The diagnostic

Every engagement begins with a diagnostic.

Three lenses, one written finding, and a scope priced against it. Proceed or not, you keep the finding.

Duration
2 to 4 weeks
Fee
USD 4,999, fixed
Deliverable
A written finding

Process, data, unit economics.

One of the three is usually binding. The diagnostic establishes which, what it costs, and what it would take to move.

01

Process

Binding

Is this a process, or a habit?

Software applied to undefined work encodes one person's version of the job and removes the discretion that was holding the operation together. Before anything is scoped, we establish whether the work, and the systems it runs through, are specified tightly enough to be encoded.

We ask: Could two people perform this work and produce the same result?

02

Data

Clear

Would two systems agree on this number?

Most integration failures are definitional rather than technical. When a business holds more than one definition of a customer, an order or a case, no integration can reconcile the disagreement, and it surfaces later as a reporting problem. We establish whether the business holds one definition or several.

We ask: Could two systems answer the same question and return the same number?

03

Unit economics

Clear

Does the build clear its own cost at this volume?

A system carries a running cost that is rarely priced at scoping: inference, infrastructure, monitoring, human review, and the maintenance that continues for as long as the system exists. We establish the volume at which the build pays for itself, and whether the business operates above that line.

We ask: What does this cost to run in year two, and at what volume does it clear it?

An illustration. Which lens binds is what the diagnostic establishes.

How it runs

What the two to four weeks actually consist of.

Run by the engineers who would build against it, which is why it can be short. No discovery team, no second explanation.

Three phases, one finding

  1. 01

    Access and interviews

    We sit with the operators who run the work rather than with the sponsors who described it. The gap between those two accounts is frequently where the constraint is.

  2. 02

    Evidence

    We pull the numbers ourselves rather than accepting the reported ones. Where a figure cannot be reconstructed from something we can see, it does not go in the finding.

  3. 03

    The finding

    Written, delivered, and walked through with whoever has to act on it.

Illustrative proportions

What we need from you

  • Read access to the relevant systems.
  • 8 hours a week from the people who perform the work.
  • Whatever has already been written internally about the problem.

What we do not need

A brief, a requirements document, or a vendor evaluation. If one exists we will read it, and say where the finding disagrees.

What it returns.

01

The constraint that is binding.

Stated plainly, with the evidence for it, in terms the operators who work inside it will recognise.

02

What that constraint currently costs.

Quantified against the operation as it runs today, not against a projection.

03

What can be built against it.

Sequenced by payback rather than ambition, with the work that pays first placed first.

04

The line in the P&L it moves.

Named explicitly, with the volume at which the build clears its own running cost.

Short, and written for someone who was not in the interviews. Clients take it to boards, lenders and their own technology teams. Every figure in it traces to something we saw, not something we were told.

Why it exists

Our advice is independent of what we build.

We charge for the finding as a document, not as the opening phase of a build. The fee covers the work whatever it concludes, including less than you asked for, or nothing at all.

A free assessment is paid for by the build that follows it. Our fee is not credited against one, because that would bring the incentive back.

Free assessment

  • Who absorbs the cost The firm
  • What pays for it The build that follows
  • What a "no" costs the firm The entire engagement
  • Likely conclusion Build something

Paid finding

  • Who absorbs the cost The client
  • What pays for it The document itself
  • What a "no" costs the firm Nothing
  • Likely conclusion Whatever the evidence supports

What it costs

A fixed fee, agreed before the work starts.

USD 4,999

Fixed, and agreed before work starts

Not adjusted by what the work finds. The fee is the same whether the finding recommends a build, a smaller one, or none at all.

The scope that follows is priced separately, by the team that would deliver it.

It is useful when

It is not useful when

  • The brief is fixed and the constraint is not open to question
  • The decision has already been made and what is wanted is a document that supports it
  • The operation the system would run inside does not exist yet, so any answer would be a forecast rather than a finding

A worked example

What the finding changed.

A consumer lender sent every approval through a manual queue, whatever the risk. Throughput was capped by the number of reviewers, and the review criteria lived in the reviewers rather than in a policy.

The diagnostic ran all three lenses. The binding one was process, and the test was whether the reviewers applied their criteria consistently enough to write down. They did.

The criteria were written down and encoded, and manual review was kept for the cases the policy refers to a person. Most decisions no longer wait in a queue.

Before · Approvals reviewed by hand, 100%
After · Approvals reviewed by hand, 40%

The difference is what the diagnostic found.

Start Here

Start with the question, not the brief.

For companies operating above $1M in revenue with a specific commercial problem.

A 30-minute call. You describe what you are considering building. We tell you which of the three lenses we would run it through first, and whether we are the right firm for it. No fee, no deck, no follow-up sequence.

Get an AI Strategy Consult
  • 0130 Minutes
  • 02No Fee
  • 03No Deck