As most of you know, late last year I started to publish an open source directory of FinOps and FinOps-adjacent companies. As of today, we are just under 300 companies in the list and I've talked to ~20% of the companies and looked at what they actually do, not what the landing page says. So between the interviews and reviewing all the companies, you stop seeing ~300 logos and start seeing the a pattern, made over and over, at different points on the same timeline.

This week I wanted to share one of the clearest patterns I have seen to date. First, reporting has become table stakes. Every serious tool has it now. Show me the spend, break it down by account and service and team, draw the trend. When every product in a category does the same thing, that thing isn't a differentiator anymore. It's the cost of showing up.

So how are the solutions differentiating in the market. The front line right now is AI cost. That's where some of the leading companies are trying to pull ahead, and it's where every new entrant is attempting to go. Which would be fine, except I've seen exactly this arc before, and so have you.

In the early-to-mid 2010s, cloud cost reporting was a differentiator because it was the easiest, most logical place to start. A tool that pulled your billing data and showed it back to you cleanly was worth buying. Then everyone built it, and it went table stakes. In that race the industry solved the easy part five times over and skipped the hard one. A few teams pushed past reporting into real control, but it took serious focus. Which is why, ten years on, most companies still can't tell you who owns a given line or stop a cost before it happens. The reporting got solved. The control never did.

Now we're in the AI cost version. Same starting move, just 10 to 12 years later. The first thing almost everyone is shipping is a view of spend. Spend by model, tokens by team, a chart of your inference bill climbing. It looks like differentiation because the resource is new. It isn't. It's the cloud reporting arc starting over on a different meter. The safe bet is that AI cost reporting is table stakes inside a year or two, and the buyer still can't control the spend, same as cloud.

That's the pattern the directory makes visible. The ecosystem sprints to the new layer before it ever controlled the last one. It did it with cloud. It's doing it again with AI, and this time the vendors are running the same play. Is that a problem? It can be, but not the way you'd expect. It comes down to what a vendor does next.

Here's where I believe things can get interesting.

Some of what's branded "AI cost" is a dashboard with a new label. The token chart is real, the data is real, and it is still just a view. It tells you what you spent after you spent it. If you already had cloud reporting, you already knew how this works. Nothing about it changes a decision. It's Visibility Theater with a fresh coat, and the tell is simple. Ask what you're supposed to do with the screen and the answer is "look at it."

And some of it is a real attempt to differentiate, and those teams are working on the part that's actually hard. Allocation that holds when the resource is a token instead of an instance, so spend traces back to a team or a feature or a customer. Cost tied to an outcome, so the number means something other than "big." A control that meets the spend before it happens instead of a report that explains it after. That's not a prettier view. That's a different product.

Its critical to know that the 20% of the companies I have spoken with, only a handful are providing meaningful AI usage, and of those, less than a handful have been able to prevent or help govern the problem.

The reason the dashboard version doesn't hold is structural, not a matter of quality. A view is a differentiator that expires. The moment the next vendor ships the same chart, and they will, it's table stakes, and you're back to competing on price and logo. That's the whole story of cloud reporting compressed into a sentence.

AI made the expiration faster. Building a spend dashboard used to take a team a quarter. Now it takes a lot less, because the same AI everyone's reporting on will help you build the report. So the easy layer fills up quicker than it ever did, which means differentiating on the view buys you less time than it did last cycle. Cheap to build was never a reason to build it, and it's a worse reason to build your whole pitch on it.

Control is the differentiator that doesn't expire, because it's the thing nobody has finished. Owning an action holds its value. A prettier chart doesn't.

If you're buying or looking to buy, you now have a clean test, and it's the same one I keep coming back to. When a vendor says they differentiate on AI cost, ask whether it changes what happens or just shows you what happened. Then make them prove it.

If the answers are all "here's the dashboard," you're buying reporting with a new label, and reporting is free now.

If you're building product, whether you're a vendor or a team standing up your own tooling, the same test points the other way. The view is table stakes. Build it, but don't mistake it for your edge. The edge is the action underneath it, and that's the part close to 300 companies still haven't finished (some more than others).

The count was never the story today. I don't track 300 companies because the market is healthy. I track 300 because the discipline hasn't settled, and it hasn't settled because reporting keeps getting solved while control keeps getting skipped. AI cost is the newest place that's about to happen. Watch which vendors sprint to the dashboard, and which ones go straight for the thing that's actually hard.

Written with the help of AI. All the ideas expressed are mine and mine alone.

If you would like your company included in the Spotlight, contact the CloudXray AI Team

Company: CloudXray AI

Category: Managed Services & Consulting

What They Do: FinOps Consulting & Advisory Services; Owners & maintainers of the single largest FinOps company directory (finops.cloudxray.ai)

Why It Matters: Companies still need guidance on implementing FinOps and understanding the landscape of companies that exist