The uncomfortable fact about most MDF programs is not that the money fails to return. It is that nobody can say either way.
The proposal lives in an email thread, the approval in a spreadsheet, the claim in a finance folder and the resulting pipeline in the CRM. Four artefacts, no thread between them. At year end the program reports how much was spent and claimed, which is activity, not outcome.
The Chain That Has to Hold
Attribution needs an unbroken chain from intent to outcome. Each link is simple; the chain almost never survives intact.
- A proposal that names a measurable expected outcome before any money is committed.
- An approval that commits funds against a budget and records what was promised.
- A campaign object that the activity belongs to, rather than a loose claim.
- A claim that attaches to that campaign with its evidence.
- Registered deals or leads associated with the campaign, so the outcome is observable.
- A payout recorded against the same campaign, closing the loop.
Where It Breaks
Almost always at link three. Without a campaign object, a claim is a standalone reimbursement with no place to hang the pipeline that followed.
- Claims filed as individual reimbursements with no campaign to group them
- Proposals approved with aspirational targets that were never measurable
- Pipeline recorded in a CRM the MDF process cannot see
- Partner-sourced leads that arrive with no campaign reference at all
- Payouts reconciled in finance with no link back to the activity
Making the Chain Hold
None of this requires new process from partners. It requires the objects to exist and to reference each other.
- Require a numeric target on every proposal, and store it rather than discussing it
- Create a campaign at approval, not at claim time
- Let partners attach registered deals to the campaign that generated them
- Keep budgets, claims and payouts on the same object as the campaign
- Report pipeline per MDF unit spent, segmented by activity type
What the Numbers Then Tell You
Once the chain holds, the reporting stops being a defence of the budget and starts being a steering mechanism.
You can see which activity types return, which partners convert funded activity into pipeline, and which campaigns consumed budget and produced nothing. That last category is the point: the value is in stopping what does not work, not in proving that the programme overall was worthwhile.
- Pipeline per unit of MDF, by activity type
- Utilisation against allocation, by partner and tier
- Time from claim submission to payout, against your published SLA
- Share of partner-sourced pipeline that traces to a funded campaign
Frequently Asked Questions
Why can most programs not prove MDF ROI?
Because the claim and the pipeline never share an object. Without a campaign that both attach to, there is nothing to join on, and the reporting degrades into spend versus budget.
What should a partner have to supply for attribution to work?
A numeric target at proposal time and a campaign reference on resulting leads. Both are light asks. Demanding full attribution modelling from partners is where programs lose cooperation.
Is pipeline per MDF unit a fair measure?
It is the most useful one available, provided you segment by activity type. Brand activity and demand generation return on different timescales, and comparing them on one number punishes the slower one unfairly.
What if the data shows poor return?
That is the programme working. The purpose of attribution is to stop funding what does not produce, not to generate a number that justifies the budget.