Data Quality

Data quality is how fit a dataset is for its purpose, usually broken down into accuracy (are the values right), completeness (is anything missing), timeliness (is it current), and consistency (do the parts agree). For most companies it is a supporting concern; for a data provider it is the entire product, because the consumer never sees your pipeline, only whether the numbers they act on are correct.

Quality is not a one-time cleanup but a standing property that has to be monitored, because sources degrade, schemas drift, and a dataset that was clean last month goes wrong quietly. The mature approach is continuous checks - freshness, volume, distribution, referential integrity - that flag a problem before a customer does, plus the provenance to trace a bad value back to its source. Poor quality is expensive precisely because it is invisible until someone has already traded or trained on it.