How we work
Fixed scope. Senior people. Working output in weeks.
Data projects have a reputation for running long, costing more than quoted and delivering something nobody uses. Most of that comes down to how the engagement is structured. Here is exactly how ours run.
01 The sequence
Four stages, and you can stop after any of them.
Discovery call
Thirty minutes. What your business does, which systems hold the data, where the reporting hurts and what you've already tried. We'll tell you honestly whether we're the right fit — including when the answer is no.
Cost: free · Output: a straight answer
Data health check
A fixed-scope, fixed-price review of your sources, models and reporting as they stand. We map what exists, find where the manual effort and the mistrust come from, and produce a prioritized, costed roadmap.
Typically 1–2 weeks · Output: findings + roadmap
Build in increments
We work through the roadmap in short cycles, starting with whatever removes the most manual effort soonest. You see something working every week rather than waiting months for a reveal.
Quoted per stage · Output: working software
Handover & enablement
Documentation, training and optional AI enablement, so your team can operate and extend what we built. Ongoing support is available, but designed to be optional rather than assumed.
Output: a platform your team owns
02 The starting point
The Data Health Check, in detail
Nearly every relationship here begins the same way. Not with a proposal, but with someone senior looking properly at what you've got and telling you the truth about it.
Fixed price, agreed before we begin. You keep the output whether or not you go on to work with us — including if the recommendation is that you don't need to.
What we look at
- Every system your reporting currently depends on, including the spreadsheets that have quietly become systems
- How data moves between them today, and how much of that movement is a person
- Where your numbers diverge between reports, and the specific reason each divergence happens
- Your existing Power BI or reporting estate — what's used, what isn't, what's slow and what's wrong
- Which metrics have an agreed definition and which have three
- Where the single points of failure are, in systems and in people
- What your team spends time on that shouldn't require a human at all
What you get
- A clear map of your current data landscape — often the first time anyone has drawn one
- A prioritized list of problems, ordered by cost to the business rather than by technical interest
- A recommended architecture, sized honestly for your business and budget
- A staged roadmap with a cost against each stage, so you can start small and stop whenever you like
- A direct recommendation, including which things aren't worth fixing
What it isn't
It isn't a sales document with a roadmap attached. Several health checks have concluded that the client needed one pipeline fixed and some training, rather than a platform — and that's what we told them. A recommendation you can't trust isn't worth paying for.
03 Working together
What we expect from each other
You get senior people, throughout
The person on your discovery call is the person doing the work. There's no bench, no account layer, and no point at which delivery quietly transfers to someone more junior. This is the main structural advantage of a small firm and we're not going to give it up.
You'll know the cost before we start
Each stage is scoped and priced up front. If we discover something that changes the shape of the work — and occasionally we do — that's a conversation before any of it happens, not a variation on an invoice afterwards.
We need someone who knows the business
The biggest risk to a data project isn't technical, it's definitional. We need reasonable access to someone who can settle what a term means when three departments disagree. A few hours a week, but genuinely needed.
We build so you can leave
Everything source-controlled, documented and handed over. We'd rather be the firm you call for the next problem than the one you can't afford to stop paying — retained support should be a choice, not a consequence of how we built it.
We'll tell you when the answer is no
If you don't need Fabric, we'll say so. If your problem is process rather than data, we'll say that too. If you'd be better served by hiring someone permanent than by engaging us, that's a conversation we're willing to have — it costs us one engagement and earns a reference.
Next step
Start with the conversation, not the proposal.
Thirty minutes on your data and where it's costing you. You'll leave with at least one useful observation whether or not we work together.