I want to include the ticketing keyword in t_keyword and I'm getting stuck on the parameters. Basically we want to be able to select all customers who bought a ticket to a performance with a particular keyword. Can anyone help?
No. Segments (keycodes) in an extraction are completely separate criteria sets joined by OR. So if I had one segment with one of your new criteria for keyword and a second segment with a criterion for season, I would get anyone who had purchased something with the keyword (regardless of season) and anyone who purchased something in that season (regardless of keyword) but not just people who bought something with that keyword in that season.
Any time you want to build a list or extraction that looks for a single transaction (a single ticket purchase or single contribution) that meets multiple criteria, you have to use IN as the operator and all the criteria have to be looking at the same table. Think about it like rows and columns on a spreadsheet. A ticket purchase generates a row on the ticket history spreadsheet. Any criteria that looks at that spreadsheet can be combined using IN (or any of the alphanumeric operators) to look for a single row, a single purchase, that meets all the criteria. But your view is an entirely different spreadsheet. So if you use criteria for the view, you will be looking for one row on the view spreadsheet that meets the view criteria and one row on the ticket history spreadsheet. For more detailed explanations of these concepts, you might want to check out the Criteria Sets document.
If you want you could add some additional columns to your view for things like season and maybe order date (be careful if you do that though that everyone understands that they still can’t use those criteria with other criteria from the ticket history unless they are looking for a HAS style result). Alternately you could add a column to your ticket history for keyword and then just add a criterion to reference that column. The trick with that is what to do if a performance has multiple keywords. You could certainly record them all in the column, but I’m not sure how or if you could set up the criterion to parse apart the data (I’m a bit of a SQL novice).
To sum up, what you have set up is definitely good and will be useful, you just need to know what results it can give you and what results it can’t give you. This all comes back to the differences between IN and HAS, which are the trickiest part of criteria sets, whether you are dealing with custom criteria or standard criteria.
Kevin Sheehan
Documentation & Learning Resources Specialist
Tessitura Network
1 888 643 5778 ext 329 Office
ksheehan@tessituranetwork.com
I would like to learn more about what Kevin describes above as a 'weighting procedure' for keywords. We would like to track people's interests using the keywords attached to the productions they have purchased, but want to make sure that they actually have an interest in the genre as opposed to just buying that type of show once. A weighting procedure sounds like something that would help.
I have searched the documentation and can't find anything other than the standard keyword setup. Can anyone point me in the right direction?
Becci Brace - TN PAC