
Which campaign deserves more budget when paid search, email and the CRM all claim the same enquiry? Before choosing a winner, you need to understand what each report is crediting.
Marketing attribution assigns credit for an outcome to the recorded interactions that preceded it. The attribution model decides how that credit is shared. A touchpoint is one eligible interaction, such as an ad click or a visit from an email.
For B2B teams, that outcome might be a qualified enquiry, an opportunity or a sale. The useful question is which campaigns introduced the prospect, supported their evaluation or brought them back to enquire.
This guide explains the main models, follows one enquiry through three credit allocations and shows how to use the findings in a budget discussion.
Start with the decision in front of you. If you need more sales conversations, report qualified enquiries. If you are reviewing pipeline, use a defined opportunity stage. A content download can be useful engagement without being either of those outcomes.
Keep those stages separate. Ten downloads, three enquiries and one opportunity could describe progression within the same group of people. Adding them together would obscure what happened.
Decide whether the report counts people, accounts or opportunities. Two contacts involved in one purchase should not automatically become two deals. Keep sourced outcomes, influenced pipeline and allocated credit clearly labelled.
Single-touch models assign all credit to one eligible interaction. Multi-touch models distribute credit across several. Use the comparison below to choose a useful reporting view, then check whether your platform supports it.
| Model | When the view is useful | Where it can mislead |
|---|---|---|
| First touch | Reviewing where recorded prospect journeys begin. | Later interactions receive no credit. Missing earlier visits can change the apparent starting point. |
| Last touch | Reviewing the eligible interaction immediately before an enquiry. | A campaign that closes the recorded journey can appear responsible for all the demand. |
| Linear | Reviewing participation across a recorded journey with equal credit per touch. |
U-shaped and W-shaped approaches are position-based variants. Check which milestones they weight and how your implementation defines them. Adobe’s model definitions explain several common credit rules.
These are general model families. GA4 attribution documentation currently lists data-driven attribution and two last-click variants. First-click, linear, time-decay and position-based models left GA4 attribution reporting in November 2023.
Suppose a prospect discovers your business through a paid social guide. They return through organic search, read a case study from an email and later submit a qualified enquiry after a paid search click.
Assume all four interactions are linked to this prospect and fall inside the reporting window. The graphic shows how three rules divide credit for that same enquiry. The figures are hypothetical.

The total stays at one enquiry. What changes is the campaign that appears strongest.
Would you cut paid social because the last-touch report gives it zero? First check whether it introduces suitable prospects who later progress. Would you increase paid search because it receives all the credit? Check whether those enquiries become opportunities and whether acquisition cost remains acceptable.
Our B2B PPC guide covers that channel-level quality and cost review. Attribution provides a view of the journey, while the budget decision also needs evidence about the outcomes you are buying.
Begin with the question you need answered. A report for discovering new audiences has a different purpose from a report for reviewing enquiries ready for sales.
Choose a main reporting view and record the outcome definition, window and data sources beside it. Use another model to investigate differences, rather than selecting whichever version makes each campaign look best.
If a channel gains credit under another model, inspect a sample of the underlying journeys. Establish whether the difference comes from its role in the journey, a changed rule or missing records before reallocating spend.
A B2B purchase can involve several people. Account mapping helps identify those buying roles. Attribution also needs a reliable connection between their campaign interactions and the opportunity being assessed.
Record campaign identifiers consistently, check that enquiry records reach the CRM and verify how contacts connect to accounts and opportunities. Keep unmatched records visible. Sharing a company domain does not prove that two people are evaluating the same purchase.
The lookback window sets how far before an outcome an interaction remains eligible. In the graphic, the first visit is 37 days before the enquiry. A 30-day window would exclude it, potentially moving first-touch credit to organic search.
That change describes a reporting boundary. It does not mean paid social stopped contributing. Record window changes so the team can interpret comparisons fairly.
Before deciding that a campaign report is wrong, trace a few enquiries through Google Ads, GA4 and the CRM. Ask whether the reports use:
A buyer’s answer to “How did you hear about us?” can add context about recommendations and conversations outside tracked activity. Keep it labelled as self-reported information rather than treating it as a complete journey.
Also separate attribution from incrementality. Attribution allocates credit. Incrementality asks what additional result occurred because of an activity, requiring an appropriate causal evaluation. A higher credit share alone cannot answer that question.
ROI requires a separate cost and financial-return calculation. Our webinar ROI guide explains that distinction for a specific campaign type.
Use attribution to choose what to investigate next. A closing campaign gaining credit may deserve a lead-quality review. An introductory campaign appearing in journeys that reach opportunities may deserve closer examination before you cut it.
For each proposed budget change, record the outcome you expect to improve, the evidence behind the decision and when you will review it. Keep the reporting rules stable during that comparison and allow time for the chosen sales outcome to develop.
For your next campaign with DemandTech’s marketing team, include the outcome and reporting requirements in the brief. You can discuss that brief with the campaign goal and current measurement gaps already defined.
Quick answers to common questions about this topic and implementation strategy.
| Equal shares are a reporting rule, not evidence of equal influence. |
| Time decay | Giving more weight to interactions closer to the outcome. | Older introductory activity receives less credit because of the weighting rule. |
| Position based | Giving specified journey milestones more weight. | The chosen milestones and percentages shape the result. They require explanation. |
| Data driven | Using platform modelling to allocate credit from observed journey data. | The result depends on the platform’s coverage and method. Unmatched activity still needs attention. |