Get oriented before your demo

Extract, Transform, Load (or Extract, Load, Transform): moving data from source systems, reshaping it, and storing it where analysts can use it. ELT performs the transformation after loading, inside the target platform.

An automated, repeatable sequence of steps that moves and prepares data, typically on a schedule or trigger.

Measuring what a dataset actually contains: data types, missing values, value ranges, patterns and uniqueness.

Checking records against rules, for example required fields, valid ranges or consistent keys, before they are used for reporting or models.

Automating parts of the machine learning workflow, such as comparing algorithms and tuning settings, so that more teams can build useful models.

Key performance indicator: a measure that shows progress toward a business goal, such as average processing time or monthly revenue.

A measure is a number that can be aggregated (sales, count). A dimension is a way to slice it (region, product, month).

Changing an assumption, such as price or demand, to see how a forecast or result would change.

Asking questions of data in everyday language and receiving charts or tables as answers.