Hello Tessitura community,
I am just starting to use Analytics to produce line graphs tracking our our marketing campaigns per production, modelling from the Production Sales default dashboards.
Does anyone know if it's possible to add a target line into these graphs? I imagine we might have to use campaign goals in order to do this...
Also, how can I transform a regular line graph to a cumulative one?
Thanks!
Krista
Hi Krista,
There was a bit of work on this kind of output in this post:
(+) Rolling Sales as they approach budget goal - Reporting & Analytics - Forums - Tessitura Network
The original post called for Performance Budget as the target. The challenge we ran into was that not all performances were represented in any given week, so we had to filter the "Target" formula to look at a specific week in which all performances have data, and then render the value from that week across all weeks.
If you use the Campaign Goal Amount instead of Performance Budget Amount, then the formula will change slightly to be more like:
MAX ( [Order Weeks Prior to Performance] , SUM ( [Campaign ID] , [Campaign Goal Amount] ))
Where the [Order Weeks Prior to Performance] (or whichever relative milestone anchor you've chosen) is set as a Filter, and using a Values type filter, set it "= 0".
As for the cumulative value, you might derive this from that other post, but check out the Running Sums section of this webinar:
Take Your Tessitura Analytics Skills to the Next Level (tessituranetwork.com)
Best,Chris
I had a follow-up thought. The challenge in the linked post is that there's a potentially different Performance Budget Amount for every performance, and they need to be summed up. If you're limiting your results to a single production season, and all performances in that production season are tied to the same campaign, then that campaign IS represented in every week of you relative milestone. And given that, you can probably just use
MAX ( [Campaign Budget Amount] )
as your target formula.
Hi Chris,
The reference to the previous post was informative (starting to see the logic in the setup...) but I should have specified that we are looking for the # quantity of tickets paid target rather than the $ budget target. Currently we don't input this data into our ticketing builds, but I wonder if this needs to be inputted somewhere or whether it could pull from the number of seats made in the build?
At this stage we are looking at the individual performance level, and each performance will be under its own widget.
Let me know if you have any additional tips for this.
Thank you!
It the number of seats in the facility, or the number of seats of a given seat type in the facility, is an effective target, then something like these could work:
( SUM ([Seat Count]) , ALL ([Weeks Prior to Performance]) )
or
( SUM ([Seat Count]) , [Seat Type Filter] , ALL ([Weeks Prior to Performance]) )
When grouping that Seats and Tickets data by Weeks Prior to Performance, a simple SUM ([Seat Count]) will only return seats associated with the given week. The ALL() function will break out of the [Weeks...] grouping and return the sum across all weeks.
If a capacity count like above isn't quite right for your target, there are 2 other options. The first is hardcoding a value into the widget. Simply "Add" and new Value, click the fx button, and type the raw number you want as the target, and click OK. If however, you need something more dynamic that will allow you to enter targets in Tessitura (e.g. via Inventory Content field or a special System Table), and have them change in your dashboards to automatically reflect the performances on which you've filtered, then I can put you in touch with someone on our consulting team to quote and implement that small customization.
Cheers,Chris
With the raw value, sometimes we have a target number that is based on a calculation made by finance and it looks like this
The blue line is Rsum([Total Ticket Paid Count])
The orange line is just the target number in the formula field as Chris says
Orange line
Thanks Chris Wallingford and Heath Wilder.
The running sum function seems to be producing a flat line at 0 for me.
I've replaced DUPCOUNT ([Ticket Paid Count]) (week-on-week graph below) to RSUM ([Total Ticket Paid Count]) - so the data filters are the same, but it doesn't show anything with a running sum function.
And using RSUM ([# of Ticket Paid Count]) produces a shape, but is overreporting by about 500 tickets?!
Is there anything I should be adding or removing in the data filters, or checking in the formulae? Is the Weeks in Date criteria an issue for the X-axis dates (noting previous examples have used the relative 'Order Days to...')?
That your Ticket Count is coming out 0 might require a support ticket.
Only price layers in the Ticket Price price category are counted for this field. If multiple Ticket Price price layers exist for a performance, the Ticket Price price layer with the min start date and min price layer type rank is counted. If you break out counts by price layer, layers in all other categories will have a count of 0.
Using DUPCOUNT( [Ticket Paid Count] ), which is also shown as [# of Ticket Paid Count], we're just getting back a count of rows such that if a ticket has multiple price layers, it is counting up all those layers If you need a workaround to 0 ticket counts, you can instead count distinct SLI ID. We'll need to filter out unsold seats which can be done with either:
Then add a value for a running sum of unique SLI ID, which could like one of these two: