Skip to main content
Back to Read
Business Dashboards5 June 2026Updated 23 September 202615 min read

Business Dashboard Development UK: KPI Dashboards That Actually Get Used (2026)

A practical guide to business dashboard development for UK SMEs: what to track, which views to build, when Power BI is enough, and when a custom command centre makes sense.

Sajad Saleem

Co-founder of Ampliflow. Builds AI automation, websites, SEO/AEO, and growth systems for UK SMEs.

Hand-cut paper collage of two paper figures beside a wall of coloured paper bar charts and a paper dial in a simple office.
Editorial illustration.
  1. 01What is a business dashboard?
  2. 02The four dashboard views most UK SMEs need first
  3. 03Power BI dashboard vs custom business dashboard
  4. 04What a business dashboard costs in the UK
  5. 05Dashboard audit checklist

A business dashboard earns its place when it changes what someone does next.

The owner wants one version of the truth. Sales wants pipeline. Operations wants workload. Marketing wants attribution. Finance wants forecast confidence. The dashboard tries to serve everyone at once, becomes a wall of charts, and the team drifts back to spreadsheets.

Quick answer: Business dashboard development is the design and build of KPI, sales, operations and revenue views that help a team make decisions faster. For most UK SMEs, the first dashboard should show the few metrics that change action this week: lead flow, follow-up risk, operational workload, revenue movement and the records that need an owner.

Last updated: 23 September 2026 · UK pricing in pounds · Worked example uses synthetic data, not client results

A KPI, or key performance indicator, is a measure tied to a business decision. A CRM is the system holding customer records and sales activity. Agree which records count before connecting them to a dashboard.

TL;DR:

  • A business dashboard should be designed around decisions, not every metric you can technically connect.
  • The first four views are usually enough: leadership KPI overview, sales pipeline, operations workload, and revenue attribution.
  • Power BI is often enough for internal reporting. A custom dashboard makes sense when the dashboard needs to work as a product, workflow surface or branded portal.
  • Before building, define each KPI's denominator, freshness, owner and decision. Otherwise the dashboard will make confusion look official.

Buying a KPI dashboard? Start with one decision

If you are looking for a developer, bring the recurring question your team cannot answer reliably. "Which quotes need follow-up today?" is a stronger brief than "we need a dashboard".

Put this in the briefExample to adapt, not a client result
DecisionWhich open quotes need an owner to follow up?
SourceThe existing quote register and contact records
DefinitionOpen means neither accepted nor declined; agree how expired quotes are handled
AccessEach salesperson sees their records; the manager sees the team
ActionOpen the quote, assign an owner and record the next step
Acceptance checkReconcile a sample against the source; test missing dates, duplicate records and restricted access

Ask suppliers to quote against the same brief. Separate the interface from data preparation, integrations, licences and ongoing support. A cheaper screen can still leave your team doing the weekly reconciliation by hand.

For Ampliflow's dashboard service, see business dashboard development. To discuss the decision and scope, Get unstuck. Use an anonymised sample; you do not need to send customer records to explain the problem.

What is a business dashboard?

A business dashboard is a live view of the numbers a team uses to make decisions. A good dashboard pulls data from tools such as analytics, CRM, ads, finance, support and operations systems, then presents the useful metrics in one place.

The useful version is selective. A chart earns its place only when somebody can say what they will do if the number moves.

The dashboard test

  1. 1

    Who uses it?

    Leadership, sales, operations, marketing, finance, account managers or clients.

  2. 2

    What decision does it support?

    Scale, pause, follow up, hire, fix, forecast, investigate or escalate.

  3. 3

    What data proves the answer?

    Only the metrics needed for that decision, with agreed definitions and denominators.

  4. 4

    What action follows?

    A dashboard should point to the next move, owner and review rhythm.

If a chart cannot answer those four questions, it probably does not belong on the first screen.

Why UK SMEs need simpler dashboards

The UK's business base is mostly small. The Department for Business and Trade's 2025 business population estimates reported 16.9 million people employed by SMEs, 60% of total UK employment, with estimated SME turnover of £2.8 trillion. ONS UK business activity, size and location data also recorded 2.73 million VAT and/or PAYE businesses in March 2025.

That matters because most SME dashboards are not enterprise BI programmes. They are owner, manager and team views for businesses where a few missed enquiries, unassigned jobs or delayed quotes can change the month.

For those businesses, the dashboard should do three things:

JobWhat the dashboard must show
Focus attentionThe small set of numbers that changed and need action
Expose stuck workLeads, quotes, jobs, tickets or invoices with no owner or next step
Create one definitionWhat counts as a lead, qualified opportunity, won job, completed work or overdue item

The value is not the chart. The value is the argument the team no longer has every Monday.

The four dashboard views most UK SMEs need first

  1. 01Leadership — are we on plan?
  2. 02Sales — where is revenue leaking?
  3. 03Operations — what needs attention?
  4. 04Customer — what is changing?

You do not need a large business intelligence programme to get value from dashboards. Most SME teams need four practical views before anything else.

Dashboard viewWho uses itQuestions it answers
KPI overviewDirectors and leadershipAre we on track? Which number changed? What needs attention this week?
Sales pipelineSales, founders, account managersWhere are leads stuck? Who needs follow-up? Which source creates qualified opportunities?
Operations workloadOperations and delivery leadsWhat is overdue? Where is capacity tight? Which work is at risk?
Revenue attributionMarketing, leadership, financeWhich channels create revenue, not just clicks or enquiries?

Start here. These views cover the decisions that usually move revenue, delivery quality and team focus.

For a ready-made service-business structure, use the KPI dashboard template for service businesses. It includes a downloadable CSV with synthetic worked data, so you can see the definitions before connecting live systems.

Worked example: turn a weekly number into today's action

Take the first week in our downloadable dashboard template. It contains 42 enquiries, 25 qualified opportunities and 18 quotes, with nine accepted quotes from that same quote cohort. These are synthetic teaching figures, not an Ampliflow client result or a benchmark for your business.

ViewCalculation or observationWhat the owner should inspect next
Qualification25 ÷ 42 = 59.5%Which sources produced unsuitable enquiries, and why?
Quote acceptance9 ÷ 18 = 50%The nine unaccepted quotes: awaiting a decision, declined or no longer relevant?
ResponseMedian first useful reply: 46 minutesThe individual late replies; a median can hide long waits
DeliveryFive overdue jobs out of 31 open jobsEach blocked job, its owner and the next agreed action

The first screen should let you move from the metric to the relevant records. A director sees the summary; the person responsible sees the work to do. Do not describe the seven qualified enquiries without quotes as lost sales: some may still be progressing normally.

Before connecting a live CRM, agree how to handle duplicates, tests, spam, reopened enquiries and missing timestamps. An automated acknowledgement is not the same as a useful human reply. Quotes sent this week and quotes accepted this week may belong to different cohorts; dividing them can produce a misleading conversion rate.

The template walkthrough explains the definitions and source breakdown. Use it to challenge the logic before commissioning a polished interface.

Ask the dashboard to explain its own numbers

Before commissioning more charts, ask the proposed dashboard to explain one change. In the same synthetic dataset, overdue work rises from 4 of 34 open jobs to 9 of 39. That is a rise from 11.8% to 23.1%, or 11.3 percentage points using the unrounded rates. It is a reason to inspect the nine jobs, not evidence that marketing, staffing or software caused the delay.

The annotated template walkthrough and downloadable decision review show the calculation, the next question and the limits of the inference. They are original teaching resources based on synthetic data, not client results.

A useful demonstration should answer four questions without the developer interpreting every screen:

  1. Which records count? Show the inclusion rule, duplicates and exclusions.
  2. How current is the answer? Show the last successful refresh, not merely the time the screen opened.
  3. What is unknown? Keep missing statuses and unavailable sources visible. An empty feed is not automatically zero activity.
  4. Who can act? Let the authorised owner inspect the underlying work; other users should see only what their role permits.

Agree an acceptance check using known example records, then try a missing timestamp, a stopped feed and a user without access. Keep a record of the expected and actual result. Do not let a convincing demonstration substitute for the permissions and failure checks your own data requires.

KPI dashboard: the leadership view

A KPI dashboard should be brutally selective. It should show the numbers leadership actually checks before making decisions.

Useful KPI dashboards often include:

  • Enquiries by source
  • Qualified opportunities
  • Pipeline value
  • Quote acceptance rate
  • Closed revenue
  • Average response time
  • Workload or open jobs
  • Month-on-month trend

The wrong version shows forty cards because every department wanted a metric included. The right version shows the numbers that decide the next management conversation.

KPIDenominatorFreshnessDecision
Qualified opportunity rateTotal new enquiriesDaily or weeklyFix source quality or qualification
First response timeNew enquiries requiring a replySame dayReassign cover or tighten routing
Quote acceptance rateQuotes sent in the matched quote cohortWeeklyReview pricing, fit or proposal clarity
Overdue work rateOpen jobs or tasksDailyAdd capacity, escalate blockers or reset expectations
Revenue by sourceWon revenue with known sourceWeekly or monthlyShift spend and attention

Give every KPI a clear definition and every rate a denominator: the total it is divided by. A count such as "five lost jobs" needs context too; five out of ten quotes tells a different story from five out of two hundred.

Sales dashboard: where revenue leaks become visible

Most sales problems are hidden in handoffs. A lead arrives. Someone replies late. A quote is sent. Follow-up depends on memory. A warm opportunity quietly cools.

A sales dashboard makes those leaks visible.

It should show:

  • New leads by source
  • Lead status
  • Follow-up age
  • Quote status
  • Pipeline value by stage
  • Lost reasons
  • Next action owner

This is where dashboards become operational rather than decorative. If a lead has no next action, the dashboard should make that obvious. The sales follow-up dashboard guide shows the stages and daily checks to agree.

For teams using AI receptionists, call handling or unified inboxes, call and message data should connect to pipeline reporting. A missed call, WhatsApp enquiry or website chat should not live in a separate world from the sales pipeline.

Operations dashboard: the command-centre view

An operations dashboard is not only for large companies. Service businesses, agencies, repair firms, clinics, professional practices and logistics teams all have the same basic problem: work is moving, but visibility is scattered.

An operations dashboard can show:

  • Open work by status
  • Overdue tasks
  • Capacity by team or role
  • Response times
  • Service-level risks
  • Upcoming deadlines
  • Delivery blockers

The aim is to spot pressure early enough to fix it. Our operations dashboard guide connects workload, overdue jobs and capacity to the next decision.

Reporting vs command centre

  1. Report

    Monthly report

    Explains what happened, usually after the decision window has passed. Useful for review meetings, but often disconnected from daily work.

  2. Live

    Business command centre

    Shows what is happening now, supports action during the week, and connects workflow, pipeline and ownership.

If the team only opens the dashboard at the end of the month, it is a report. If it changes what happens today, it is a command centre.

Power BI dashboard vs custom business dashboard

Power BI is a strong choice when the job is internal reporting, data modelling and analysis. It is especially useful when the business already works inside the Microsoft ecosystem.

The comparison should not pretend Power BI cannot handle permissions, external access or embedding. Microsoft documents Power BI apps and audiences, external guest sharing through Microsoft Entra B2B, embedded analytics for customers and row-level security for embedded analytics.

The honest comparison is about fit, control and operating complexity.

NeedPower BI dashboardCustom dashboard
Internal reportingStrong fitPossible, but may be more than needed
Microsoft data stackStrong fitDepends on wider system
External stakeholder accessPossible with the right tenant, licence and sharing setupStrong fit when the experience must be controlled end to end
Workflow actions inside the dashboardPossible in adjacent tools, less natural as the core interfaceStrong fit
Bespoke permissions and user journeysPossible, but can become admin-heavyStrong fit
AI-assisted summaries or routingPossible with setupStrong fit when tied to workflow context
Product-like user experienceLimited by BI/reporting conventionsStrong fit

The honest answer is not "custom is always better". If Power BI answers the decision and the team uses it, use Power BI.

Build custom when the dashboard is part of the operating system of the business, not just a place to view charts.

What a business dashboard costs in the UK

Dashboard pricing depends on the number of data sources, the complexity of the data model, whether it is internal or client-facing, and whether the dashboard includes workflow logic.

Broader dashboard work may sit inside:

NeedTypical route
A simple reporting dashboardAmpliDash standalone
Dashboard plus marketing, SEO and lead trackingManaged growth plan
Dashboard plus workflow automationAutomation project
Dashboard inside an internal tool or SaaS productApp or MVP build
Client-facing portal with permissionsCustom app scope

The mistake is pricing the dashboard before mapping the decisions. Two dashboards can look similar and have completely different build complexity because the data behind them is different.

For a dashboard that needs workflow actions, permissions or a portal, Ampliflow's Apps and MVPs route starts from £997 for a focused first build. That is not a price for every integration or reporting project. Bring a sample of the data, the decision to support and the people who need access; we can establish the scope before quoting. Data cleanup, licences and ongoing operation need explicit owners and costs.

What to ask a dashboard developer before committing

Ask for a demonstration using a small, anonymised sample of your actual data shape. It should answer one decision end to end: source record, calculation, displayed result and action. A polished screenshot cannot establish whether those steps work.

  1. Can we trace a number back to its source? Agree definitions and check a sample manually.
  2. What happens when a feed stops updating? Show the last successful refresh and a visible stale-data state.
  3. Can each role see only the records it should? Test permissions with separate accounts, not just an administrator preview.
  4. Who owns corrections and ongoing changes? Name the operational owner and the route for a changed definition.
  5. What can we export or hand over? Confirm data, documentation, access and licensing before signing.

If the existing spreadsheet or BI tool already answers the question reliably, keep it. Commission a custom interface when the missing workflow or access model justifies maintaining another system.

Dashboard audit checklist

Before building a dashboard, answer these questions.

Business dashboard audit

  1. 1

    Users

    Who needs access, and what should each role see?

  2. 2

    Decisions

    What decisions should the dashboard support every week?

  3. 3

    Data sources

    Which tools hold the source data, and who owns access?

  4. 4

    Definitions

    What exactly counts as a lead, qualified opportunity, booking, sale or completed job?

  5. 5

    Freshness

    Does this need to update live, daily, weekly or monthly?

  6. 6

    Actions

    What should happen when a number moves in the wrong direction?

Do this first and the build gets simpler. Skip it and the dashboard becomes a design exercise with no operating value.

How Ampliflow builds business dashboards

Our dashboard work starts with the business decision, not the chart library.

We map the data sources, decide which views each role needs, clean the KPI definitions, then build the smallest useful dashboard first. For many teams, that means a leadership KPI view, pipeline view and operations view. For more complex teams, it becomes a business command centre connected to automation, AI receptionists, apps and internal workflows.

The goal is simple: fewer tabs, fewer spreadsheet reports, and a clearer view of what needs attention.

Bring one recurring decision, an anonymised sample and the action you wish the screen could support. That is a useful starting point for a clarity chat.

For the service overview, see AmpliDash: business dashboard development. If the dashboard needs to become a full internal tool or customer portal, see apps and MVP development.

FAQ

What is the difference between a KPI dashboard and a business intelligence dashboard?

A KPI dashboard focuses on the core metrics a team checks regularly. A business intelligence dashboard usually goes deeper into analysis, filtering, trends and data modelling. Many businesses need both: a simple KPI overview for leadership and deeper BI views for investigation.

Is Power BI enough for a small business?

Power BI can be enough when the dashboard is mainly for internal reporting and the team is comfortable using Microsoft tools. A custom dashboard is usually a better fit when you need a product-like interface, workflow actions, tight brand control, integrated AI behaviour or customer-facing journeys.

What data sources can a business dashboard connect to?

Common sources include Google Analytics, Google Ads, Meta Ads, CRM systems, email platforms, finance tools, booking systems, spreadsheets and custom databases. The exact answer depends on API access, data quality and permissions.

How often should a dashboard update?

Update frequency should match the decision. Sales and operations dashboards often need live or daily updates. Board reporting may only need weekly or monthly updates. Real-time data is useful when someone can act on it quickly.

What is the first dashboard a business should build?

Start with the dashboard that supports the most expensive recurring decision. For most SMEs, that is either lead pipeline visibility, revenue attribution or operations workload.

Business dashboards

Start with the decision your dashboard needs to support

Bring the report you keep rebuilding, the systems it pulls from and the question your team needs answered. We agree the data, the views and who gets access before we quote.

Decisions and KPIs
Data sources and quality
Reporting views
Team access
Explore business dashboards

Use an anonymised example. There is no need to send customer records to explain the problem.