Knowledge Base

Favoring, Not-Favoring Fields Based on Input’s Like “Unknown”

See More Articles

v3 platform This article refers to the v3 Dedupely platform. If you use a v4 Dedupely integration please visit the documentation under the v4 tab here.

Every now and then, your data will contain these types of “filler values” (note, “unknown.last.name”):

You’ll see that, by default, Dedupely might select these types of values for the final merge. Let’s take a look at the default selection when looking at the custom merge dialog:

As you can see, Dedupely’s algorithm isn’t smart enough (or human enough) to know that names with “unknown” shouldn’t be selected.

There’s a solution for this. Go to Settings > Advanced > Prioritized Fields

Here you’ll see a section called “Based on field input…”.

From here, it’s easy to set up a rule to avoid having Dedupely ever favor field input with those keywords:

This is not case sensitive, so it works for “Unknown” or “UNKNOWN” as well. Also, you’ll notice John Smith’s field actually has “.last.name” added as well. This will still work since it searches the text for that “unknown” keyword.

Let’s Save that, and when it’s done re-scanning, then view the custom merge dialog again to see the preview:

You can see that “Smith” gets automatically selected now instead of “unknown.last.name”.

A quick note: If the other last name field was blank, “unknown.last.name” would get selected since it takes a better-than-nothing approach to the merge.