Frequently asked questions.

Duncan Boyne is a UK Power BI consultant who helps businesses turn data scattered across ERPs, CRMs, SQL databases and spreadsheets into audit-ready, decision-ready reporting that leaders trust. The work combines proper data modelling underneath with design thinking on top: visual hierarchy, colour theory and accessibility.

A Power BI Health Check is a flat £950, fixed: a scored report card across six areas, eight checks in each, the issues that matter ranked by impact, a practical roadmap and a walkthrough call, all within five working days.

Yes. Sage 200 is a specialism: connecting it to Power BI, modelling it properly, and building financial reporting where the accounts reconcile back to Sage to the penny. There is a live example built on an anonymised Sage 200 estate in the portfolio.

Duncan is based in Great Yarmouth, Norfolk, and works with businesses across Norwich, East Anglia and the wider UK, remotely and on-site. Typical clients run on Sage 200, other ERPs, CRMs, SQL databases and spreadsheets across manufacturing, wholesale and distribution, professional services and construction.

The difference is trust and usability. Numbers are traceable from the visual back to source and reconcile to the penny, measures are consistent everywhere they appear, and reports are designed with a clear visual hierarchy so people can see what changed and what is at risk, not just a wall of numbers nobody opens. Accessibility is part of the build too: contrast that passes, alt text on every visual, and colour never carrying a meaning on its own.

That's the point of it. Copilot answers questions using whatever your semantic model tells it, so if measures disagree or numbers don't reconcile, AI repeats those errors faster and with more confidence. Reconciled data, consistent measures and documented logic are the groundwork that makes Copilot safe to switch on, and they're exactly what these engagements build.

Yes, as a rolling monthly care plan in three tiers, cancel any time. Watch is £250 a month: refresh failure monitoring and alerting, a scored monthly health report, and any failure diagnosed within 24 hours, with no build hours included. Care is £750 a month and adds hands-on work: one request at a time, fixes back within 48 hours, up to four hours of build work and a quarterly model review. Partner is £1,500 a month for two requests in flight, up to ten hours and ownership of the reporting roadmap.

Most people start with a fixed-price Power BI Health Check or a free 20-minute intro call. You can pay online or book a call directly from this site.

A Power BI Health Check is £950, a flat fixed fee with no day rates. That covers six areas of your Power BI estate with eight scored checks in each, 48 checks in total, each returned as Healthy, Watch or Critical against a written standard. You get a scored report card within five working days, the issues ranked by impact, a roadmap you can hand to a developer, and a walkthrough call. It covers one workspace, up to 15 reports and up to 3 semantic models. Payable by card online or by invoice, and the fee credits in full against work booked afterwards.

Remote by default, which keeps it simple and fast. On-site works across Norfolk and East Anglia, and further afield by arrangement.

Read access to the reports and model, or a copy of the file. Nothing gets changed without agreeing it first.

The scored report card and roadmap land within five working days of the session, followed by a walkthrough call.

Anything Power BI touches: Sage 200, Dynamics 365, Epicor, Infor Visual, SQL databases, CRMs and spreadsheets.

It's a fixed-price £2,950 assessment of whether your semantic models will give Copilot correct answers. Copilot does not fix a model, it inherits one: ambiguous field names, missing descriptions, no synonyms and measures nobody trusts all turn into confident wrong answers, faster than a human could produce them. The assessment scores what is there, lists what to change, and tells you what the capacity will actually cost.

Thirty real business questions, written with your team, each with a verified expected answer worked out from source. Those get run against your model before any changes and again afterwards, so AI readiness stops being a matter of opinion and becomes a pass rate you can see move. It is also re-runnable, which makes it the natural thing to re-test after any significant model change.

No, and finding out what you would need is part of the point. Copilot in Power BI needs an F2 capacity as a floor, but sustained real use pushes much higher, and the gap between those two numbers is where budgets get embarrassed. You get capacity and licence cost modelling as a named deliverable, so the finance conversation happens before the bill does.

No. Plenty of Copilot workshops are a day of slides and a demo on sample data. The deliverables here are specific to your estate: your models scored, your tenant settings reviewed, your questions tested, your capacity costed, and a written list of who should not be given Copilot access yet and why.

Then that is the finding, and it is worth more than a green light would be. Switching Copilot on over a model that disagrees with itself does not create a reporting problem, it broadcasts one. The report tells you what to fix first and what it would take, and the fee credits against that remediation work if you book it within 90 days.

A fixed-price £5,950 assessment, five days of work delivered over about three weeks, that answers whether you should move to Microsoft Fabric, what it would cost over three years, and which parts of your estate should not move at all. You get a maturity benchmark, a target architecture with the reasoning written down, capacity sizing, a 90-day plan and an 18-month roadmap.

Yes, and sometimes it should be. Plenty of estates are being pushed toward Fabric on momentum rather than a business case. The assessment is equally useful as evidence for slowing a migration down or scoping it much smaller, and the three-year cost model is usually the part that changes the conversation.

Because Fabric capacity is the cost that surprises people. The difference between the smallest workable SKU and what sustained real usage needs is very large, and it is easy to pilot happily on something that will not carry production. Sizing it with your actual refresh patterns and user counts is more useful than any feature comparison.

It helps, and I am one. It means the assessment lines up with how Microsoft expects these engagements to be scoped, and it removes the procurement objection that comes with hiring an individual. It does not mean I sell you licences: I sell no Microsoft seats or software, so nothing in the recommendation is shaped by resale.

You get a decision you can defend to a board, in writing, with the numbers behind it. If you then want the migration delivered, that is scoped separately and half of this fee comes off it if you book within 90 days.

It's the core engagement: turning scattered systems and spreadsheets into one properly modelled source of truth, with trusted measures and reports designed to be used. It's scoped per project rather than sold off a rate card.

Builds are scoped to your systems and the decisions you need to support, so pricing follows the scope. It starts with a discovery conversation, not a rate card.

ERPs including Sage 200, CRMs, SQL databases and spreadsheets. If Power BI can reach it, it can go into one modelled layer.

Yes. You get documentation that survives me leaving, training so your team owns it, and support when you need it. No lock-in.

Start with the free one-day assessment: every report gets catalogued and complexity banded, and you get a fixed total in writing before committing to anything. Migration is then £600 per report with no minimum, so a single report is a perfectly good place to start. The shared semantic model gets built before the reports, so each report sits on the same definitions rather than its own copy of the logic, and every migrated figure is validated side by side against the original before sign-off. You also get a retirement plan for the old platform with the licence saving quantified.

£600 per report, with no minimum order — migrate one report to see how it goes, or the whole estate at once. A free one-day assessment produces the inventory and a fixed total before you commit to anything, so you know the number in advance rather than watching a day rate accumulate. Around £500 to £1,000 per dashboard is the accepted market rate, and publishing it means you can work out your own budget before you ever email me.

Tableau, Qlik and spreadsheet estates most often, and legacy SSRS or Excel-based reporting suites regularly. The work is the same shape in each case: understand what the report is actually for, rebuild it properly in Power BI, and validate the numbers side by side.

Only if that is what you want. Like-for-like is available and sometimes necessary for regulatory or audit reasons. Where the original was poor, a redesign costs the same per report, because rebuilding a bad report faithfully is not a service worth selling. That gets agreed report by report during the assessment.

That is usually where the value is. Ten Tableau workbooks tend to contain ten slightly different definitions of the same measure, which is why nobody trusts the numbers. Migration is the moment to collapse those into one shared semantic model, and the assessment flags where the definitions currently disagree.

Every migrated figure is validated side by side against the original before sign-off. If a number differs, either the old report was wrong or the new one is, and you get told which. That check is part of the price rather than an extra.

When you stop using an old system, the obligation to produce its data does not stop with it. Auditors, HMRC and your own finance team still need the history, so the system limps on for years and keeps costing money. This turns it into a queryable reporting archive with an audit pack ready to go, so you can switch the old system off and stop paying for it. From £1,950 for a single system, from £6,500 for a finance-grade archive with full reconciliation.

Because they are genuinely different jobs. Archiving a CRM or an operational system with a couple of years of history is a few days of work: model it, make it queryable, document it, check the record counts and totals tie out. A finance system is a bigger job because the reconciliation has to stand up to an auditor: trial balance, P&L, balance sheet, drill down to individual transactions, tied back at £0 variance. Paying finance-grade money to archive a small CRM would be daft, and I would rather say so on the page than in a quote.

Anything whose database can be attached or exported. Sage, Dynamics, Epicor, Infor, bespoke SQL systems and most CRMs have all been done. The work is driven by the database rather than the badge on the front, so an unusual or heavily customised system is not a problem, it just changes the scoping conversation.

On the finance-grade version, a trial balance reconciles back to the source system and if it does not tie out at £0 variance it does not ship. On the single-system version, record counts and key totals are reconciled and the reconciliation is included in the pack. Either way you get the evidence rather than an assurance, because an approximate archive invites a decision that cannot be defended later.

No. The data is attached read-only or worked from a backup, so nothing is written back and nothing live gets slowed down. If the system is already switched off, a backup file is all that is needed.

No. The archive is delivered into your own Power BI tenant and the data stays with you, which is where it belongs. If you also need a restorable copy of the original database retained somewhere for a regulator or an insurer, that is a job for your IT provider or a specialist archiving vendor rather than something I take on.

Long enough that deleting it is rarely an option, which is exactly why so many dead systems are still running. Company and tax records have statutory retention periods measured in years, and an auditor, an HMRC enquiry or a due diligence process can ask for them at any point inside that window. The question is not whether to keep the data, it is whether you keep paying a licence and a server bill for the privilege.

Any level. That ranges from a "What is Power BI and why would we use it?" first look for teams brand new to it, through report-building foundations, to advanced data modelling, DAX, Power Query, performance tuning and report design grounded in visual hierarchy, colour theory and accessibility.

Because generic training on demo files rarely sticks. The prep half-day clears the practical blockers before anyone sits down to learn: Power BI installed, the right licences in place, and access to your data agreed. It also lets me mould the material to your systems and the questions your business asks, so your team learns on the data they'll be using the next morning.

Yes. Regular sessions for a developer or a team work well as a retainer. Email me and we'll shape it around what you need.

Yes, for your internal use. You keep the recording and the written summary.

An unblocking session is £100. Bring a specific problem, we work it live over a screen share, and you leave unblocked with a written summary of what changed and why.

A 1-2-1 day is from £750 online or £850 in person; a small group (up to 4) from £950 or £1,100; a team day (up to 10) from £1,250 or £1,500. Rates depend on level and prep, and are confirmed before anything is booked.

There are three rolling monthly tiers. Watch is £250: monitoring, alerting, a scored monthly health report and any failure diagnosed within 24 hours, with no build hours included. Care is £750: everything in Watch plus one request at a time, fixes back within 48 hours, up to four hours of build work and a quarterly model review. Partner is £1,500: two requests in flight, up to ten hours, a monthly working session and I own the reporting roadmap. All three are rolling, cancel any time with an email.

Because selling four hours of hands-on time for £250 priced my time at about £62 an hour, and that is not sustainable for anyone: a plan that loses money gets resented, then rationed, then quietly dropped. Watch now buys the thing that genuinely prevents most incidents, which is monitoring and a monthly scored report, and it is honest about not including build time. If you want hands-on capacity, Care at £750 is the tier that has it.

Because nobody knows how many hours a change takes until it is done, so an hour count makes both of us watch a clock instead of the work. One request at a time means you queue as much as you like, I work them in order, and nothing gets rationed at the end of the month because a timesheet says so. It also caps my workload honestly rather than through goodwill.

Refresh failures are diagnosed within 24 hours of your report on every tier, usually much sooner, and on Watch you often hear from me before you have noticed. You get a plain-English note covering what broke, why it broke and what fixes it.

Your estate scored against the same standards as a Health Check, tracked month on month so a slow decline shows up as a trend rather than as a crisis. It is the reason the cheapest tier is still worth paying for in a month where nothing broke and you raised nothing.

No, on any tier. The plan buys availability, monitoring and fast answers rather than banked hours, and hours that roll over turn a care plan into a savings account with a consultant attached. Anything bigger than the tier covers gets scoped and quoted before it starts.

Yes. Most estates I support were built by someone who has since moved on. A short handover review gets me familiar with your models and reports; from then on they're covered like anything I built myself.

Considerably, and that is the honest comparison rather than a comparison against another consultant's day rate. A BI analyst in the UK costs upwards of £45,000 plus employer costs, holiday and the risk that they leave and take the knowledge with them. Partner at £1,500 a month is well under half of that, and the documentation stays with you either way.