DataMosaicCompare it properly.
Files, tables and databases, compared step by step, from the formula to the query.
All posts
RSS feedInventory accuracy formula: three measures on one count
One store, ten lines, one count. Worked out three ways, it scores 50%, 94.8% and 90.9%, and the honest answer depends on the question you are asking.
Compare data between two databases: SQL Server to Oracle in 3 steps
Tills on SQL Server, the ERP on Oracle, one invented day of sales: totals first, then a row hash both engines agree on, and the five type traps that make equal rows look different.
Compare two columns with VLOOKUP, XLOOKUP or Power Query
Three ways to match two lists on a key, run on one invented price file, and the two traps that make a lookup say a row is missing when it isn't.
Compare two tables in SQL for differences: three queries that work
EXCEPT both ways, a full outer join on the key, and a hash per row, run on one invented day of sales, with the NULL traps that make each one lie.
Compare two CSV files for differences: in Excel, Python and PowerShell
When the rows come in a different order, match them on a key first. Three ways to do it, and the four things that make identical files look different.
How to compare two Excel files and highlight differences
Three ways, from one formula to a lookup on a key, and how to tell which one your files need.
Data reconciliation: techniques, and a worked retail example
Counts, totals, keys, fields, hashes and grain: what each technique catches and misses, worked through on one chain’s Monday morning.