Data connection
Business Central, SAP Business One, Sage, SharePoint workbooks and bank files read into one semantic model, refreshed on a schedule.
Dashboards built on a reconciled data model, with access by role, for companies that run on an ERP and a stack of Excel workbooks.
A dashboard is the last step. Most of the work is the model underneath it.
Business Central, SAP Business One, Sage, SharePoint workbooks and bank files read into one semantic model, refreshed on a schedule.
Every page is checked against the figures finance already trusts, such as the board pack or the trial balance, before anyone sees it.
Row-level security by entity, cluster or hotel, and separate published apps for operations and the board.
Bonsaï Camp runs a Power BI Analytics and AI programme so your analysts can extend the model without us.
We use Power BI when it is the right tool, and we say so when it is not.
You mostly need to read figures from one or two sources, the audience is internal, and nobody has to approve, correct or enter anything in the same tool.
The work needs data entry, approval workflows, alerts assigned to a named owner, or a check that stops a bad import. That is a custom application or a Business Cockpit.
Power BI for the board view, an application for the operational side, fed by the same data. Read how we decide.
Real systems, in production or in delivery. Client names are withheld.

Management no longer waits weeks for an assembled board pack. Consolidated P&L, budget and working capital, live from the ERP.
Constraint. Five entities, an on-premises ledger and a board pack assembled by hand after month end.

Sales, margin, cash and supply chain read nightly from SAP, from the P&L waterfall down to each product range.
Constraint. Figures lived in SAP and in forecast workbooks and were assembled by hand each month.

Each morning, management knows where the month will land against budget, from one set of figures reconciled with accounting.
Constraint. Hotels reported on separate timetables and group figures were assembled afterwards.
Thirty minutes is enough to say whether Power BI is the right answer.