Power BI consultant UK SMEs use instead of hiring an analyst

What does a Power BI consultant in the UK build for a small business?

A Power BI consultant builds the pipeline from your business systems to a set of reports that update on schedule: connections to sources, a clean data model, calculated measures such as gross margin and debtor days, report pages for each audience, and the refresh and security settings that keep it running.

For a UK SME, the real product is not the charts. It is the end of the monthly ritual where someone exports from Xero or Sage, pastes into a workbook, fixes the formulas that broke, and emails a PDF nobody fully trusts. A good Power BI consultant UK businesses hire removes that ritual and leaves you with numbers that match the accounts.

The work usually has five stages: agree the questions, connect and clean the sources, build the model, design the pages, and set up refresh plus handover. Each is visible to you. You see reconciled totals before any chart is drawn, because a beautiful dashboard with the wrong turnover figure is worse than no dashboard.

We are three freelance developers working from India. Another of us leads data modelling, SQL and cloud setup, the third of us runs discovery and KPI design with your directors, and one of us builds any custom connectors or web dashboards. You talk to the people doing the work.

Power BI or Looker Studio: which should a UK SME choose?

Choose Power BI if your business runs on Microsoft 365, needs a proper data model joining several sources, or wants row-level security. Choose Looker Studio if you run on Google Workspace, mostly report on Google Ads, GA4 and Sheets, and want viewers to have no per-user licence cost.

Power BI is stronger at modelling: relationships between sales, customers, products and dates, time-intelligence calculations like year-to-date and same period last year, and security roles. Its licensing is per user for sharing, which Microsoft's documentation explains in detail. Looker Studio is simpler, browser-based and free for standard use, and it shines for marketing reporting, but complex joins across finance and sales data are harder and slower in it.

Plenty of UK SMEs use both: Looker Studio for the marketing team's channel reports and Power BI for the finance and operations pack. As a Power BI consultant UK clients hire for honest advice, we will tell you when Looker Studio is enough, because that saves you licences and build time.

What data sources do UK SMEs typically connect to Power BI?

Most UK SMEs we speak to have an accounts package, a sales channel, a CRM or job tool, and several spreadsheets that fill the gaps. The accounts package is usually Xero, Sage or QuickBooks; sales come from Shopify, WooCommerce, Amazon or an EPOS system.

Each source behaves differently. Cloud tools with APIs, such as Xero, Shopify and HubSpot, can be pulled automatically. Desktop software installed on an office PC or server needs a gateway so Power BI can reach it. Spreadsheets are the trickiest, because people change column names, add tabs and type “n/a” into number columns; we move them to a structured table or SharePoint list with validation before connecting.

• Accounts: Xero, Sage 50, Sage Accounting, Sage 200, QuickBooks Online, FreeAgent.
• Sales channels: Shopify, WooCommerce, Amazon Seller Central, eBay, EPOS exports.
• Customers and pipeline: HubSpot, Pipedrive, Zoho CRM, Salesforce.
• Operations: job management tools, stock systems, timesheets.
• Marketing: GA4, Google Ads, Meta Ads, Search Console.
• Everything else: Excel workbooks and Google Sheets kept by individuals.

Shopify merchants who want the shop side improved before measuring it should see our Shopify developer page for UK brands.

How do you get Sage data into Power BI?

It depends which Sage you run. Cloud Sage products expose data through APIs or exports that can be scheduled; Sage 50 installed on a local PC or server needs a connection on that machine plus an on-premises data gateway so Power BI can refresh it.

Microsoft describes the on-premises data gateway as locally installed Windows software that connects the Power BI service to data sources on your network, using outbound connections so you do not need to open inbound ports. That matters for a small office: no firewall changes, but the machine hosting the gateway must be on and connected when refreshes run. A PC that is switched off at 6pm means a 7am refresh fails.

For Sage 50 desktop, data is usually read on the machine where the company data lives, through the database connection Sage provides or through scheduled exports, then refreshed via the gateway. For Sage 200, the reporting database route depends on your edition and hosting. We check your setup in discovery rather than assuming one method.

Microsoft's gateway documentation distinguishes standard mode, which supports shared connections and multiple users, from personal mode for one user's own models. For a business, we install standard mode on an always-on machine, register it to your tenant, and document who administers it.

Xero, Shopify and other cloud sources: API pulls and a small data store

Cloud sources are pulled through their APIs, either by a Power BI connector where a reliable one exists, or by a small scheduled script that lands the data in a database Power BI reads. The second route is often more dependable for SMEs.

Xero's API gives access to invoices, bills, contacts, bank transactions and reports such as profit and loss. Shopify's API gives orders, line items, refunds, products and customers. Pulling these directly into Power BI every refresh can be slow and fragile, especially with years of history, so we often store them in a small cloud database (Azure SQL, PostgreSQL or BigQuery), pull only new and changed records, and point Power BI at that.

This design also solves joins. Shopify knows the order; Xero knows the invoice, payment and VAT; your stock spreadsheet knows the landed cost. In the data store, they meet on order numbers and SKUs, so you can see true margin by product and channel instead of three partial views.

• Incremental loads, so each refresh fetches only what changed.
• History kept even if a source limits how far back its API goes.
• Reconciliation checks comparing totals with the source reports.
• Logging, so a failed pull is spotted before the board meeting.

If you also want Xero talking to your other systems, not just reporting, see our Xero integration page.

Designing KPIs directors will actually read

Start from decisions, not data. Ask each director which three numbers would change what they do next month, then design one page around those. A dashboard with forty visuals gets opened twice and ignored.

For most UK SMEs, the director page covers cash position and forecast, revenue against budget and last year, gross margin by channel or product group, aged debtors and debtor days, and one operational measure such as stock cover, utilisation or order fulfilment time. Each figure sits beside its comparison and a simple indicator of direction, with a drill-through to detail for anyone who wants it.

Definitions are written down. “Revenue” might mean invoiced sales excluding VAT, or orders placed including shipping, and those give different answers. We agree every definition with your finance lead, document it on a glossary page inside the report, and build the measure once so every page uses the same logic.

Setting up Power BI for UK year ends, VAT quarters and the tax year

A UK dashboard needs a date table that understands your company's financial year, your VAT periods and, where relevant, the personal tax year that runs from 6 April to 5 April. Without it, “year to date” means the calendar year and finance ignores the report.

UK companies choose their own accounting reference date, so we never assume March or December. The date table gets columns for financial year, financial quarter and period number, plus flags for your VAT quarter pattern, whether that is January, February or March stagger. Measures such as year-to-date revenue and same period last year then follow your year automatically.

Businesses under Making Tax Digital for VAT already keep digital records in software like Xero or Sage, which makes the finance side of the model easier: the VAT-period view in Power BI can be reconciled to the returns your accountant files. The dashboard is for management insight, not for filing; your accountant and the accounting software remain the source for returns.

Retailers may also want trading weeks or a 4-4-5 calendar. That is just more columns in the same date table, agreed in discovery.

How often can Power BI dashboards refresh?

On shared capacity, which most SMEs on Power BI Pro use, Microsoft allows up to eight scheduled refreshes per semantic model per day. On Premium, Premium Per User or Fabric capacity, you can schedule up to 48.

Microsoft's data refresh documentation notes that the eight-slot limit on shared capacity applies to scheduled refreshes and API-triggered ones, and that the quota resets daily. For most management reporting, eight is plenty: a refresh at 6am, noon and 5pm keeps directors current. Sales-floor screens that need near real-time figures are a different design problem.

There is a trap worth knowing. Microsoft states that if nobody visits any dashboard or report built on a semantic model for two months, Power BI treats it as inactive and pauses the refresh schedule, emailing the owner. Seasonal reports, like a year-end pack, can quietly stop updating. We set the owner to a monitored mailbox and add refresh-failure alerts to a named person.

• Schedule refreshes after source systems finish overnight jobs.
• Keep the gateway machine on during refresh windows.
• Send failure alerts to someone who will act, not a departed employee.
• Use incremental refresh for large histories to keep refreshes fast.

Do I need Power BI Pro? Licensing for a Microsoft 365 business

If you want to share Power BI reports with colleagues in the standard way, the person publishing and the people viewing generally need Power BI Pro licences. Free licences let people build reports for their own use only, unless the content sits on Premium or Fabric F64 or larger capacity.

Microsoft's licensing documentation sets out three per-user types: Free, Pro and Premium Per User. Pro users can publish, share and collaborate with other Pro users. PPU adds most Premium features, but sharing PPU workspace content requires viewers to have PPU too. Capacity licences, meaning Fabric F64 and above or Power BI Premium, allow Pro or PPU users to share content with colleagues who have free licences.

For a small team of directors and managers, Pro for each viewer is usually the simplest answer. For a business wanting reports seen by many staff, the arithmetic changes, and either capacity, Looker Studio or a custom web dashboard may work out cheaper. Check first whether your existing Microsoft 365 subscription already includes a Power BI licence; Microsoft's documentation explains how.

We give you a licence recommendation with the quote, based on who needs to view, edit and publish. Licences are bought and billed in your own Microsoft account.

How much does a Power BI consultant cost in the UK?

Quotes vary widely, mostly because of how many sources are involved and how clean they are. With BtechWaleTech, dashboard and data builds start from US$600, a custom web dashboard starts from US$900, and ongoing care starts from US$120/mo after two free months.

When comparing a Power BI consultant UK quote with ours, compare deliverables rather than day rates. Ask how many sources, how many report pages, whether reconciliation to your accounts is included, whether gateway setup and refresh monitoring are included, what documentation you get and how much training. A cheap quote that stops at “connected to Xero” leaves you to discover that VAT, credit notes and multi-currency invoices are not handled.

The biggest cost driver is data quality. If product codes differ between Shopify and your stock spreadsheet, or customers appear under three spellings in Sage, somebody has to map them. We price that honestly in the quote after a quick look at your data, rather than discovering it halfway through.

Licences sit outside our quote. Power BI Pro or capacity is billed by Microsoft, Looker Studio has no licence cost for standard use, and any cloud database for API pulls is billed to your own Azure, AWS or Google account.

One-off Power BI build or monthly support: which suits a UK SME?

Choose a one-off build when your questions are stable and your sources rarely change. Add monthly support when you expect new reports, new sources or staff who will lean on someone for changes and explanations.

A one-off build ends with a documented model, trained users and two months of free maintenance to catch problems like a changed column name or an expired connection. Many SMEs never need more. After that, maintenance from US$120/mo covers refresh monitoring, fixes when a source changes its API, and small tweaks.

Monthly support suits businesses whose needs keep moving: a new branch, a second Shopify store, a lender asking for a covenant report, a director wanting a new cut of the numbers each quarter. It is agreed in your written quote, with the scope stated clearly, so you know what is included and what is quoted separately.

Giving a remote Power BI consultant access to your data safely

Give the minimum access needed, in accounts you control, and remove it when the work ends. Your data should stay in your Microsoft or Google tenant and your source systems, not be copied onto a consultant's laptop.

In practice we ask for a guest or named user in your Microsoft 365 tenant with rights to a specific workspace, read-only or API access to accounting and sales tools, and, where a gateway is needed, a short screen-share session so your IT contact installs it with us. Credentials are stored in the Power BI service or your cloud key vault, not in files.

Under UK GDPR, you remain the controller of any personal data, such as customer names in sales data. Where dashboards do not need personal data, we leave it out or aggregate it. Row-level security limits what each viewer sees. You should record the arrangement in your own processing records and check any international transfer safeguards with your adviser, since we work from India.

• Named accounts only; no shared passwords.
• Read-only access to source systems wherever possible.
• Personal data excluded unless a report genuinely needs it.
• Access reviewed at handover and removed on request.

How long does a Power BI dashboard project take?

Most SME builds take two to four weeks from approval: about a week for discovery and source connection, a week for the model and reconciliation, and one or two weeks for report pages, refresh setup, testing and training.

The first week is where time is won or lost. If access to Xero, Sage or Shopify arrives on day one, we can pull sample data immediately and spot quality problems early. If the gateway machine needs IT approval, or the key spreadsheet belongs to someone on holiday, the start slips. We send a one-page access checklist with the quote so you can prepare.

Reconciliation is a formal step. Before we design visuals, you receive a simple table of monthly totals from the model beside the same totals from your accounting reports. Your finance lead signs it off. Only then do pages get built, which keeps the review focused on layout rather than disputes about numbers.

Larger projects, such as several companies consolidated, multi-currency group reporting or a custom web dashboard, take longer, and the quote says so.

Why SME dashboards fail, and how to avoid it

Most failed dashboards fail for human reasons, not technical ones: nobody agreed the definitions, the numbers did not match the accounts, or refresh broke quietly and trust evaporated. Each of these is preventable.

• Numbers that do not reconcile: fixed by a signed-off reconciliation table before any design work.
• Too many visuals: fixed by designing each page around a few decisions.
• Silent refresh failures: fixed by alerts to a named person and a visible “last refreshed” time on every page.
• Gateway on a laptop: fixed by installing it on an always-on machine in standard mode.
• Spreadsheet chaos: fixed by moving manual inputs into validated tables.
• One person understands it: fixed by documentation and training for at least two staff.
• Licence surprises: fixed by a licence plan agreed before building.

Red flags when choosing a consultant include refusing to show reconciled totals, building in their own tenant rather than yours, and promising real-time dashboards from desktop software without discussing the gateway. Want automation that feeds data into systems as well as reports? Our AI automation guide for UK SMEs covers that side.

Working with a Power BI consultant in India from the UK

India is four and a half hours ahead of the UK in summer and five and a half in winter. Your late morning and afternoon sit inside our working day, and overnight UK time is a handy slot for heavy refresh testing without disturbing anyone.

Discovery happens on a video call with your directors and finance lead, usually an hour, screen-sharing your current spreadsheets. After that, we share progress as short recordings and a test workspace you can open any time. Questions go on WhatsApp, which we answer seven days a week.

Quotes and invoices are in USD, paid from the UK via Wise, bank wire or PayPal. The itemised quote arrives in about two working days and nothing is billed before written approval. Invoices come from India; your accountant can advise on VAT treatment of overseas services.

A worked example: a hypothetical Midlands parts supplier on Sage 50 and Shopify

Suppose an engineering parts supplier near Coventry runs Sage 50 on an office server, sells to trade customers by phone and to the public through a Shopify store, and tracks landed costs in a spreadsheet. The managing director wants one monthly view of margin by channel. This is an illustration, not a client story.

Discovery would produce a short KPI list: revenue and gross margin by channel, top customers by margin, aged debtors, and stock value by product group. The finance manager would confirm that revenue means invoiced sales excluding VAT and that the financial year ends 30 June.

The build would install a standard-mode gateway on the always-on office server to reach Sage 50, pull Shopify orders nightly into a small cloud database via its API, and move the landed-cost spreadsheet into a validated SharePoint list. In the model, Shopify orders and Sage invoices meet on SKU, with landed cost attached, giving true margin per channel.

Reconciled monthly totals would go to the finance manager before any page is designed. The final report would have a director page, a finance page with VAT-quarter views and a sales page, refreshed three times a day, with alerts to the finance manager. Given one desktop source, one cloud source and one spreadsheet, the quote would likely sit a little above the US$600 starting price.

Power BI consultant UK checklist before you start

Prepare these and your project will move faster, whoever you hire. Most of them are decisions only your business can make.

• A list of the five to ten questions your dashboard must answer.
• Named owner for each KPI definition, usually the finance lead.
• Your financial year end and VAT quarter pattern.
• Access to accounting, sales and CRM systems, ideally read-only.
• The spreadsheets people rely on, with the person who maintains each.
• An always-on machine for the gateway if any source is desktop software.
• A decision on who needs to view, edit and publish, for licensing.
• Two staff members who will be trained to maintain the reports.
• Agreement that reconciled totals are signed off before design.
• A plan for support after the free maintenance period ends.

Ready? Send us your sources and the questions on WhatsApp, or read our bespoke software guide if you suspect the real problem is the systems, not the reporting.