JM, development coordinator at Neon Museum here for my weekly segmentation question. I post these in various community groups but thought this may be most relevant for database teams...
About half a dozen of our board members have purchased tickets to our Duck Duck Shed season. Why is that this segment only has a count of 2?
I have tried variations with the board constituency and the season in separate segments. But that seemed to show who *didn't* have ticket, not who did.
I also tried changing the operator, but I don't think that brought me closer to my goal. Hoping this community has some ideas!
From the last example, each (non-suppression) segment only contains accounts that weren't in the previous selection. I'd you want Board Member+ purchased tickets in season X you'll want those criteria in the same segment.
Thank you, problem solved.
I am also working on another Extraction where I am trying to find donors from last fiscal year (I used between and the dates) who are also donors a specific month this fiscal year (I can use run date plus one month since it's from today). I don't think 2 date fields can go in one segment, right? My issue here is that the results list is larger after creating 2 segments. I want the segment to be smaller, narrowed down after giving this FY.
What do you recommend?
Okay, very important: segments don't form a funnel, as it were, which I think seems to be what you're trying to do. Your restrictions in one segment don't restrict following segments. So segment A can be Customers who bought X and your next segment can be Customers who didn't buy X, and you'll get the full list for each. Suppression segments will remove customers from future consideration, and segments will also not include customers who have already matched a previous segment, but the criteria don't affect future segments.
So let's say 40000 people bought tickets in season A. You have 20 board members, 14 of which have bought tickets. If your segments are:
Board Members and bought tickets: 14
Bought tickets: 39986 (doesn't include the previously identified customers)
Board Members: 6 (doesn't include the previously selected Board Members)
Now, I'm pretty sure you can have multiple instances of the same criteria in a list/extraction segment (not at my desk), but you have to be able to use has/does not have, or be in a position where you are defining different date periods for the same connected item (i.e. a contribution). Since you probably need more than one criteria to define your contributions anyway, what I would do in this case is make at least one (dynamic!) list. So you could make a list of donors from the last fiscal year. Then in a second list, or your segment, you would say "contributions in this specific month" AND "In List XXXX".
You could also presumably define one of those things negatively ("does not have a contribution in this time period") and then use that as a suppression, but I think using a list is cleaner and less likely to trip you up.
My eyes do glaze over at Extraction documentation from the point of "Inclusion Segment Order" with the rainbow-colored people, and beyond it. But I think your explanation helps clarify the functionality.
Thank you, Gawain!
In case it helps for future, the best metaphor I've heard to explain how this functionality works is this:
Think of a sieve over a sieve over a sieve (etc) over a bucket. Each sieve is a segment.. The criteria determines what gets caught by this sieve. When you run the extraction, you're pouring sand into the topmost tier. Some granules get caught in the top. Others flow through to the next, but once something is caught, it's caught. It can't be in two places at once. If a given sieve (remember it's a segment in real life) is a suppression, you're putting that whole sieve to the side and not using its contents. Each sieve that's an inclusion segment, you take the contents and do what you want with them.
I didn't coin this, and am paraphrasing from memory, but in my experience this makes it "click" for lots of folks.
I've used a stream and waterfall analogy when doing my slide show about extractions
If every constituent is a fish swimming down a stream they'll either find a pond made for a quality they hold or move on down the stream
eg: the Board fish might get end up in top pond, donors in the next etc. till you reach then end. A fish can only appear in one pond (ie: there are no Schrodinger's fish in an extraction)
I used to use gravity coin sorter and/or marble size sorter, but nobody uses coins or marbles any more. I didn't find any fish sorters on tiktok, but here's a rhinestone sorter using the sieve method www.tiktok.com/.../7065159146504719663. Also found egg and fruit sorters by size and weight, but I think this might work.
this whole thread reminds me of a youtube video i got sucked into about potato harvests and how the potatoes kept falling through the sorter machine at various points depending on their size...
You're more than welcome, I hope that helps. As you can see, it's something we've all had to spend a lot of time getting our heads around!