It comes up early with almost every new partner. How many vendors do you carry? Our answer, somewhere around a dozen, tends to earn a raised eyebrow, because the cloud marketplaces we get held up against are advertising two hundred and counting. On a side-by-side slide we lose that comparison every time. For a while that bothered us, and then it stopped, once we'd worked out what the slide was actually measuring.
If your problem is pure procurement, a big marketplace will serve you well, and there's no point pretending it won't: one sign-up and one bill instead of ten, with almost any product available to put in front of a client by end of day.
The easy question and the expensive one
But finding a backup vendor was never the hard part of anyone's week. There are dozens of competent ones and a client will happily sign for whichever logo you bring them. The questions that actually cost money surface much later. Can the appliance still cope once the dataset has quietly doubled, and if it can't, how long before anyone actually notices? Does the immutable copy that reads so reassuringly in the brochure survive the day an admin account gets phished? Somewhere in that mix there's usually a client running Proxmox next to a bit of legacy VMware, where the choice between agent-based and storage-level protection quietly decides how ugly their next recovery window looks. A grid of vendor names can't help you with a single one of those.
What a vendor's support desk can and can't do
The fair objection here is that a marketplace doesn't abandon you when things get hard. It escalates the ugly tickets to the vendor, whose own engineers know the product far better than any reseller ever will. That's completely true, and when a problem really does live inside one product, those engineers are exactly who you want on the phone.
It's just that the failures which genuinely hurt a client mostly don't live inside one product. They turn up in the joins. A backup agent behaving oddly against a particular hypervisor while it tries to land an immutable copy offsite, with a chunk of the client's data sitting over in Microsoft 365 where that vendor has no visibility at all. Three different support desks will each confirm their own piece is healthy, and every one of them will be telling the truth. The gap between the pieces is real work, and it belongs to nobody on that list. Somebody still has to stand in that gap, and for a lot of our partners that somebody ends up being us.
Why we turn vendors away
Our rule for taking a product on is not a glamorous one. Our engineers need to be able to design with it, price it without guessing and keep it alive when it starts playing up, and in practice that means we've usually had the thing in the lab long enough to break it ourselves at least once. Anything that can't clear that bar, we let go, even when there's obvious money in it, because carrying a product we can't properly support would leave us taking a clip of the invoice and adding nothing a client couldn't get elsewhere. That same rule is why our comparison write-ups are willing to come down on one side, and why a partner who calls up wanting to thrash out StoneFly versus QSAN for an awkward workload gets an actual argument back down the line rather than a polite brochure. Those opinions are on the site already, so you don't have to take the claim on trust; the vendors we back are listed right there with the reasoning attached.
The trade, downsides included
None of this makes us the right fit for everyone. A marketplace will out-run us the week you want some unfamiliar vendor onboarded in a hurry, and we've got nothing that spins up a fringe SKU at two in the morning without one of us involved somewhere. A larger MSP that already runs a serious engineering practice in-house, and mainly wants a clean procurement rail with a single monthly invoice, will very likely be happier there than with us. We made our peace with that a fair while ago, and we'll cheerfully point those partners at a marketplace ourselves. The ones we're actually built for tend to be after something else.
Mostly what they want is a way off the treadmill of competing on price, and that only opens up once you know a small set of products well enough to hold a view a client will pay actual money for. The partner programme is shaped around that idea more than around volume. For partners across Australia, New Zealand, Fiji and Papua New Guinea it usually looks fairly ordinary from the outside: somebody local who knows the lineup, quotes put together by the same people who understand how the parts sit together, and on the Keepit side of things, honest margin visibility through KCM, the Keepit Channel Meter so nobody's pricing a deal by feel. That whole arrangement only works because the list is short enough for the knowledge behind it to be real, which is the entire point.
If that's nearer to how you'd rather run things, have a look at how the partnership works and we'll go through which of our vendors genuinely fit what you sell, and which of them honestly don't.