About us

Built by practitioners.
For people who run real systems.

We have been the people carrying the pager. That experience shapes everything we build.

A Melt Cafe branded coffee cup in front of monitors displaying observability dashboards

What we believe

Founded in Leeds by Rob Chapman and Peter Holmes. Between them, decades of platform and reliability experience across UK enterprise, finance, public sector, and observability vendors. The Melt Cafe is the consultancy they wish had existed when they were on the other side of the table.

Patterns from production

Our playbooks come from years of running critical systems at scale, not from vendor whitepapers. You benefit from what we have already seen work elsewhere.

Fewer incidents, not just faster recovery

Most incidents have signals hours before they fire. We help you spot them earlier, so the next outage is something you noticed before your customers did.

Business questions first

We start with the questions the business actually needs answered. Dashboards and alerts follow from there, not the other way around.

Cost discipline

Observability bills creep. We help you keep what is load-bearing, lose what is not, and document the difference so the next quarterly finance review is short.

Knowledge transfer

We work alongside your engineers, not over their shoulders. The aim is for your team to stop needing us, not to lock you in.

Managed service if you want it

If you would rather not run the platform yourselves, we will. A monthly retainer covering monitoring, alerts, and the inevitable 2am pages.

How we work

Engagements vary in shape. Some are a week. Some are a year. They tend to follow this rhythm.

Observability With Grafana book cover
  • Listen first

    We spend the first week with your team and your tooling. Before we suggest anything, we want to know what is actually broken versus what people think is broken.

  • Plan what is worth doing

    Not every fix is worth the engineering cost. We rank options by impact and effort and let you choose. We will tell you when the answer is to do nothing.

  • Deliver in the open

    We work in your tools, on your tickets, with your team in the channel. No black-box deliverables. You see the work as it happens.

  • Hand over and stay reachable

    Engagements end. We document what we did and why, train the team, and stay available for the next time you need a second pair of eyes.

The founders

Two veterans of large platform teams who wanted to do observability properly, for traditional businesses that need their systems to just work.

Rob Chapman

Rob Chapman

Three decades on the full application lifecycle, including spells at the Environment Agency, BT Global Services, Microsoft, Anaplan, and Grafana. Spends his time on cost-aware architecture and the cultural side of observability.

Peter Holmes

Peter Holmes

Twenty years across digital operations, platform engineering, and SRE leadership. Roles at Boots UK, Fujitsu Services, Anaplan, NHS Digital, and Grafana. Particularly focused on day-two operations: keeping things running smoothly and affordably after the project ends.

Get in touch

If you want to know whether we can help, the cheapest way to find out is a thirty-minute call.