Attribution modelling
Attribution modelling your team can read and your CFO will accept.
Most companies we meet already have attribution reports. What they do not have is a number anyone will defend in a budget meeting, because nobody can say what it counts. We build the model into your CRM, write the definitions down so marketing and sales are reading the same thing, and hand it over to your team to run.
Built in HubSpot or Salesforce, documented, and handed over. Written-first, no call required.
The work
Built by someone who opens the CRM
Attribution goes wrong in the data long before it goes wrong in the model. We start in the records, fix what is breaking the join between a contact and a closed deal, and only then argue about weighting.
Ask a senior operator ↗
How it works
What an attribution build looks like
A fixed-scope engagement, usually six to eight weeks depending on how much repair the data needs. Tap a step to see what happens in it.
The build
What an attribution model is made of
Six pieces, and skipping any one of them is why the last attempt did not stick. Tap a node or a chip.
The definitions are the piece people skip, and they are the reason the model holds or falls over. We walk through them in closed deal attribution.
Why us
Attribution that survives the first argument
Most attribution work is delivered as a dashboard. The dashboard is the easy part. What decides whether it lasts is whether sales agreed to the definitions before it was built.
We fix the data before the model
A weighting applied to records that will not join is a decorated guess. We repair what is breaking the link between contacts and closed deals first, which is unglamorous and is also the reason the number ends up believable.
Definitions signed off by both teams
Sales approves what counts as a touch before we build anything. It is a short document and it prevents the quarterly argument that kills most attribution projects.
Built in your CRM, not beside it
HubSpot or Salesforce native, reading live records. No separate tool to license, and nothing that stops working when someone forgets to run the export.
Handed over, not held hostage
You get the documentation and a walkthrough. If you never call us again the model still runs, which is the outcome we are aiming for.
What changes
From three reports that disagree to one number people use
The aim is a figure the marketing lead, the sales director and the CFO can all look at without starting a negotiation about what it means.
The argument this settles is covered in full in closed deal attribution.
Who it is for
And who it is not
Marketing spend big enough to argue about
Once the budget is material, someone will ask what it bought. If the honest answer today is a shrug and a channel report, this is the work that fixes it.
Sales and marketing reporting different numbers
Both pulling from the same CRM and reaching different conclusions. That is a definitions problem, and it does not resolve on its own.
Not a fix for an empty CRM
If the records are thin because nothing is being logged, attribution is the wrong project. Start with the cleanup, then come back to this.
Not a dashboard purchase
If what you want is a tool installed and someone else to decide what it counts, an agency retainer will be cheaper. The value here is the decisions, and those need your people in the room.
Questions
Attribution modelling, answered.
What is attribution modelling?
Attribution modelling is the set of rules that decides which marketing activity gets credit for revenue. It covers the weighting between touches, the window in which a touch still counts, and the definitions of marketing-sourced and marketing-influenced. The model matters less than whether everyone reading the report agrees on what it counts.
How long does it take to implement attribution in HubSpot or Salesforce?
Six to eight weeks for most B2B companies, and the variable is data repair rather than build time. If contacts and closed deals already join cleanly it is faster. If deals were created without contacts, or reps have been typing people in by hand, that gets fixed first because a model built on records that will not join produces a confident wrong answer.
Do we need a multi-touch model?
Often not. Multi-touch is worth the complexity when sales cycles are long, spend is spread across several channels, and the touch data is captured reliably. When cycles are short or the tracking has gaps, a simpler model tells you nearly the same thing and is far easier to defend. We recommend the model your data can honestly support, which is sometimes less than what was hoped for.
Why do our sales and marketing numbers disagree?
Almost always because the two teams are using different definitions rather than different data. Marketing counts influenced revenue, sales counts sourced, and neither figure is labelled. Writing the definitions down and having both teams sign off is the single change that resolves it, and it costs nothing but a meeting.
Will our team be able to run it after you leave?
That is the point of the handover. You get the documented definitions, a walkthrough with whoever will maintain the model, and the list of things that quietly break attribution - a renamed property, a rep bypassing a form, a campaign created outside the structure. A model only the consultant understands is one we consider unfinished.
Ask us directly
Do your numbers disagree?
Tell us what your reports currently say and where they contradict each other. We will read it and tell you straight whether this is an attribution problem or a data problem underneath it, because the two need different work and only one of them is this.
You will not get a pitch deck or a junior, and there is no obligation. Just a real read, in writing.
See our profile on Clutch ↗Ask about attribution modelling
Write us a few lines about where marketing is now. You’ll get back a senior read on whether fractional fits - and what the first ninety days would focus on.