Dashboard development, from the data source to the screen
Dashboard development for companies whose numbers live in five systems and a spreadsheet. We decide with you which measures matter, connect the data from your accounting, CRM, online store, production or product database, model it so the numbers add up, and build the dashboards: in a BI tool for your own team, or custom-built into your app for your customers. Data refreshes on its own, access rules are enforced, and your company owns everything.
- You see the result before it goes live
- Fixed price for the agreed scope, in writing before work starts
- Source code, domain and accounts in your company's name
What is dashboard development?
A dashboard is a screen that answers a recurring question at a glance: are we on target this month, which orders are late, which customers are at risk, how is the new feature used. Dashboard development is everything needed to make that screen trustworthy. Most of the work is invisible: agreeing what each measure means, connecting the source systems, cleaning and modelling the data, scheduling refreshes and controlling who sees what. The charts come last, and they are only as good as what is underneath.
That is why a dashboard project starts with questions, not with chart types. Who looks at it, how often, and what do they do differently depending on what they see? A dashboard that does not change a decision is decoration.
What we build
- Management dashboards. Sales, margin, cash, pipeline and headcount on one page, with the same definitions finance uses.
- Operational dashboards. Order queues, delivery status, stock alerts, support tickets or production progress, refreshed every few minutes for the people who act on them.
- Embedded customer dashboards. Usage, results or savings shown to your customers inside your web or mobile app, each customer seeing only their own data.
- Product analytics. Activation, retention and feature use from your own product data, without sending personal data to a third-party analytics tool.
- Data pipelines and warehouses. The connections, transformations and storage the dashboards run on, so the same number appears the same everywhere.
Signs you need dashboard development
- Someone spends a day each month copying figures from several systems into a spreadsheet for the management meeting.
- Two reports show different numbers for sales, margin or active customers, and the meeting is spent arguing about which is right.
- Your team finds out about late orders, stock-outs or failed payments from customers rather than from a screen.
- Customers ask for reports on their own usage or results, and your staff build them by hand.
- You bought a BI tool, but only a single colleague knows how to change the reports, and the data behind them is a weekly export.
Each of those is a data problem first and a chart problem second, and each is solved by the same steps: agree the definitions, connect the sources once, refresh automatically and show the result to the people who act on it.
BI tool or custom-built dashboard?
Both are legitimate, and we build both. Business intelligence tools such as Microsoft Power BI, Google Looker Studio, Tableau and Metabase are quick to start and let your own staff add reports. Custom dashboards, built with the same web technology as your app, give you full control over design, speed and access, which matters when the dashboard is part of what your customers pay for.
| Situation | BI tool | Custom-built |
|---|---|---|
| Audience is your own staff | Usually the right choice | Rarely worth it |
| Questions change every month | Staff can add reports themselves | Each change is development work |
| Dashboard is part of your product | Embedding possible, with licence and branding limits | Your design, your access rules |
| Many external users | Licences can grow with every viewer | Running cost follows use |
| Instant response inside an app | Depends on the tool and its embedding | Built for your app's speed |
| Ownership of the logic | Inside the tool's files and service | In your repository as code |
What each route costs to build and run is explained in our guide to dashboard development cost.
Measures first: one definition for each number
The most common reason dashboards are ignored is that two of them show different numbers for the same thing. Is revenue booked when the order is placed, delivered or paid? Is an active customer someone who logged in, bought or did not cancel? We write a short definition for every measure on the dashboard, agree it with the people who own the number, and build it once in the data model so every report uses the same logic. The definitions are visible next to the charts, so nobody has to guess.
Access: who sees which rows
A dashboard often shows data some viewers must not see: other regions' sales, other customers' records, salaries. The safe pattern is to filter at the data level, not by hiding a chart. In Power BI that is row-level security: Microsoft's documentation (read 11 November 2026) describes it as filters, defined within roles, that restrict data access at the row level for users of a semantic model, and notes that it applies to viewers, not to workspace administrators, members or contributors. In a custom dashboard we enforce the same rule in the database queries and test it with real accounts. Either way, a user who should not see a row never receives it.
Charts that people can read
Good dashboard design is restraint: few charts, the most important number first, consistent colours and labels in plain words. It is also accessibility. The W3C's guidance on WCAG 2.2 success criterion 1.4.11 (read 11 November 2026) requires the parts of graphics needed to understand content to have a contrast ratio of at least 3:1 against adjacent colours, and explains, for example, that lines in a line graph should meet 3:1 against their background. We follow that, never rely on colour alone to carry meaning, label values directly where possible, and check every dashboard on a phone as well as a large screen.
The data underneath
Most dashboards draw on several sources: accounting, CRM, online store, payment provider, ad platforms, production or the product database. We connect them through their official interfaces, load the data into a warehouse or database in an EU region where needed, transform it into a clean model and schedule refreshes. Every pipeline has monitoring, so a failed refresh raises an alert instead of quietly showing yesterday's numbers as today's. Where the questions behind the numbers still need working out, our data analytics consulting starts there.
How we build it
- Questions and audience. Who uses each dashboard, how often and which decisions it supports.
- Measure definitions and data map. Every number defined, every source located, data issues listed.
- Scope and price in writing. Dashboards, sources, refresh, access rules and what the first phase leaves out, signed before work starts.
- Data first. Pipelines and the data model, checked against figures your finance or operations team already trusts.
- Dashboards and review. First version on real data, reviewed with the people who use it, then refined.
- Handover. Documentation, measure definitions and admin access to your team, with monitoring in place.
You own it
Reports, data models, pipelines and any code sit in your company's own accounts and repository. Data stays in your systems or in a warehouse in your company's name in an EU data region, and personal data is processed under a written data processing agreement. You can keep us to add dashboards, maintain them yourself or hand them to another team.
Dashboards often sit on top of systems we build: see custom CRM, light ERP, internal tools and business process automation. Already on Power BI and need the model, DAX or security fixed? See our Power BI consultant service. For device fleets and sensor data, see IoT software development; for meter, grid and charging data, energy software development; and for the difference between descriptive, diagnostic, predictive and prescriptive work, our guide to types of analytics. See prices.
Questions we get
What is dashboard development?
It is the work of turning data from your systems into a screen that answers recurring questions at a glance. It covers choosing the measures, connecting and cleaning the data, modelling it so numbers add up, designing the charts and setting who sees what. The visible dashboard is the last and smallest part of that work.
Should we use Power BI or build a custom dashboard?
Use a BI tool such as Power BI, Looker Studio, Tableau or Metabase when the audience is your own staff and the questions change often. Build a custom dashboard when it is part of your product for customers, needs your own branding and access rules, or must respond instantly inside your app. Many companies need both.
How often does the data refresh?
As often as the decision needs. Most management dashboards refresh nightly or hourly; operational ones every few minutes; some, such as a live order queue, in real time. More frequent refresh costs more to run, so we set it per dashboard, not by default.
Can our customers see their own data in our app?
Yes. An embedded dashboard shows each customer only their own records, enforced in the data layer, not only in the screen. We test it with real accounts from several customers before launch.
What if our data is messy?
It usually is. The first phase often fixes definitions and duplicates in the source systems, because a dashboard on top of bad data spreads wrong numbers faster. We list the data issues we find and fix them at the source where possible.
Who owns the dashboards and the data?
Your company. Reports, data models and any code sit in your company's own accounts and repository, the data stays in your systems or an EU-hosted warehouse in your name, and you can change supplier at any time.
Get a written quote
Tell us in a few lines what you need. We reply within one working day with a written quote: scope, timeline and a fixed price or an hour estimate. Asking is free and commits you to nothing.