Skip to content
Flamingo’s AI Token Pricing Model Shifts Margin Risk for MSPs – With Rich Freeman artwork

Flamingo’s AI Token Pricing Model Shifts Margin Risk for MSPs – With Rich Freeman

Business of Tech: Daily 10-Minute IT Services Insights

September 24, 2026

A growing shift is materializing in the IT services market as software vendors move away from traditional fixed software pricing and towards models based on AI token consumption, effectively redefining the sources of margin and exposing providers to variable cost structures.

Speakers Dave Sobel, Rich Freeman

TopicsTech NewsNewsBusiness

Dave Sobel (0:14)

Software just got priced at a dollar a device. The money moved to the tokens. There's now a vendor selling to MSPs that makes almost none of its margin on the software and almost all of it on AI consumption. And its founder thinks the entire industry ends up there. At the same moment, the largest AI vendor in the world is quietly starting letting its biggest customers off the consumption meter altogether.

Rich Freeman spent the past two months reporting on both ends of that trade. Today, we're going to work out who's holding the middle. Welcome to the Business of Tech Lounge. This is where we break down what's changing in the IT services market, what it means for providers and vendors, and what to actually do about it. If you're running or supporting an MSP, this is about making sense of the environment you're operating in, not just the headlines. Now, to make this conversation possible, a message from our sponsor.

Every tool you bolt onto Microsoft 365 is another console, another login, another thing that eats your team's time. Proofpoint 365 Total Protection gives that time back. Security backup and compliance from Microsoft 365 in one multi-tenant platform. Built MSP first. Enterprise-grade protection tailored for your small and mid-size clients with the operational simplicity that protects your margin.

Modern threats covered, resilience built in, one pane of glass. Built on Hornet Security, now part of Proofpoint. Learn more at proofpointtotalprotection.com.

So my guest today is Rich Freeman. He writes Channelholic, co-hosts the MSP Chat Podcast, and helps lead the consultancy Channel Mastered. Before that, executive editor at Channel Pro. He's been reading this market longer than many of the companies in it have existed. Rich, welcome back.

Rich Freeman (2:16)

Well, thank you very much, Dave. I am delighted to be back.

Dave Sobel (2:20)

Well, I'm excited to have you. Audience, if you've got questions, drop them in the chat as you go. We'll take them in real time. Rich, I'm going to dive in. I want to start with one of your more recent pieces here because I think it's the most concrete thing anybody's published on this. Flamingo, a young startup you first wrote about in January. They build the whole MSP stack from scratch. That's RMM all the way through SIM, and charges a dollar a device.

10 million AI tokens included, and you pay by consumption after that.

And that's how GitHub and the SAP bill as well, and how nobody I think in the channel does. Now, Asarov told you plainly, the software is not where the money is. He says it's in the tokens. Now, most of his 400 partners are burning 10 to 20 times that allotment every month. So the dollar isn't the price. It's the part of the bill you can see, but the real cost is the overage. It's variable and the rate that would let anybody calculate it really isn't published anywhere. And by year end, they're gonna host their own models and partners get twice the tokens for the same money.

I read that as a margin statement. Today, they resell frontier tokens on a thin spread. Once they own the inference, their cost collapses, but the price stays pegged. So I kind of want to ask you to get your take on it. Is Flamingo a software company that happens to build by consumption? Or is it an inference reseller with a free RMM attached to do customer acquisition?

Rich Freeman (3:54)

I'm gonna go with the former of those two options there. And by the way, Dave, let me correct you ever so slightly about the pricing model.

It is a dollar per device per month right now with 10 million tokens included. If you pay annually, it's 80 cents.

Dave Sobel (4:12)

Okay, got it.

Rich Freeman (4:13)

And 25 million tokens instead of 10 And it's also worth knowing that's where they're at right now. So they are on a roadmap that will eventually get them to 20 different modules within the suite. Eight of those are shipping right now. That's what you're paying a dollar or 80 cents for. Down the road, when all 20 are out there, the pricing will go up to $3 and then presumably some discounted amount if you're paying for the year in advance.

That whole model basically tells you where Flamingo thinks the value is in what a software vendor for MSPs does right now. According to Flamingo, they make money at 80 cents per device. Not a whole bunch, but they're not subsidizing you on the software. But the whole idea is, from their perspective, and they're not the only ones in the industry thinking this way, is that in an era when AI coding tools make it possible to duplicate what more or less any application out there is doing, pretty quickly, the value that you're delivering to your customer, whoever that customer is, in this case it's MSPs, isn't coming from the software. In Flamingo's view, the value that they're providing is in the automation. The automation is enabled and delivered by the AI.

32 more minutes of transcript below

Thousands of transcripts fetched by people building searchable podcast archives

Fetch the whole transcript

The demo key returns a sample episode in full, no card needed:

request
curl -H "x-api-key: pt_demo" \
  https://spoken.md/transcripts/1000651996090

Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.

From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.

Using your own key:

request
curl -H "x-api-key: YOUR_KEY" \
  https://spoken.md/transcripts/1000791355112