Say you work for a financial services company in the lending business. You receive loan applications that you process in exactly 2 days. For every 20 applications, you approve one. This means your approval ratio is 5%.
Say below is how your applications and approvals look like. Your approvals lag the applications by 2 days.
Now let’s look at your approval ratio. If we don’t consider the time lag of 2 days, your approval ratio varies from 4% to 10% (red line). This is incorrect. I just prepared this mock data assuming 5% approval ratio (green line).
This is how time lag impacts your derived metrics.
Time lag is also a critical input to root cause analysis. For details, see Root Cause Analysis.
Let’s look at a couple of examples of varying time lags in the real world.
- Uber: if the number of available cars in an area drop, it might impact the number of bookings in that area, almost immediately.
- Ecommerce: if a significant number of workers call in sick for a day in a warehouse, it will impact the number of shipments for the day and the day after. The number of deliveries will get impacted with a further delay of 1-2 days. Order cancelations and returns might also increase with their own time lags.