When we’re scheduling a review, the analysis is usually the core, but most of the elapsed time goes somewhere else.
In many cases, this can’t be helped, regardless of whether we’re familiar with the environment and team.
Here are a few key things that affect the timeframe:
Access. To systems and models (the specifics depend on the nature of the review). That means a request, approvals, and someone setting it all up. It's not unusual for this to go through multiple iterations. This often adds the most time.
Documentation. The documents may be spread across shared drives, someone’s inbox, a Confluence space, etc. The person who wrote the model spec may be on extended leave, or have moved on.
The data extract. The first extract usually won’t be the right one. Date ranges, filters, fields across multiple systems, and very often a difference in understanding what a field name means or represents. Three or four rounds is not abnormal, and each one involves a request, a wait and a check.
People. The person who knows the system is usually busy with several other things. Interviews get scheduled, moved, and scheduled again.
Root cause analysis. If this is part of the review, it can take a long time, usually spent on testing various hypotheses.
So a review that would need four weeks of core analysis can take more than three months end to end.
Worth knowing and planning for this upfront, so that there are no timeline surprises later.
Disclaimer: The info in this article is not legal advice. It may not be relevant to your circumstances. It was written for specific contexts within banks and insurers, may not apply to other contexts, and may not be relevant to other types of organisations.