Unfamiliar data.Start by getting
your bearings.

Preflight examines structured data and helps you notice what deserves a closer look.

An illustrative examination

12 source files

Records before 2014 make up 3.3% of the data

but account for 88% of repeated Customer_ID occurrences

Treating those records separately changes the observed duplicate occurrence rate in the remaining data:

2.4% → 0.3%

Look closer

Synthetic data, for illustration.