Data quality – Consistency

You can’t achieve high data quality without consistency.

Consistency here refers to gathering data from every interaction.

The alternative is having gaps in your data, where a set of interactions are not reported.

Consistency is insidious, the thought of having inconsistent data can make analysts unsure of their work.

The reason is that you can’t tell an interaction has happened or not.

You can’t assume that the user has done a specific interaction or not.

In the best-case scenario, the unreported interaction is not important.

Worst case, an important interaction is unreported.

The problem is, you won’t be 100% sure it is unimportant until you start tracking that interaction.

Only the data can tell you the importance of that specific interaction.

– Samy