An analysis toolkit used across the team

Situation

A large professional services practice ran the same core financial analysis at the start of nearly every engagement: profitability by customer and product, working capital behaviour, cost structure and cash trends. Each team built it again from scratch, in a new spreadsheet, on a new deadline.

The work was slow, it varied in quality depending on who did it, and much of the first week of an engagement went into producing analysis that was essentially standard.

Constraint

Client data never arrives twice in the same shape. Different systems, different chart of accounts structures, different levels of detail and different file formats. Any tool that only worked on tidy data would be used once and abandoned.

Adoption was the other constraint. The people who would use it were finance and advisory professionals, not developers, and they were under deadline pressure. A tool requiring training or a change of habit would have been ignored.

What was built

An analysis toolkit that took raw client data, standardised it, and produced the full set of standard analyses automatically, with consistent output ready to go into a client document.

It was built to handle messy input rather than to assume clean input, with the mapping step made explicit so a user could tell the tool how a client’s data was structured instead of reshaping the file by hand. Because the analysis was repeatable, teams could rerun it as better data arrived instead of treating the first cut as final.

It also went further than the manual version. Once the standard analysis was free, deeper cuts of the data that nobody had time for previously became routine.

Who it was built with

Teams across the practice, through use. Early versions were run alongside manual analysis on live engagements so the output could be checked, then the toolkit absorbed what people asked for. Feedback from the people under deadline pressure decided what was built next.

Outcome

The toolkit was released across the practice and used on many large clients. It removed days of manual analysis per engagement, reduced cost, made output consistent between teams, and produced insight from client data that the manual process had not had time to reach.

What this looks like for you

If your team rebuilds the same analysis every month or every engagement, the work is standard even though the data is not. Automate the standardising step and the analysis follows for free. See FP&A automation and AI.