Reporting automation: from 2.5 days to 2 hours
Manufacturing company
Karol Jurewicz led this project at a manufacturing company where he worked before founding cm-opti. We use the same method in client projects today.
Industry and type of organization
A manufacturing company. The monthly management report was produced by hand, assembled from many data sources.
Business problem
The monthly report for the board was produced by hand and took 2.5 days: gathering numbers from many sources, putting them together and checking them. The board saw the previous month's results a few days late, and every manual transfer of numbers risked a mistake.
Behind this lay a deeper, non-technical problem: the report included many indicators, but only some of them actually influenced the board's decisions.
Approach and scope of work
The project started with the question of which decisions this report was meant to support, and with defining the KPIs. Then the data was organized and unified into a single source of truth, so that everyone computes the same indicators the same way. The recurring collection, cleaning and merging of data from many sources, known as the ETL process, was automated in Power Query. Changes to the ERP system were also designed so that the data comes out of it ready for the report, without manual rework.
The final step was the visualization: the state of the company and the key metrics on a single screen, so that they answer the question of how the business is doing before anyone has time to ask it.
Business impact
2.5 days → 2 hours
preparing the monthly report
the same report, a fraction of the time
The time recovered went back to the team for analytical work, instead of assembling numbers by hand. The report is now ready almost immediately after the month closes and shows the current picture of the company, so decisions have a solid basis.
Scale of the implementation
A monthly reporting cycle, data from many sources. The process was standardized once and can be repeated in every subsequent period without building it from scratch.
If you suspect your data would answer questions you settle today by gut feel, describe your reporting process to us. We will see what the data can tell you.
Related articles in the cm-opti knowledge base
Service area: Machine learning and forecasting (data analysis, reporting and decision support).