fills. Keep marketing-sourced pipeline and marketing-influenced pipeline separate.
- Send downstream outcomes back to the platforms: Where supported, feed qualified-lead, opportunity and customer stages into ad accounts so reporting and bidding reflect pipeline quality rather than raw submissions.
- Add self-reported attribution: Use an open-text “How did you hear about us?” field to surface podcasts, communities, referrals and other influences that left no trackable click.
- Test incrementality where volume permits: Compare an exposed group with a control group to estimate whether the advertising caused additional conversions, not just appeared on the same path.
Finally, define a reporting window for each channel and publish it beside the result. For example, high-intent search may justify a shorter window than paid social or brand activity.
Immaculate dashboards are not possible. What is possible: s a measurement stack honest enough to support better budget decisions.
How to tell if your attribution is working
In simple terms, your attribution setup is working when it helps the business make those better budget decisions.
A practical check:
- The dashboard and the sales team describe the same pipeline. Reported channel performance should broadly match what sales is hearing from prospects and seeing in active opportunities.
- Every result has a stated lookback window. Anyone reading the report should know how far back the model searched for eligible interactions.
- Branded search is separated from demand creation. It may capture a large share of conversions, but it should not automatically receive credit for the conversion.
- Major budget decisions use more than platform conversions. Pipeline data, model comparisons, self-reported responses and controlled tests should all inform what happens next.
Remember: attribution helps decide where to place budget. It does not tell you what message or creative will persuade the audience once the ad reaches them.
How Inspired Marketing measures paid media
Nobody needs another PPC dashboard full of green arrows and no clear answer.
We start with the plumbing that sustains marketing measurement. CRM stages need to be clear, and lead definitions need to mean the same thing to marketing and sales. The ad platforms need to receive more than a stream of form fills.
Where the setup allows it, we feed qualified-lead, opportunity and closed-won stages back into the platforms. We add a simple “How did you hear about us?” field to high-intent forms. We compare more than one view of performance, separate sourced pipeline from influenced pipeline, and check the numbers against what sales is seeing in real conversations.
This is not glamorous work. But it is the work that makes the budget defensible.
Send us your current PPC report, and we’ll tell you what it may be hiding.
