We recently surveyed 685 senior partner and ecosystem leaders about how AI is landing across partner programs. One pair of numbers has stayed with me.
Sixty-three percent are using AI or automation for ecosystem analytics and reporting. For MDF and program management, it’s 34%.
That gap is not about difficulty. Reporting is not an easier problem than MDF. The difference is whose data each one runs on.
The work that stops at your own front door
Reporting, partner communications, enablement and training all sit inside your organization. You own the data, you set the rules, and you can automate as far as your appetite takes you. Unsurprisingly, that’s where partner teams have gone first, and they’ve gotten real value out of it.
MDF is a different animal. Approving a claim means knowing what the partner actually ran, what it produced, and whether the spend did what it was supposed to do. That evidence lives in their systems, in their format, on their timeline. You are not automating a process so much as negotiating one.
Due to the nature of how different partners work, ecosystems aren’t integrated into important workflows like revenue tracking or attribution reporting. Every partner brings its own rules on tools, data sharing, and privacy. When you’re operating one-to-many, that becomes a real constraint on what you can automate, no matter what technology you’ve bought.
Which is why MDF accountability has been a manual, relationship-dependent, quarterly-argument kind of process for as long as most of us have been doing this.
The teams that got past it
The survey did surface a group pulling ahead. Organizations that have integrated AI broadly across sales, marketing, and other functions are running AI on MDF and program management at 44%, against 29% for everyone else.
What separates them is worth noting, because it isn’t a tool. That group has more governance in place, not less. Formal policies, data access controls, partner usage guidelines, defined human oversight.
That sounds backwards until you accept what the real constraint is. You cannot automate an approval you haven’t defined, and you cannot define it alone. Governance is how two companies agree on what counts as proof before anyone tries to put software on top of it.
Accountability is an agreement problem first
Most MDF conversations I’m in start with process. Faster approvals, cleaner submissions, better templates, a portal somebody hates.
Those things matter, and they’re downstream of a harder question. Do you and your partners agree on what a funded program is supposed to produce, and what evidence settles it? If the answer is no, better process just moves the disagreement to a new place in the calendar.
Get the agreement right and the rest becomes tractable. Automation has something to attach to. Measurement stops being a reconstruction exercise at quarter end. And the accountability conversation changes from chasing receipts to comparing results.
Join us on October 7
I’m hosting a Voice of Partner Marketing session on exactly this, with Angharad West from Kyndryl, Paige Johnson from Pega, and Jordan Tarantino from Samsara. Three people running MDF at real scale, talking about what they’ve changed about adoption, accountability, and impact.
Register here: https://www.pipeline360.com/webinar/unlocking-mdf-driving-adoption-accountability-and-impact/
We’ll publish the full research later this fall. In the meantime, a question worth sitting with: if you had to prove the return on last quarter’s MDF spend this week, how much of that evidence could you get without asking a partner for it?