Method

The playbook, not the pitch.

This is the process we run on every engagement: the stages, the gate between them, and the things we refuse to do. We publish it because a three-person studio has nothing to sell but judgement, and judgement you cannot inspect is just a claim.

How it runs

Five stages, one gate.

The assessment: two to five days, fixed price, fixed scope. Nobody enters the embed that follows without a written decision.

01

Access

Day zero, before we arrive

Credentials, a systems inventory and a stakeholder map.

02

Shadow

The first days on site

A record of how the work actually happens, taken from the people doing it, not the people describing it.

03

Map

Running alongside the shadowing

A gap map: how the business runs against how it says it runs, with every line tied to a name or a file.

04

Build

Every engagement, at every length, including two days

A working system, built against your real data, not a mockup of one.

05

Show and hand over

The last day

A ranked list of what else is worth building, and a written decision on whether to continue.

The gate

One hard gate. Five questions, answered with evidence.

At the end of the assessment we write a go or no-go, answering five questions with evidence rather than impression. A file reference or a quote answers each one, or the answer is no.

  1. Is there a named person in your business, not just an executive sponsor, whose week actually changes if this works?
  2. Is there a committed use case, rather than a general mandate that has no way to fail and so never finishes?
  3. Is there a named technical contact on your side, so integration questions do not route through whoever has a spare hour?
  4. Does the best opportunity we found clear the cost of the embed? If it does not, we say so.
  5. Is the system of record trustworthy enough to build against, or would that surface in week six at ten times the cost?

The principal who was not on your engagement reviews the decision and writes their own sentence, not a signature. The two people who spent days inside your business, and are proud of what they built, are the worst placed to judge whether to leave. Walking away is an expected outcome, not a failed one: the assessment was paid for, something ran on your data, and you were told something true.

If the answer is yes, the embed runs for roughly twelve weeks, with three dated guardrails: the first integration working inside 14 days, production inside 90, and handover in week twelve. Disengagement is scheduled at kickoff, not negotiated at the end.

What we refuse to do

The trust is in what we will not do.

We do not record hours spent on your engagement.

No hours-spent field, no utilisation figure, no billable ratio, anywhere. The moment that number exists, someone attaches a target to it, and an engineer optimising for hours is a consultant, whatever the title says.

The decision to continue past the assessment is not written by the people who ran it.

A principal who was not on your engagement reviews the gate and writes their own sentence, not a signature. The two people who spent days inside your business, and are proud of what they built, are the worst placed to judge whether to leave.

We do not let a twelve-week embed drift into an open-ended one.

What counts as finished is written into the agreement at kickoff, not negotiated once we are inside. A longer engagement is a separate decision, with its own scope and price, made after this one is done.

We do not hand you a claim we cannot point at.

Everything in a document we give you is tied to a name, a quote or a file, or it is marked an assumption with an owner against it. Never neither.

We profit if you decide your business needs this.

We sell assessments and embeds. When we tell you your business needs one, that conclusion earns us money, every time we say it. If a consultant has ever told you what you wanted to hear rather than what was true, you already know why that matters here too. Weigh what we say with that in mind.

If we walked on Monday

What would still be running on Friday.

  • The source, in a repository you own outright, not one we host on your behalf.
  • A runbook: how the system works, how to change it, and who to call when it breaks.
  • Monitoring already wired in, so a failure is visible before your customers find it for you.
  • At least one of your own people trained to run it day to day.
  • Nothing left that only we can operate.
Next

Ready to see what a two-day assessment finds in your business?