A Combat Sports Promotion Built Its Own Streaming Platform in Five Months

Most people who want to stream their own events start by pricing platforms, and stop when they reach the per-subscriber line. The arithmetic is discouraging in a specific way: the better your card does, the more you owe. You are billed for your own success, every month, permanently.

This promotion did not start there. It started with a domain and a cheap virtual server, and five months later it had a working platform with just under 400 accounts on it. Everything below comes from our own billing records rather than from anybody's recollection. The promotion is anonymous at their preference, and a few figures are rounded.

What they had on day one

Not much, and that turns out to be the point. In early March they joined a discovery call and, before the second email in the thread, had already bought two things: a domain, and the basic plan from a budget hosting company at around $40 a month.

That was the entire stack. No platform, no player, no apps, no way to take money. What they did have was an audience and events worth watching, which is the one part nobody can sell you.

Two days after that call, their player was live on their own domain. If you want the sequence in detail, how it works walks through the same steps.

Why renting was the wrong shape for this

The obvious move is a hosted platform, and for a lot of businesses that is the right answer. For combat sports it fits badly, and not because of quality. Three things about this category fight the rental model directly.

• The billing runs opposite to the business. Fight promotions are spiky. A card lands, a few thousand people arrive in a week, then it is quiet until the next one. On a per-subscriber plan every spike becomes a bill, and the bill arrives whether the spike converted into anything or not.
• Pay-per-view is the model, not a feature. Selling one event to somebody who will never subscribe monthly is the core transaction in this sport. On rented platforms single-event sales are a tier you climb to, when they exist at all.
• The audience is on phones and televisions. At peak this platform served 465 connected devices against roughly 390 accounts — people watch on more than one screen, and they watch on the sofa. Apps are not a nice-to-have in this category; they are where the event is consumed.

Put against public pricing, the third point is the expensive one. Apps are the thing rented platforms charge the most to unlock (Uscreen public pricing, August 2026):

Read the last column as well. A transaction fee is charged on ticket sales, which in a pay-per-view business is charged on the busiest night of the quarter. The plan that ships apps starts at $449 a month, and still takes a percentage of every ticket.

What we did

We installed the platform on the server they already owned. That sentence contains the whole cost difference, so it is worth being literal about it: the machine is theirs, rented directly from their host at their price, and nobody sits between them and it.

• A web player on their own domain — not a page on somebody else's platform with their logo in the corner.
• A package structure with more than one price — a monthly plan and a discounted annual one living in a single package, so fans choose at checkout instead of being sent to two catalogues.
• Checkout wired to their own payment account — money moves from the viewer to the promotion. We never hold it, which means no payout schedule and no reserve after an event.
• Subscriber management they run themselves — adding, editing and cancelling accounts without opening a support ticket with us.
• Multi-device access — one account, several screens, which is how a household actually watches a card.

The result, in money

Here is the growth, taken from daily platform snapshots rather than from a dashboard screenshot:

And here is what the platform charged for it. The fee is $0.20 per active subscriber per month, calculated from daily snapshots, so it follows the real audience rather than a plan tier:

Their busiest billed month cost $61.86, plus the server they rent themselves. For comparison, at the same subscriber count on public pricing: App Essentials is $449 a month billed annually plus $0.99 per member, which at 375 members is roughly $820 a month. The Growth plan looks cheaper at $149, but carries a $1.99 per-member fee and ships no apps at all — for this audience that makes it not a cheaper option but an unusable one. Our pricing is here and the full comparison is here.

The gap is not a discount. It is the difference between owning the shelf and renting it, and it widens every month, because one side grows with your success and the other does not.

"But I don't want to run a server"

Nobody does. It is worth separating two things that usually get bundled together.

Renting a virtual machine is not system administration. It is a checkout page and a monthly card charge, the same as any other subscription, and the plan this promotion started on is the cheapest tier of a budget host at around $40 a month. What people actually mean is that they do not want to maintain one — and that is fair, so we install it, configure it, and it stays configured.

The trade you are making is straightforward: you take on a machine that costs tens of dollars a month, and in exchange your platform bill stops being a function of your own growth. If the underlying delivery side is what you are curious about — encoders, restreaming, the media server itself — that is FastoCloud's territory rather than this article's.

What owning it lets you add

This is where owning starts to matter more than the monthly saving. On a rented platform you file a feature request and wait. On your own, you commission something once and it is yours.

• Pay-per-view on a single event — priced as a day pass, sitting on the same package as the monthly plan, so a casual buyer and a season-long fan are served by one catalogue.
• A free tier, or none at all — a package with no price attached is free, so you can run a funnel with free preliminary cards and paid main events, or skip free entirely. That is a decision you make, not a plan feature you buy.
• Your own apps on mobile and TV — published under your name, bought once at a fixed price rather than unlocked by moving up a plan forever.
• Pages Google can index — fighter pages, event pages, results. Rented platforms generally have nowhere for that content to live, so every ticket has to be bought with ads or social reach.
• Whatever your federation needs — rankings, weight classes, event archives. The feature list covers what ships as standard; anything beyond it is scoped per project.

To keep the picture honest: the figures above cover the platform and the server only. Apps and custom modules are quoted separately, per project, and none of them are folded into the numbers you just read.

Questions we get from promoters

How long until we can sell a ticket?

The player here was live two days after the first technical call. Content, pricing and artwork take longer than the platform does, and building the audience takes longest of all. This promotion arrived with events and followers already in hand, which is the correct order to do it in.

What happens on a big fight night?

Capacity is bounded by your server rather than by a plan tier, so you can size the machine up before a card and back down afterwards, paying your host for the days you actually need. The platform fee tracks subscribers, not traffic, so a viewing spike does not by itself produce a bill.

Do you take a cut of ticket sales?

No. Payments go to your own payment account, so there is no revenue share, no payout schedule and no reserve held after an event. We charge per active subscriber and nothing else. In a business where one night can be most of a quarter, waiting weeks to be paid is a real problem.

Can we sell one event without a subscription?

Yes, and it is standard rather than a higher tier. A package can carry several prices at once, so the same event is a day pass for one buyer and part of a monthly plan for another, decided at checkout. That mix is the normal shape of combat sports revenue.

We already have a following on social. Does that transfer?

It is the only thing that reliably does. Software can be installed in an afternoon; an audience cannot. Every promotion we have seen grow quickly on its own platform arrived with people already watching it somewhere else.

What if we outgrow the server?

You resize it. That is a hosting decision made with your host, not a renegotiation with us, and it does not change what the platform costs. Growing out of a machine is a good problem, and it is the one problem a plan tier handles worst.

If you are pricing platforms for a promotion and the per-subscriber line has started to look like a tax on your own growth, it is worth twenty minutes. Check the pricing, or see how this compares to Uscreen.