Geospatial Match - Need Intersect to ID Zip Within Place / DMA / CBSA

[RickPack]

It looks to me like one needs to use "Target Intersects Universe" of the Spatial Match tool in order to identify zip codes in a larger geographic space (Place, DMA, CBSA, etc.), and not "Target Contains Universe" or "Target Within Universe". 

The Raleigh "Place" from the Allocate Input tool is an Incorporated Place (thanks, Mandy Gray @mandycgray) according to the Census. Allocate indicates its Key value as 37550000 (a concatenation of the place’s State FIPS + Place FIPS) and that is "Raleigh City" according to https://www2.census.gov/programs-surveys/popest/geographies/2019/all-geocodes-v2019.xlsx. The Place FIPS is not "00000" for "Incorporated Places".

I think the spaces within the Place boundaries causes only "Intersects" to work (compare Raleigh_Place.png to Zip_27612.png - both from the Browse tool). This is probably only true for some Places.

One can see the results of the attached Spatial_Match_InsersectNeeded.yxmd for an example using Raleigh, NC zip code 27612. You need the Business Insights data package for this to run without changing the source for the Allocate tools (currently Experian US - Most Recent Vintage), and need Business Insights or Location Insights to have Allocate tools.

RALEIGH PLACE

RALEIGH, NC ZIP 27612

[CharlieS · Accepted solution]

I'm not sure exactly what you're looking for here, but I'll chime in with a few thoughts for anyone else reading:

- The documentation on the Spatial Match tool has a handy set of images for each match type: https://help.alteryx.com/current/designer/spatial-match-tool 

- It seems like your goal is to produce a list of ZIPs for each geography (place, DMA, CBSA). In  this case, you're definitely going to need to use "Intersects" as the spatial match method. ZIPs as polygons are a hot mess and don't consistently share boundaries with any other geography. Between that and the other situations you listed above, "Intersects" is the choice here.

- For this situation, it works both ways, but I would recommend placing your geography of interest as your "Target" and the list of ZIPs as the "Universe" input. This organization is often beneficial for subsequent data aggregation, but also for the reason below. 

- For speed purposes, I would highly recommend saving off the complete list of ZIP polygons from the Allocate Input as a .yxdb file. Reference this .yxdb file as the universe "Use Records from File of Database" configuration inside the Spatial Match tool. When you use a .yxdb here, the Spatial Match tool can use the spatial index that's included in .yxdb file spatial objects. In short, it only loads what is necessary to the scenario (instead of loading ever ZIP in the country, then matching).

I'm always happy to talk spatial. 

[RickPack]

Thank you, @CharlieS! I neglected to communicate an interest in having someone confirm my thoughts. I also hoped to help others who might have invested time like I did in using Spatial Match in correctly to find all the Zips within a Place, DMA, or CBSA.

[mandycgray]

Very cool, Rick!  Its been a pleasure working with you.