Guide
Multi-touch attribution for service firms
Connect referrals, campaigns, and sales conversations to deals, then check what the report can support.
What does multi-touch attribution measure?
Multi-touch attribution divides credit for a conversion or deal across recorded interactions. The result depends on the model, the reporting window, and the data available. Keep those settings visible when using the report to plan channel spending.
Keep reported sources and observed activity separate
Ask how the client heard about you, then store that answer alongside the activity you can verify. A buyer may name a referral while the website visit carries a campaign tag. Keep both records.
Client-reported source
Record the answer, the date you asked, and the referring person or event when known. Keep unknown as a valid value.
Observed interactions
Keep dates, campaign tags, and relevant sales activity. Define which interactions qualify for the report so repeated visits or emails do not change the rules by accident.
Contact and deal links
Check that activity belongs to the right people and deal. HubSpot documents these associations as a requirement for including activities in deal and revenue attribution.
Choose a model your tool supports
Start with the question you want the report to answer. Write down the model and conversion event before comparing channels.
First or last interaction
These rules assign all credit to one recorded interaction. Define where the journey starts and ends, such as first contact through deal close.
Linear allocation
This rule divides credit equally across eligible interactions. Use the example below to check the arithmetic and confirm what your tool counts as an interaction.
Platform differences
Google Analytics currently offers data-driven, paid and organic last-click, and Google paid channels last-click attribution. It removed first-click, linear, time-decay, and position-based models in November 2023. Check the options in the tool you use.
Check that allocated revenue adds up
This fictional example uses one deal and three eligible interactions under a linear rule. It illustrates credit allocation only. Attributed credit alone does not establish how much extra revenue a channel caused.
Example: one closed deal worth 12,000 Recorded referral: 4,000 credit Campaign response: 4,000 credit Sales meeting: 4,000 credit Total allocated: 12,000 Model: linear Window: from recorded referral to deal close
Avoid counting the deal several times
If you show the full deal value next to every touched channel, label that view as influenced revenue. Do not add those channel totals together as company revenue.
Show missing data
Report how many deals have no source or linked activity. Keep those deals visible when comparing the report with total closed revenue.
Make comparisons you can explain
Review the underlying deals before changing spending. For a small set of wins, one large contract can change the channel ranking.
Use a consistent group of deals
State the close-date range, currency, and value definition. Separate new clients from renewals or expansion when they follow different sales processes.
Show the settings and limits
Include the model, lookback window, deal count, and missing-source count. Check whether the provider samples interactions or limits the history available on your plan.
Test the report during a CRM trial
Create sample contacts, a deal, a referral, and dated activities. Verify associations, the selected model, and the total. Repeat with a missing source and a second contact on the same deal.
Frequently asked questions
Sources
- Google Analytics: attribution models and current options
- HubSpot: attribution models, interactions, and associations
Provider documentation checked on 17 September 2026. Features and report options depend on the tool and plan.