Home/Starter Services
Fixed-scope first engagementsA bounded first engagement that ends in a decision.
Every one of these has a fixed scope, a fixed fee agreed before it starts, and a date it finishes on. You get a specific outcome and the evidence behind it — not a retainer that quietly becomes the relationship.
Built for whoever owns the decision — a data or technology leader, an IT director starting from a standing start, or a line-of-business owner with a P&L and a problem. Anyone who needs a first engagement they can justify internally and stop cleanly if it does not work.
Several ways in. Every one of them ends.
- AI & Data Assessment Sprint
- GenAI Pilot
- AI Governance Quickstart
- Executive AI Workshop
- Data Platform Accelerator
- Cyber Risk Sprint
You leave with Two weeks · a scored picture, ranked gaps and sized costs
You leave with One function working under production conditions, with its evaluation harness
You leave with A named review path, a decision log, and a policy drafted for sign-off
You leave with A leadership team that agrees, and a shortlist of use cases worth funding
You leave with A working foundation, its pipelines, and the runbook to operate it
You leave with A scored posture, ranked exposures, and what closing each one takes
When you have not yet isolated the right intervention.
This is the most common first engagement, and deliberately so: it is the one that tells you which of the others you actually need — or that you need none of them.
AI & Data Assessment Sprint
Know what is actually there, what it blocks, and what the paths cost — scored against your own systems rather than a questionnaire you filled in about yourself. Everything else on this page is easier to choose once it is done.
- Readiness scorecard
- Gap register
- Order-of-magnitude cost view
- Prioritized next steps
One of these fixes what the sprint finds.
The Assessment Sprint scores six dimensions. Each engagement below repairs or deepens one of them, which is why its last page names the one you need. All five run today; the Assessment Sprint is simply the first with a full page of its own.
builds · One working use case
GenAI Pilot
One function, one use case, taken from definition to a working system under production conditions. The scope cap is the point: more than that and it stops being a pilot and becomes an implementation.
Worked examples: predictive maintenance in Operations · content and campaign productivity in Marketing · service deflection in Customer Experience · document processing in Finance and Legal.
- You leave with
- A running system, its evaluation harness, and what it costs at volume
- Scope
- One function · one use case · a capped number of processes, all three fixed in the proposal
fixes · Governance and review path
AI Governance Quickstart
A review path, an approval record and a policy your legal and risk teams will actually clear — built so the people who blocked you can sign it.
- You leave with
- A named review path, a decision log, and a policy drafted for sign-off
- Scope
- Scoped with you
fixes · Adoption and literacy
Executive AI Workshop
Your leadership team in one room, working out what AI is actually for in your organization and which use cases are worth funding.
For a full programme rather than a first session, see AI & Data Literacy.
- You leave with
- A shortlist of use cases ready to fund, and a leadership team that agrees on them
- Scope
- Scoped with you
fixes · Data foundation
Data Platform Accelerator
A lakehouse or warehouse foundation stood up properly on Microsoft Fabric, Databricks or Snowflake. Opens with a diagnostic of what you already have, so nothing gets rebuilt that did not need rebuilding.
- You leave with
- A working foundation, its pipelines, and the runbook to operate it
- Scope
- Scoped with you
goes deep on · Security and AI risk
Cyber Risk Sprint
The Assessment Sprint screens this dimension in three questions. This one goes deep on it, including the exposure your existing review was never designed to catch.
- You leave with
- A scored posture, ranked exposures, and what closing each one takes
- Scope
- Scoped with you
Something else entirely
If the thing you are stuck on is not on this page, say so. We would rather scope the right engagement than sell you the nearest one.
Tell us the decisionFour rules, and we hold ourselves to them.
Consulting goes wrong in predictable ways. Each of these constraints exists because it removes one specific way it goes wrong, and they apply to every engagement above.
01
Fixed scope
Written down and agreed before anything starts. If the scope genuinely has to change, that is a conversation before it happens rather than an invoice after it.
Removes: scope that expands to fill the budget
02
Fixed fee
Agreed in writing upfront. No hourly billing, and no change orders for work inside the agreed scope. You know the number before you commit, not after.
Removes: the bill nobody can predict
03
It ends in a decision
Every one finishes with something you can act on — a recommendation, a working prototype, a cleared policy — presented live, not emailed as a deck and left to interpretation.
Removes: the engagement with no natural end
04
The output is yours
Everything is handed over in source and is yours to act on however you choose, including with another firm or entirely in-house. Continuing with us is a decision you make afterwards.
Removes: deliverables only we can execute
What this work has moved.
Each engagement below was delivered in the practice behind one of the services above, and evidences that service directly. The figures are the ones our clients reported.
Healthcare · Business intelligence
A multi-site care provider
Scattered HR spreadsheets replaced by an integrated analytics layer over their core people system, so leaders work from live talent data instead of weeks-old reports.
Evidence for the Data Platform Accelerator
Read the storyManufacturing · AI and machine learning
A global manufacturer
One function, one use case: a sales forecasting and demand planning model that learns continuously, feeds straight into the operations workflow, and lifted inventory efficiency alongside the forecast.
Evidence for the GenAI Pilot
Read the storyHealthcare · AI and data literacy
A regional health organization
The dashboards already existed and nobody used them. The problem was confidence, not tooling — so we ran a structured literacy program across departments.
Evidence for the Executive AI Workshop
Read the storyDr. Mechie Nkengla
Founder & Chief AI & Data Strategist · Ph.D. Applied Mathematics
The thesis behind the Assessment Sprint — that AI programs stall on data readiness, governance, adoption and unit economics rather than on the technology — was the subject of our IBM TechXchange session on scaling past the pilot.
All case studiesStart from the decision, not the service.
The most common mistake is buying the engagement that matches the technology you are excited about rather than the decision you are stuck on. Find your sentence below.
“I need a number to take to a budget conversation.”
Assessment Sprint. It exists for exactly this. Two weeks, and you leave with sized options and the reasoning attached.
“Leadership wants to see AI doing something real.”
GenAI Pilot. One use case, working, under production conditions — rather than a demo that quietly runs on a laptop.
“Legal or risk has stopped us and we do not know what would satisfy them.”
AI Governance Quickstart. The review path and the record, built so the people who blocked you can sign it.
“We are using more AI and data than our security process was designed to review.”
Cyber Risk Sprint. Your posture scored against the exposure AI actually introduces, not the exposure your last review was written for.
“My leadership team does not agree on what AI is even for.”
Executive AI Workshop. Alignment first. Building anything while the people funding it disagree about its purpose is how pilots get cancelled at the readout.
“Our foundation cannot carry what we want to build on it.”
Data Platform Accelerator. Worth doing the Assessment Sprint first if the scope is not already settled.
“I own a P&L and one number I need AI to move.”
GenAI Pilot. A line-of-business problem with a budget behind it rarely needs a data strategy first — it needs one use case built well, in your function, against that number.
“We are starting from almost nothing and do not know what to ask for.”
Assessment Sprint, deliberately scoped small. Starting from a standing start is a normal place to begin and a cheap one to get right; it is far more expensive to discover the gaps after you have bought a platform.
“We already know what to build and just need the hands.”
None of these. That is Staff Augmentation, and buying a sprint first would waste your time.
The questions we get every time.
What do these cost?
We do not publish numbers, because the honest answer moves with how many systems and teams are in scope and a figure on a web page would be wrong for most readers. What we commit to is the shape: a fixed fee agreed in writing before any work starts, no hourly billing, and no change orders for work inside the agreed scope. You will have the number before you commit to anything.
How do you fix a fee before you know the scope?
There is a scoping conversation first, and it costs you nothing. We work out how many systems, how many teams and how much existing material there is, then write the scope and the fee into the proposal together. If that conversation shows the engagement is larger than a starter service, we say so rather than fixing a fee we would have to renegotiate in week two.
What do you need from us before kickoff?
Access and some people’s time, and the specifics depend entirely on your setup. Whatever it is gets written into the proposal before you commit, so nobody discovers a hidden burden halfway through. If the access you can realistically give us would not support the engagement, that is worth finding out before the contract rather than after.
Do we have to continue with you afterwards?
No, and every one of these is built on that assumption. The output is handed over in source and is yours to act on however you choose, including with another firm or entirely in-house. A deliverable you can only execute by rehiring the people who wrote it is not a deliverable, it is a lead magnet.
What if none of these fit?
Then we will say so. Some problems need an implementation programme rather than a bounded first engagement, and some need nothing from us at all. Both are better answers than selling you a sprint that was never going to resolve the thing you are stuck on. Tell us what you are trying to decide and we will be direct about it.
Can we run more than one?
Usually sequentially rather than at once, and usually you should not need to. The Assessment Sprint frequently changes which of the others makes sense, so running it first and deciding afterwards tends to be cheaper than committing to two upfront.
Tell us what you are trying to decide.
Not what you want to buy. Describe the decision you are stuck on and we will tell you which of these answers it, whether it needs one at all, and what it would take to scope.
Prefer to look first? Score yourself free in five minutes and bring us the result.
Not sure where to start? Tell us the decision, not the service.