There's a quiet bait-and-switch in a lot of cloud "support." You sign up expecting an operations partner and you get a ticket queue — somewhere to log a problem and wait. That's break-fix. It's fine for some things. It is not the same as having someone run your cloud, and the gap shows up at the worst possible moment.
If you're evaluating providers, here are the questions that separate the two — and they're worth asking directly, because the answers are revealing.
"Who owns the problem at 3am?"
With break-fix, you do — they advise, you execute. With a real operations partner, they take control, coordinate the recovery and keep you informed. Ask exactly what they do when an incident fires out of hours, and who is awake to do it.
"What's your response time, and is it a commitment or an aspiration?"
A serious provider will give you a defined response time for high-severity incidents and stand behind it. Vague answers here tell you everything.
"Is this SRE-led or ticket-led?"
The hyperscalers' own support plans are largely reactive and case-based — they keep the platform available, but you still run your environment. A modern operations partner works the way good engineering teams work: monitoring, runbooks, blameless reviews and continuous improvement, not just closing tickets.
"What do you do when nothing is broken?"
This is the one that exposes break-fix fastest. If the answer is "wait for the next ticket," it's break-fix. A real partner is improving things between incidents — reliability, security posture, cost, automation — so there are fewer incidents to begin with.
"Can I see what you see?"
Operating in the open is a good sign. Shared dashboards, transparent post-incident reviews, real telemetry you can look at yourself. If visibility is a black box, accountability usually is too. We build on Datadog precisely so the customer can see the same signals we do.
"What do your accreditations actually mean?"
Logos are easy. Ask what a credential required and what it lets them do for you. Being a Datadog Advanced Partner and the world's first "Powered by Datadog" accredited MSP means something specific about how we operate observability — you can read more about what it means in practice on our partners page. A wall of unexplained badges means less.
The underlying point: you're not buying a place to send tickets. You're buying an operating model and the people who run it. Ask the questions above of anyone you're considering, including us. The right partner will be glad you did.
Critical Support is built around that distinction on purpose — an embedded engineering partnership, not a queue.