Revenue attribution is the practice of assigning credit for closed deals to the marketing and sales touchpoints that contributed to them. It sounds methodological and dry — and the models themselves are relatively simple. The complexity is in the implications: which channels get investment, which activities get credit, and how marketing justifies its budget to the board. Choosing the wrong attribution model doesn't just miscount history — it systematically misdirects future investment.
Why attribution model choice determines channel investment
Consider a B2B deal that closes after 14 months. The buyer's journey included: a LinkedIn Sponsored Content ad impression (Month 1), an organic blog visit (Month 2), a webinar registration via email (Month 5), three SDR outreach emails (Months 7-8), a demo request from organic search (Month 10), and a closing call (Month 14). Under different attribution models, the credit allocation looks like this:
The investment implications are stark. Under last-touch, all credit goes to Google Search — and your annual planning will direct budget toward search and away from LinkedIn, content, and webinars. Under W-shaped, LinkedIn and webinars receive credit, and budget allocation reflects the full buyer journey. The model you choose determines the channels that get funded.
The six attribution models compared
How it works: 100% of credit to the first marketing touchpoint. Pro: Values awareness and demand generation investment — the channels that bring buyers into the funnel. Con: Ignores everything that happened between first touch and close. A channel that creates early awareness but never converts would look excellent under this model. Best for: Understanding which channels initiate the buying journey — used as a supplementary metric alongside other models, not as the primary attribution model.
How it works: 100% of credit to the final touchpoint before conversion. Pro: Simple to implement in any CRM or MAP without additional infrastructure. Con: Systematically under-credits every awareness and consideration touchpoint, over-credits the conversion trigger. Under last-touch attribution, branded search and retargeting appear to drive all revenue — because buyers who have been nurtured for 12 months use search to navigate back to a vendor they already know. Best for: Nothing. Avoid as a primary model. It is the most common and the most misleading.
How it works: Equal credit distributed across all touchpoints in the journey. Pro: Values the full journey. No touchpoint is invisible. Con: Treats a casual blog view and a 45-minute demo equally. Does not reflect the differential impact of touchpoints on buying decisions. Under linear attribution, channels with high contact frequency (email nurture) appear to drive disproportionate pipeline — not because they are the most influential, but because they have the most touchpoints. Best for: Establishing a baseline understanding of which channels have presence across the buyer journey.
How it works: 30% to first touch, 30% to lead creation (MQL), 30% to opportunity creation, 10% distributed across remaining touches. Pro: Values both demand generation (first touch) and conversion (opportunity creation). Reflects the two most strategically important moments in the buying journey. Con: The percentage splits are arbitrary — they are not validated against your specific data. A company with a 6-month sales cycle may need different weights than one with an 18-month cycle. Best for: Mid-market B2B programmes as the primary attribution model. More accurate than first/last touch; more practical than data-driven.
How it works: Machine learning assigns credit based on the statistical contribution of each touchpoint to actual conversion outcomes, derived from historical deal data. Pro: Eliminates arbitrary weight assumptions. Credit allocation reflects actual buying behaviour in your market. Con: Requires 1,000+ closed deals to produce statistically reliable models. Only available in Google Analytics 4 (for web conversions) and enterprise attribution platforms (Bizible, LeanData, Rockerbox). Requires significant MAP and CRM integration. Best for: Enterprise B2B programmes with high deal volumes and dedicated RevOps resource.
How it works: Combines data-driven or W-shaped multi-touch attribution with self-reported attribution ("How did you first hear about us?") to capture dark funnel touchpoints — LinkedIn posts seen but not clicked, word of mouth, podcast mentions, CTV impressions — that never appear in CRM tracking. Pro: Captures the full buyer journey including channels with no trackable click. Con: Requires consistent self-reported attribution data collection at every conversion point, and discipline in data analysis. Self-reported data has recall bias. Best for: Any B2B programme that invests in brand awareness, content, or social channels — which is to say, most programmes above £5M ARR.
Building your attribution infrastructure
Every paid media link, every email campaign link, every social media link should carry UTM parameters (source, medium, campaign, content, term). Without consistent UTMs, touchpoint data is incomplete and attribution models are unreliable. Build a UTM naming convention document and enforce it across every team member who creates campaign links.
Most CRM and MAP platforms capture the most recent source, not the original source. Implement a first-touch attribution cookie that captures the original traffic source on a buyer's first visit and passes it to the CRM when they convert — even if their 20th visit (the conversion visit) came from direct or branded search. In HubSpot, this is available natively. In Salesforce, it requires custom field configuration or a third-party tool.
"How did you first hear about us?" on every demo request form, discovery call script, and customer onboarding survey. Analyse the results quarterly alongside your multi-touch attribution data. The discrepancies between what self-reported attribution shows and what CRM tracking shows will reveal the dark funnel channels that are influencing your buyers invisibly.
Before committing to a primary attribution model, run first touch, last touch, and W-shaped in parallel for one quarter. Compare the channel rankings under each model. Where the models agree, you have high confidence. Where they disagree significantly, investigate — the disagreement reveals which channels are producing touchpoints at different stages of the journey, and which model best reflects your actual buying behaviour.
The dark funnel: what attribution misses
- LinkedIn organic content: A buyer who reads your CEO's LinkedIn post, saves it, thinks of you when a buying trigger occurs, and searches your brand name — appears as branded search in your attribution model. LinkedIn organic gets zero credit. But it was the trigger. Self-reported attribution is the only way to capture this.
- Word of mouth and referrals: The highest-converting B2B acquisition channel in almost every sector. Entirely invisible in digital attribution. Track referral source at point of first contact, and include it in your attribution analysis as a separate category.
- Event and conference conversations: A conversation at an industry event that leads to a demo request 3 months later will be attributed to whatever digital channel the buyer used to find your demo booking page. The event is the real first touch. Track this in your CRM using an "event" lead source field populated by your SDR or AE.
- Podcast and CTV impressions: Brand awareness from audio and video advertising has no trackable click. It shows up in your attribution data as direct traffic or branded search. The only way to measure it is through brand lift studies, geographic cohort analysis, or self-reported attribution.
Our RevOps and analytics programmes build the attribution infrastructure, pipeline reporting, and sales-marketing alignment frameworks that let you prove marketing ROI.