Compare CSV and TSV exports

Choose your files

Open Two files. Choose the earlier export for Baseline, then the later export for New output. In this example, both files list three people and their payments. We want to see whether anybody's payment changed. Open Comparison options and look for CSV and TSV options.

Read the file correctly

Start with Automatic for both Delimiter and Text encoding. A delimiter is the character between fields. Here it's a comma. In a tab-separated export it's a tab. If your supplier specifies a semicolon or a pipe, you can select that separator yourself. Text encoding controls how characters are read. If accented names look wrong, check which encoding the supplier used, then select it here.

Look at the first line of our file. Record ID, Name, and Amount are column headings, so leave First row contains column names checked. If your first line already contains a person's payment, clear the box. You can then refer to columns by number. The first column is one, the second is two, and so on.

Match people across exports

Leave Match rows by set to Row position and select Compare. We get twelve differences. Look at the first pair. Alice is first in the baseline, while Chloë is first in the new file. Their amounts differ because we're looking at two different people. Row position is useful when someone changing the order is itself something you need to catch.

For the payment check, we want to follow each person wherever they moved. Open Comparison options again. Under Match rows by, select Key column. In the Key column field, type Record ID, the name from our header. Each ID must appear only once per file, and every record needs one. A duplicate or empty key produces an error identifying the row to check.

Select Compare again. The result now says Same. Follow Alice's ID, one. Her amount is one hundred in both files. The other two people also kept their original amounts. The first run answered whether each row stayed in place. This run answers whether the records themselves changed.

Leave out a known date field

For a second example, choose the two inventory files from the Guided Showcase. Switch Match rows by back to Row position. With Ignore columns empty, Compare finds six differences. Four are update dates on existing items. For a stock count, type Last Updated in Ignore columns. Use the heading exactly as it appears. You can enter several column names or numbers, separated by commas.

Review the stock changes

Select Compare. Two findings remain. On the ergonomic chair row, Units went from eighteen to sixteen. The other finding is a new row for a task lamp with twenty-two units. Ignoring the date field kept both of those changes visible. To include the dates in your next check, clear Ignore columns and run Compare again.