Using AutoCat and Regex

AutoCat is certainly one of Tiller’s “killer features” and its ability to process Regex expressions makes it even more powerful. But, I learned something about using AutoCat Regex that seems obvious in hindsight but worth sharing.

If you want to use one rule to assign one category to several different vendors, you cannot use the same rule to simplify the description. Normally, I prefer to use AutoCat to Categorize and replace the complicated Description I get from the bank but I can’t do that if I also want to use one rule to assign the same category to multiple vendors.

For example, I use one very simple Regex expression to categorize several fast food restaurants as “Fast Food.”

On the AutoCat tab: I set the Category to “Fast Food” and the Description Regex column to:
mcdonald’s | mcdonalds | wendys | wendy’s

Using Regex makes the rule case insensitive which automatically catches many more transactions than a standard AutoCat rule where case matters. As an aside, I have no idea why some of the transactions we get from our banks are in all caps and others aren’t but we do. Likewise, I don’t know why the name of the vendor in the Description downloaded from Wells Fargo changes for no reason. A local restaurant called The Purple Onion might be described as Purple Onion, The Purple Onion, purple on, purp xxx. But, I digress!

To overcome the inconsistency and case sensitivity and to reduce the number of rules needed to categorize similar vendors, I started using Regex in AutoCat. It works beautifully but, it has one drawback that I didn’t initially consider.

If I include different fast food restaurants in the same rule, I can’t simplify the Description. I have to live with the long and inconsistent description I get from Wells Fargo or change it manually or create a separate rule for each restaurant.

Of course, the same problem would occur if I wanted to classify Lowe’s, HomeDepot, and Tractor Supply as “Hardware.” The categorization would work but I’d have to live with the complicated description I get from Wells Fargo or change them manually.

One solution – sort of: I could create separate AutoCat rules for each vendor using all of the variations of their name that I get from the bank in lower case separated by the “pipe” character in the Description Regex column. And, in the same rule, I could add a simple and consistent Description in the Description column.

As I said, this seems pretty obvious in hindsight but maybe worth sharing if anyone on the Forum is considering this approach.

To be very clear, this is not a flaw in Tiller.

Hi @ScottC - hoping you saw this great discussion here:

Thanks, Heather.

I appreciate the “heads up.” I’m trying to learn how to use REGEX functions along with the FILTER and QUERY functions to customize my views of the Transactions table.