How can we help?

Ask a question or search our knowledge base:

Using Advanced Text Matching in Sift Workflows


Users of Sift Workflows sometimes want to build policies on text fields. This includes criteria like:

  • Block all email addresses of the form johndoe*
  • Approve all IPs that are of the form 10.*.*.*
  • Block listings that contain ‘Disney’ or ‘disney’
  • Review all orders where the item name contains ‘gift card’

To enable building criteria of this form, we have Advanced Text Matching options:

  • Partial vs whole word matching → For any new criteria you set up on text fields, Sift Workflows will match on whole words by default. This means that a filter for ‘text me’ will not trigger on ‘text megan’. However, you can force an override to match on substrings (ie partial words) via the Substring Match checkbox. Something is a substring if it exists anywhere in the string, no matter what's to the left or right of it. So, if you check the "Substring Match" box, then ‘car‘ would match ‘A car for sale‘, ‘card‘, ‘‘, etc. If you do not check "Substring Match", we'll only find ‘car‘ if it's not bordered by a non-letter, non-number character, including the start or end of the string. So, ‘car‘ would match ‘car.‘, ‘free-car‘, ‘I have a car for sale‘, ‘‘, etc, but not ‘card‘, ‘5car4‘, or ‘I have a carforssale‘.
  • Case sensitivity → By default, Sift will ignore case on text fields. So ‘’ will trigger on ‘email me at’. This may be overridden by checking the "Case Sensitive" checkbox.
  • Wildcards → We also have support for wildcards. The two wildcards we support are ‘*’ (for 0-many characters) and ‘%’ (for 0-1 character). For example, ‘hack*’ will trigger on ‘’, ‘’, ‘‘, etc. ‘free bitcoin%%‘ will trigger on ‘free bitcoins!‘, ‘free bitcoins‘, and 'free bitcoin', etc. 

Please reach out to if you have any questions. 

Have more questions? Submit a request