The costs nobody owns

Past about ten engineers, a SaaS platform starts paying for three things that no one is responsible for: capacity it is not using, on-call that runs on habit rather than a rotation, and an audit trail that lives in someone's head instead of the repository.

None of those appears as a line item, which is exactly why they survive. The cloud bill is one number, the incident is one outage, the audit is one date. Nobody ever adds them up.

This audit adds them up. We read the cloud account, the clusters, the pipelines, and the runbooks, then report what each problem costs in dollars, in risk, and in hours.

It is read-only. Nothing in your environment changes, we never need write access, and your team's time goes to a 30-minute kickoff and a 60-minute walkthrough at the end.

What we look at

Four surfaces, one report.

Every finding is tied to one of these, so the report reads as a plan rather than a list of complaints.

Cloud spend

Rightsizing, commitment coverage, storage tiers, egress, idle non-production environments, and the resources nobody remembers creating. Waste is usually concentrated in a handful of items, so the fix is smaller than the number suggests.

Reliability

Whether your SLOs are written down or assumed. Alert volume measured against alert value. The on-call rotation, its escalation path, and what actually pages at 3 a.m. Single points of failure, and whether the backups have ever been restored.

Delivery

How much of the platform is genuinely in Terraform, and how far the running system has drifted from it. How long a deploy takes, what a rollback looks like, and whether a new engineer can ship on their first day.

Audit trail

The controls a SOC 2, PCI DSS, or HIPAA auditor actually tests: access, change management, logging, and encryption. Not the policy document — the evidence you could hand over next month.

What you get

  • Findings ranked by dollars, risk, and effort, not by how interesting they were to find.
  • A dollar figure per finding, and a total at the top of the report.
  • A 90-day plan: what to do first, what to do next, and what to leave alone.
  • The evidence behind each finding — queries, configuration, and screenshots an auditor would accept.
  • A 60-minute walkthrough with the engineer who did the work, not a salesperson.

What it costs

Three scopes, from $2,500. Focused covers one cloud and one cluster; Full covers two clouds and three clusters over a week; Extended adds a second week and a framework mapping. Every scope is fixed price before we start, with no hourly billing.

On the Azure estate I run, rightsizing and storage automation took monthly spend from $40,000 to $18,000, about $264,000 a year. That is one platform and it is not unusual. One finding of that size pays for this audit many times over — and if the audit finds nothing worth acting on, the report says that too.

The report is yours whether or not you hire us for the work that follows. Implementation is a separate engagement, quoted after the audit when the scope is known rather than guessed.

Scopes

Start at the size of platform you have.

The Focused fee credits toward a Full or Extended audit within 90 days.

Focused

$2,500 Two to three days

One cloud account and one cluster. The spend and reliability picture, without the compliance mapping. This is the tier for a team that needs the numbers before it can justify anything larger.

  • Cloud spend for one account, last 90 days
  • SLOs, alerts, and on-call for the production cluster
  • The top ten findings, ranked and costed
  • A 90-day plan
  • A 60-minute walkthrough

Not included: Delivery pipeline and drift review, Compliance evidence mapping, and More than one cloud or one cluster.

Seed and early-stage teams, or a first look before a larger commitment.

Extended

$7,500 Up to two weeks

For a multi-cloud platform, more than three clusters, or a team that needs the control mapping in the same engagement.

  • Everything in Full, at larger scale
  • Control-to-platform mapping for one framework
  • An evidence index for the technical controls in scope
  • An auditor-ready walkthrough

Not included: Implementing the controls; that is the platform project.

Multi-cloud platforms and teams preparing for a specific audit.

Who it is for

  • 10–150 engineers running Kubernetes on Azure or AWS
  • A SOC 2, PCI DSS, or HIPAA audit on the calendar
  • A board or a CFO asking what the cloud bill actually buys
  • A new VP of Engineering who inherited the platform and the bill
  • An incident that should not have been possible

Who it is not for

  • Pre-product teams with no meaningful cloud spend yet
  • Anyone looking for staff augmentation or a body in standup
  • Teams who need the findings implemented the same week — that is the next engagement
  • Companies that need a provider to take over operations entirely

How it works

One week, four steps.

  1. Scoping call

    Thirty minutes. We agree the scope, sign the NDA, and confirm the price. No charge and no obligation.

  2. Access

    A read-only cloud role, Kubernetes read, repository read, and a billing export. Nothing that can change your environment.

  3. Analysis

    Four working days. We follow the money and the failure modes, and we test the claims in your runbooks.

  4. Report and walkthrough

    The written findings, ranked and costed, plus an hour to walk through it and argue with the priorities.

The honest part

No access we don't need, and no work we can't justify.

A fixed price means there is no reason to invent findings. If the platform is in better shape than you think, the report says so and you have a document that proves it to your board.

Read-only

We never hold a credential that can change your production environment. That is a design choice, not a promise.

Fixed price

Quoted from $2,500 before work begins. No hourly billing, no change orders for finding more than expected.

Yours to keep

The report, the evidence, and the plan are yours whether or not you hire us for the implementation.

Book the audit.

From $2,500, read-only access, fixed before we start. Tell us what you run and we will confirm the scope and a start date.