There is a pot of money in the technology channel that should be one of the best growth engines in the industry. Market development funds — MDF — is the budget vendors hand to their partners to go and create demand. The logic is sound. Vendors can't reach every buyer themselves, so they fund the people who can: the distributors and resellers who sit close to the end customer. Done well, MDF is how you light up the long tail — the thousands of smaller partners and smaller deals that no vendor field team will ever touch directly.

Done badly, it's a bonfire. And right now, it's mostly a bonfire.

I've watched this from both sides. I run a channel software business now, but I spent years inside the channel itself — including a stretch at OpenLogic, an IBM reseller. I know exactly what an MDF-funded campaign produces, because I had to work the output. And I'll say the thing nobody in a QBR will say out loud: I would rather have had one genuinely good lead a year than the skip-load of rubbish our campaigns generated. One real buyer, properly understood, beats a thousand rows in a spreadsheet. It isn't close.

The lead form is where it all dies

Here's how the money actually gets spent. A vendor releases MDF. A partner builds a campaign — an email, an ad, a landing page. The landing page has a form. The form asks for a name and an email. And that form is where the whole thing dies.

A landing-page form converts at one, maybe two percent. That number is treated as a law of nature in this industry, when it should be treated as a scandal. Ninety-eight percent of the people the campaign paid to attract arrive, look, and leave — and we learn nothing about any of them. We don't know who they were, what they wanted, or why they didn't fill the form in. The campaign spend bought their attention for a few seconds and then threw it in the bin.

And the two percent who do fill it in? We barely know more about them. A name and an email tells you nothing about intent. It can't, because it never asked. The form is a turnstile, not a conversation. It cannot tell the difference between a CIO with budget and a renewal next quarter, and a student doing a project who'll never buy anything from anyone. They look identical on the form. They land in the same list. They get the same follow-up.

Qualification is the bit nobody can afford

So the list goes to a BDR to qualify. And this is where the economics fall apart completely.

Human qualification is expensive. A BDR working a list has to call every name, get past the fact that most numbers are wrong and most people don't pick up, and have the same opening conversation over and over to work out — usually in the first thirty seconds — that the lead is worthless. Most of them are worthless. So the BDR is being paid to discover, one painful call at a time, what the form should have told us and didn't: that the student and the CIO are not the same person.

It is too expensive to qualify a bad list properly, so nobody does. They skim it. They call the ones with recognisable company names, give up on the rest, and the genuinely interested buyer who happened to look like everyone else on the spreadsheet gets one rushed voicemail and is never called again. The one real lead in the batch dies next to the nine hundred fake ones, because we had no way of telling them apart and no budget to find out.

That is the core failure. MDF doesn't fail because the money is wrong or the partners are lazy. It fails because the mechanism we use to capture demand — the form — destroys the one thing that matters, which is knowing how interested someone actually is.

All those wasted conversations

Now think about what we threw away to get there.

Every one of those ninety-eight percent who left had something going on in their head. Some were idly curious. Some were a year out. But some — a precious few — were ready. They'd arrived because the campaign hit a real need, they wanted to understand what they were buying, when, what it would replace, and what the objection was they couldn't get past. They were ready to have the conversation that leads to a deal.

And we offered them a form.

They had questions, and there was nobody and nothing to answer them. So they left, and we recorded nothing, and the MDF that paid to bring them to the page bought us precisely zero understanding of the most valuable person who visited it. The conversation that would have been an asset — the actual substance of buying intent — was never even allowed to start. We don't just fail to capture those conversations. We prevent them.

AI changes the economics of the first conversation

Here's what's changed, and why this is now fixable rather than just annoying.

The reason we used a form was never that forms are good. It's that the alternative — a human ready to talk to every single visitor the moment they land, at any hour, in any volume — was impossible to staff. You cannot put a BDR behind every landing page waiting for someone to want a chat. So we put a form there instead and accepted the carnage.

That constraint is gone. An AI assistant can hold the first conversation with everyone who arrives — properly, not a scripted chatbot dead-end, but a real exchange that asks what they're looking at, what they're trying to solve, where they are in the process. It does the thing the form never could: it finds out how interested the person actually is. The CIO with a Q3 renewal has a very different conversation from the student, and the AI knows the difference by the third exchange — because it asked, the way a good BDR would, except it can do it ten thousand times at once and at no marginal cost.

This is the unlock. AI doesn't just replace the form with something nicer. It changes the economics of qualification. The expensive bit — working out who's real — moves to the cheapest possible place: the very first interaction, automated, with everyone, before a single BDR minute is spent. The human effort then goes only where it's warranted: onto the handful of genuinely interested buyers, who arrive at the salesperson already understood. Briefed, not cold.

And the ninety-eight percent who'd have bounced off a form? Now some of them talk, even briefly. And even a short conversation that ends in "not yet" is data we never had before — a record of what people came looking for and why they didn't proceed. That's campaign intelligence the form could never produce.

One conversation isn't enough — the system has to learn

But filtering the first conversation, on its own, isn't the whole answer. A lead isn't a single moment. It's a journey — first chat, follow-up, a call, a second call, a proposal. The interest reveals itself over time, and it moves. The student stays a student. The real buyer gets warmer, or goes cold, or surfaces a new objection at the third touch that wasn't there at the first.

So the thing that captures the first conversation has to be part of a system that watches the whole progression — that carries what was learned on the landing page into the salesperson's call, and what was learned on the call back into how the next campaign is built. A collective system, analysing conversations as the lead moves through its life, getting sharper each time. Every conversation should make the next one better: the chat this morning informs the call this afternoon, which informs the campaign next month.

Without that, you've just built a better turnstile. With it, you've built something that compounds — where the partner network's thousandth conversation is qualified better than its first, because the system has seen the pattern before. That is what turns MDF from a cost into an engine.

Make the money prove itself

There's a commercial sting in the tail, and it's the reason this matters now rather than someday.

MDF budgets are under pressure precisely because nobody can prove they work. When a vendor reviews the spend, an MDF-funded campaign that produced a spreadsheet of unqualified names has no defence — there's no line from the money to any pipeline, because the form severed it at the first step. So the budget gets cut, or pushed toward whatever's easiest to measure, and the long tail goes dark again.

Capture the conversations instead of binning them, qualify them with AI at the door, carry them through a system that learns, and attribute the outcome back to the campaign that started it — and suddenly the spend defends itself. Not with a list, but with a chain of real conversations and the deals they became.

MDF could be the thing that finally fires the long tail. The money is there. The partners are there. The buyers are there — they're literally landing on the page. We've just been greeting the best of them with a form and a turnstile, and wondering why the channel underperforms.

I'd still rather have one great lead than a thousand bad ones. The difference now is that we can finally tell which is which — and stop throwing the great one away.