Comparing Data Formula?

[MJTOWNEND]

Hi,

I have two sets of data (customer names). The two sets of data are from different sources. Even though they are the same list of customers they are all have slightly different names (not correlated).

They will be partly the same, maybe one or two words the same.

Is there a formula that will be able to highlight and link the two sets of data together?

I don't want to go through manually and link these.

Many thanks,

Matt the novice

 

[Asala42]

These situations can be tricky, especially if there's no rhyme or reason on the differences between lists. There's usually a few things you can do to narrow down the manual effort though.

If it were me I'd start by trying out an exact match formula like VLOOKUP or COUNTIF. This will give you an idea how many records you have that are not matching. Even if only say 10% of your records match exactly, at least you're reduced your manual effort by that much.

From there I'd test use a helper column next to each table that have a better shot at lining up in both tables. It really depends on what you're usually seeing to cause mis-matches in names you know to be the same. Here's some common examples:

=TRIM(CLEAN(B2))
Removes trailing and leading spaces, common with some export data.

=LEFT(B2,10)
Will produce the first 10 characters. This might be enough to find a match for most entries.

=IFERROR(LEFT(B2,FIND(" ",B2)-1),B2)
Will output the characters in a string that are to the left of the first space (if say these are company names and some).

We can probably help you better if you illustrate the most common differences in text between your mismatches by showing some examples. My plan would be to match the records you can, and then sort the mismatches side by side.

As an aside, is this an ad hoc 1 time effort or something you'll need to repeat? If it's something you expect to repeat, you may want to use this effort to build a table that you'll be able to use later (NameX from Table1 = NameY from Table2). Then at least existing customers would be easy to match later.

 

[Joyner]

To add to that, some sample data and the differences and similarities will help, but most likely using a combination of the lookups and wildcards may get you most of the way there.