What is the best way for sports teams to get all their data in one place?
Over the last six weeks I've had the same conversation with a lot of clubs. Football, rugby, a couple of governing bodies. Different sports, same question: we've got data in five systems and we want it in one place, how do we do that?
Until now, when you strip it back, there have only been three answers.
Hire someone and build it yourself.
Pay a consulting firm to build it.
Buy a product that says it does it for you.
I've seen all three play out and I've been on the selling side of the second one for a long time.
Option one: hire in-house
The pitch is obvious. Get a data engineer, maybe a data scientist, have them in a seat in the building and you own the thing you build.
Here's what usually happens. The club doesn't have a data team, it has one hire, maybe two, and the person managing them has never built data infrastructure. So nobody has the skills or experience to judge whether the work is any good or how experienced the hire really is, and there's no way to find out until the thing either works or it doesn't.
The hire, meanwhile, doesn't know sport. They don't know what a Monday morning readiness meeting needs or what the analyst is going to ask of the data on match day plus two, so they build what looks right from a technology point of view. That takes six to twelve months. Then the performance staff try to build a report on it and find out it isn't shaped the way they need. A lot of money has gone out, there isn't much to show for it, and the conversation about what to do next is an awkward one.
Then there's the other problem, which nobody budgets for. When that one person leaves, they leave with the keys. There's rarely any documentation. The next hire spends months working out what's there and then, more often than not, unpicks it and starts again.
I commented on a post on LinkedIn last week that this is a direction problem rather than a capability one. The hire could code, what nobody could give them was a brief, and I still think that's most of the story.
Option two: hire a consulting firm
I owned and ran a consulting firm for years before Axon, so I'll be careful here, because the instinct behind hiring a consultancy is sound. You want people who've done it before and can deliver, and that's what you think you're paying for.
What I know from the inside is that the senior person in the room when the job is sold is very often not the person who builds it. Delivery goes to someone more junior, and that person doesn't know sport either, so you get the same problem as option one above. Things get scoped before anyone really understands what the reporting has to do, the build takes six to twelve months, and when the output doesn't fit the team's needs, the consulting firm will point out that it was delivered to scope. Every change after that is extra.
I've seen this a lot. I know of teams and governing bodies that have spent six-figure sums, sometimes seven, building a data platform, found out the result wasn't fit for purpose, and are now going through the same process again with a different firm to rebuild it. Ouch.
There's a newer version of this problem too. Warehouses built this way three or four years ago were built for dashboards. They weren't built for an AI tool to query, and a lot of them are now being upgraded or replaced to make that possible, which is a second build on top of the first.
Option three: buy an athlete management system
Most AMS products have been around for ten, fifteen even twenty years and have sold for most of that time on the promise that they'll centralise your data. For collecting and displaying it themselves, they mostly do.
The issue is how they're built. An AMS was designed to hold the data it collects and show it back to you inside the platform. It was designed before cloud data warehouses were normal and well before an AI tool could sit on a warehouse and answer questions directly. So the data lives in the vendor's system and comes out through an export or an API, and anything you want to do with it alongside your GPS, your force plates, your match data, you do somewhere else.
We've worked with a number of clubs that bought an AMS for exactly this reason and realised a year or two in that it doesn't give them a central dataset at all. So they go back to option one or option two, hiring someone or paying someone to pull the data out of the AMS and join it to everything else. So they've paid twice, once for the product and once for the build around it.
The fourth option, the reason Axon exists
A lot of teams can't afford the way the first three go. I don't mean the headline price, I mean the drawn-out version where it costs more than anyone expected and still doesn't do what the performance staff needed.
What we built is the thing all three were trying to get to. A dedicated data warehouse per club, with the integrations to Catapult, VALD, STATSports, Opta, the wearables and the rest already built and already maintained, and we've hit and fixed most of the ingestion problems a club will ever run into, because we've been doing this since 2020. It's live in days, instead of months, and the risk of it not being fit for purpose is massively reduced, because this is the only thing we do.
On top of that sits the AI layer. The MCP connection carries the metadata an AI tool needs to understand sports data, so your staff point Claude or whichever tool they use at the warehouse and ask it questions. And when the team build reporting and analysis, usually an HTML report these days, Axon HQ is where it gets published, with access controlled and the data behind it staying current. We've seen clubs pass these reports around on a shared drive and it isn't good, for security or for keeping anything up to date. Published into HQ, the athlete management side, the warehouse and the reporting are one connected thing.
Which answers the obvious question about option three. Axon has the AMS functions too: athlete management, medical and availability, wellness forms, an athlete app. The difference is that they sit on top of the warehouse rather than instead of it. The club defines its own measures and builds its own reports and its own AI workflows on the data, and we run the platform underneath. In all three of the other options you end up living with somebody else's idea of what you needed, and it's usually a year in before anyone realises they'll never get what they need.
About the author
Jon Foote is the founder and CEO of Axon Perform. He has spent twenty years in data and analytics, the last six of them inside high-performance sport, working with the All Blacks since 2020 and with national teams and clubs across rugby, football and cricket since. Before Axon he founded and ran a data consultancy, which is where most of what he says about the second option comes from.