Skip to content
Technology transformation expertise firm · ARDIANE Group Innovate, secure, succeed

Glossary

Self-service BI: definition of self-service business intelligence and its limitations

Domain

Data

In brief

Self-service BI refers to a way of organising business intelligence in which business users build their own reports and analyses, without ordering them from the central BI team. A term from common usage, self-service BI is practised within a framework known to the organisation: tools provided, prepared data and publication rules.

Page contents

Definition of self-service BI

Who builds the report: the BI team, or the person who needs it? Self-service BI refers to the second case. Business users (management controllers, operations managers, research analysts) put together their own reports, dashboards and analyses, with tools the organisation provides.

The term comes from common usage and vendor vocabulary, which distinguish degrees of it. Microsoft’s documentation (learn.microsoft.com, Fabric adoption roadmap, article on content ownership) describes three strategies: business-led self-service, in which the business unit owns and manages all its content; managed self-service, in which a central team manages the data and the business manages its reports; and a so-called enterprise strategy, in which the central team manages everything. The Tableau Blueprint guide (help.tableau.com) distinguishes three governance models: centralised, delegated and self-governing. These pages were read on 12 November 2026.

Self-service BI, central reporting, exploration by analysts, shadow BI: not to be confused

Practice Who produces What separates it from self-service BI
Central team reporting The BI team, for all readers The business reads reports; it does not put them together
Exploration by data analysts Specialists, in SQL or in notebooks, on detailed data A distinction of usage: it relies on code, which self-service does not require of its authors
Shadow BI Users, from extracts and local files It escapes the framework: neither known nor approved

The expression shadow BI, from common usage, is modelled on shadow IT, and the last row depends on the framework rather than the tool. Microsoft’s documentation thus distinguishes business-led self-service from shadow IT: in both cases business users create and manage content, but shadow IT bypasses IT and is not approved. Our shadow AI entry describes the same mechanism for AI tools.

What self-service requires, and what it leaves open

Both sets of documentation describe three foundations:

  • prepared and reused data: Microsoft speaks of semantic models shared between many report authors, the role played by a semantic layer;
  • a visible separation between certified content and exploratory content: Tableau Blueprint recommends “sandbox” projects distinct from certified projects;
  • support for authors, training and documentation first and foremost.

To these is added a data catalogue, so that authors can find the authoritative source.

Self-service leaves three questions open. Authors who start from raw tables may calculate the same indicator in several ways. Content useful to one team may cope poorly with growing volume, a case Microsoft anticipates by organising its transfer to the central team. And putting together a dashboard does not say how to present it to a governance body, the subject of our data storytelling entry.

Self-service at work: time to fill tracked by region

Consider a temporary employment agency whose branch managers want to track the time taken to fill assignments. The BI team publishes a certified model (assignments, branches, qualifications) in which the “time to fill” indicator is already defined. A regional manager puts together, in her workspace, her dashboard by branch and by trade, without writing a query.

She adds a local file of targets: this part remains exploratory until it has been reviewed. When other regions ask for the same dashboard, it is put forward for certification, and the targets join the shared model.

Our business intelligence and analytics page describes how to open up self-service by levels of autonomy and certify content within it.

FAQ

Frequently asked questions

Another question? Ask us

What is meant by self-service BI?

Self-service BI means that reports and dashboards are put together by the people who use them, instead of being ordered from a specialist team. The term comes from common usage, and tool vendors describe degrees of it. Microsoft's documentation distinguishes business-led self-service from managed self-service, in which a central team manages the data; the Tableau Blueprint guide describes three governance models: centralised, delegated and self-governing.

How does self-service BI differ from shadow BI?

Self-service BI is a practice known to and approved by the organisation, which provides its tools, data and rules. Shadow BI, a term from common usage, refers to reports and files produced by users outside that framework, from untracked extracts. Microsoft's documentation draws the same line with shadow IT, which it describes as a way of bypassing IT.

Does self-service BI remove the need for a BI team?

Self-service BI shifts the BI team's work without removing it. In the managed self-service described in Microsoft's documentation, a central team manages the data, while business users are responsible for their reports. The BI team then prepares shared models, certifies content and supports authors.

A project to bring to completion?

A first conversation is enough to understand your challenge, identify the right people and tell you frankly what we can bring.