Service Business intelligence

Power BI consulting: from your business systems to metrics with a single calculation rule

For leadership and finance teams who spend meetings arguing over which number is right: we integrate your business systems, write each calculation rule once and deliver Power BI dashboards in which every department sees the same metric. Your team is trained to carry the work forward.

Summary

What tends to go wrong

Each system holds part of the story: the ERP has invoicing, the CRM has the pipeline, the finance spreadsheet has the adjustments. Pulling it together takes days, the calculation changes depending on who built the report, and meetings end up arguing over which number is right instead of what to do about it.

Who it is for

Executives, finance teams and managers who track the business through hand-built spreadsheets, and IT teams who need to give those teams a reliable foundation for analysis.

Deliverables

What gets delivered

  • Integration (ETL) routines with scheduled loads into the central repository
  • A data warehouse with a dimensional model
  • Power BI dashboards with the agreed metrics
  • Training for dashboard users and for those who will maintain them
  • A map of data sources and a list of priority metrics, coming out of the assessment
  • Metrics dictionary and a set of questions with reference answers

Steps

How the engagement runs

  1. Questions

    The decisions the dashboards will support and the metrics that measure them.

  2. Sources

    Mapping the source systems and spreadsheets and the quality of their data.

  3. Model

    Integration, data warehouse and calculation rules validated with the business.

  4. Dashboards

    Build, check the figures with users and publish.

  5. Training

    Preparing users and the people who will take the work forward.

Roles

Who takes care of what

Trustwise

Integration, modeling, dashboard development and training.

Your company

Name a business owner for each metric, provide access to the sources, validate the figures and hold the Power BI licenses.

Engagement

Where to start

It starts with an assessment of your data sources and the questions management needs answered. The proposal that follows, either a project or ongoing support, can be split into deliveries by area, for example sales first and finance next. Commercial terms are set out in the proposal.

Scope

What is included

  • Data integration (ETL)

    Extraction from source systems, cleansing and scheduled loading into a central repository.

  • Data warehouse and data model

    A dimensional star-schema model with facts, dimensions and business rules written down in one place.

  • Data analysis

    Exploring sales, inventory, customer, marketing and finance data to answer specific management questions.

  • Power BI dashboards

    Interactive Power BI reports, backed by SQL Server, with the metrics management follows.

  • Team training

    Training for the people who use the dashboards and for those who will maintain them, built around your own project.

  • Dashboard support

    Report performance tuning, gateway and scheduled refresh, and evolving the metrics over time.

  • Metrics dictionary and reconciliation

    Each metric with its rule, granularity, source and owning department in writing, and the dashboard reconciled with the source system.

Warning signs

Signs that a BI layer is missing

Having dashboards is not the same as having BI. What matters is where the numbers come from and whether everyone uses the same rule to get to them.

  • Management reporting that relies on exporting reports and pasting them into spreadsheets
  • Different departments presenting different values for the same metric
  • Dashboards that exist, but nobody can explain where the numbers come from
  • Slow Power BI reports or refreshes that keep failing
  • Calculation rules known only to one or two people

Approach

How a BI project is built

The visible part of a BI project is the dashboard, but most of the work happens before it. First we list the questions management wants answered and the sources that hold the data. Then comes integration: ETL routines that extract, clean and load the data into a data warehouse, usually on SQL Server. On top of it we build a star-schema model with fact and dimension tables, where each business rule is written exactly once. Only then do the Power BI dashboards come in. We follow that order so every figure on the dashboard can be checked, rule by rule, against the source system.

Rollout

Once the dashboard is ready

A dashboard has to stay current and reach the right people. These items are usually discussed at this stage.

  • Publishing to the Power BI service, with workspaces by department
  • Gateway for on-premises data and scheduled refresh
  • Access based on each user's role, with department-level permissions and row-level security
  • Documentation of the measures and sources behind each dashboard

Metrics

Every number with a rule, a source and a check

The data engineering behind a dashboard usually stays out of sight. We document it: what goes into each metric, at what level of detail it is calculated and how to check the result against the source system. The same approved model can serve as the basis for a natural-language question pilot, but the BI project is contracted and delivered without depending on AI.

  • A metrics dictionary with calculation rule, granularity, source and owning department
  • A dashboard reconciled with the source, with any differences explained
  • A set of questions with reference answers, used to check the dashboard and, if there is a pilot, the natural-language answers

Technologies

Platforms and tools

  • Microsoft Power BI
  • Microsoft SQL Server as the data warehouse platform
  • DAX and Power Query
  • Microsoft ETL tooling such as SSIS or Azure Data Factory
  • Natural-language features in Power BI and Microsoft Fabric, evaluated against the client's capacity, licenses and permissions

FAQ

Questions we often hear

Do we need a data warehouse, or can Power BI connect straight to our system?

For a one-off dashboard, a direct connection may be enough. When there are several sources, history to keep or calculation rules that must be the same for everyone, a data warehouse is usually the way to go: it takes query load off the production system, keeps the history and holds the rules in one place. The assessment shows which route fits your case.

Why does a Power BI report become slow?

The usual suspects are a model without a star schema, columns and tables loaded but never used, DAX measures that force row-by-row calculation, and DirectQuery against tables without suitable indexes. In many cases the fix lies in the model rather than in a new license or a bigger server.

Who maintains the dashboards after the project?

The aim is for your team to be able to do it, which is why training is part of the service. If you prefer, Trustwise can keep evolving the dashboards under an ongoing support agreement.

Can you use data that lives in spreadsheets today?

Yes. Spreadsheets often hold targets and adjustments that exist in no system. They can feed the model as long as their layout is stable and someone is responsible for keeping them up to date.

Which Power BI licenses will we need?

That depends on how many people will consume the dashboards, the data volume and how often it refreshes. Buying licenses is your company's decision; the assessment lays out the options and what each one means in your situation.

Does a BI project require an AI project?

No. The BI project stands on its own. If the company later wants to test natural-language questions about its metrics, the approved model and the set of reference answers are the starting point, and that is handled as a separate pilot after capacity, licenses and permissions have been checked.