T-Stats and Disintegrating Data

Thanks. To follow up, I took a snap shot of the T-stats report a few years ago of historical sales, and returned to that same report again yesterday. The snapshot and the T-stats report differed in ticket numbers but not dollars.

I can understand constituents changing based on merging. But, having number of tickets sold with past shows change is odd indeed. No activity would have likely occurred on closed seasons, but anything is possible I guess.

David Anderson | Director of Marketing

Kansas City Ballet | Todd Bolender Center for Dance & Creativity

500 W. Pershing Road, Kansas City, MO 64108

d: 816.216.5587 | o: 816.931.2232 x1323 | f: 816.931.1172

Buy tickets now at kcballet.org

From: Tessitura Marketing Forum [mailto:forums-marketing@tessituranetwork.com] On Behalf Of Amanda Freeman
Sent: Tuesday, September 12, 2017 10:49 AM
To: David Anderson <danderson@kcballet.org>
Subject: Re: [Tessitura Marketing Forum] T-Stats and Disintegrating Data

Hi David,

In general you'll likely get more replies to T-Stats questions (especially if you're hoping for a response from JCA) if you post in the T-Stats forum.

In terms of your current question, T-Stats is reloaded regularly from Tessitura, so if the performances in question had any activity since you saved those reports, it is possible for the results in T-Stats to change, either due to returns or exchanges. Constituent counts can always go down due to merges, where if a duplicate and an original both had separate orders, after they're merged that will be reported as one constituent instead of two. For ticket counts to go down, it would generally indicate some sort of activity in Tessitura, or possibly your T-Stats load failed recently. 

Another thing to consider: when you mention saving the report to Excel and pulling it again later, do you mean you had the same report still saved in the same workspace and you just opened it up to see the current results? Or did you recreate the report? If the latter, it's important to keep in mind that Excel won't display the criteria selected, so unless you were looking at the actual same report in T-Stats, it may be difficult to confirm the current version of the report is actually identical to the previous. If the money is identical, that would be odd for it to differ on ticket counts alone, but if the situation is that the dollars are close enough, but the rest isn't, then it may be either due to activity in Tessitura, or a difference in criteria.

In general if you have concerns about the results you see in T-Stats there are some recipes you can always use to compare T-Stats to standard Tessitura reports. For example for tickets, this is a good one. If you find that recipe does not match, then you'll want to have one of your consortium's designated support contacts open a ticket with the T-Stats support team so they can dig into it. If that recipe does match but your comparison to what you exported previously does not, then the difference is likely due to one of the possibilities I mentioned above.