This as an example of the one that got away. The analyst that worked on this audit carefully planned the analysis, went through all the effort of carrying out the analysis, documenting her work, and having it peer reviewed. But as the engagement progressed, it became clear that the results of the analysis would not meaningfully contribute to the audit findings or recommendations.

It was a tough call, and caused a lot of disappointment for the analyst, but in the end we don’t do analysis because it’s interesting of because data is a shiny buzzword. We do it to add value to our audits. No regerets. Just a mature approach to performance audit data science.

There were however side benefits from the work done. The domain knowledge gained by the analyst has already been applied to other audits in the transport space, and the insights from the analysis have informed our annual planning process.

Click here to read the report.