ArcGIS Community Analyst

Micro-Targeting with the Updated Tapestry Segmentation Data on the Community Analyst API (Part I of II)

Some Tapestry Segments

Community Analyst Data Release

You heard last week that you can access new data through Community Analyst.

Well, in celebration of this Community Analyst data release, I’m going to talk about accessing one of my favorite datasets through the Community Analyst APIs—the Esri Tapestry Segmentation  dataset—and discuss how and why it should be leveraged by organizations large and small.

Before we get into the nitty gritty, let me describe a real life use case of Tapestry Segmentation.  I’m confident that this use case will inspire your own!

Charity Outreach

OK. So, what is this Esri Tapestry Segmentation stuff and what makes it so special that a geek like me gets excited?

Well, let me use the example of a charitable organization.

Many charitable organizations rely on individual donors to sustain them.  Many of these organizations are constantly under pressure to identify potential donors and solicit donations from them to keep their bills paid and pass on value and consideration to the causes they are advocating.

Well, this can be a delicate balance.  With increased public scrutiny, some charitable organizations are under fire for what many perceive as spending too much time and resources on this marketing aspect and not enough on the actual charitable work and/or direct aid to their causes.  How can they responsibly balance these competing priorities?

One obvious solution is to really know who their donors are and know where to find more of them.  So what does this really mean?

Esri Tapestry Segmentation – Outstanding Enterprise-grade Data for Profiling People, Communities, Areas, and Locations 

If they have a good idea of the profile of their “best” and “most reliable” donors—their “donor base”—then, they can more effectively seek them out in outreach efforts while greatly minimizing the chances of needlessly spending valuable time and resources on fruitless efforts.  I guess what I’m really saying here on a meta-level is that, the most responsible and effective organizations thoughtfully coordinate their efforts, decisions, and strategies based on reliable and actionable intelligence rather than on whims, feelings, or “knee jerk” reactions—because, of course, having a better picture, model, or view of reality significantly increases your chances of success!

OK.  Let me put it bluntly.  I’m not a professional marketer, statistician, demographer, or data miner (even though I work with a lot of them)—I’m actually a GIS geek with enough of a technical background to make me slightly “dangerous.”  With Tapestry Segmentation, you can be all of that and more with relatively little effort.  Just imagine leveraging advanced market segmentation methodologies which were once only accessibly to the largest and most lucrative organizations who are able to hire these crazy people and crunch a lot of (expensive) data—and, no, you don’t need a marketing MBA to understand it either!

Esri Tapestry Segmentation – What is It?

The dataset behind Tapestry Segmentation “classifies” every area into one of 65 household categories or segments.  Each segment represents a set of dominant characteristics among the households within the area.  These characteristics can represent a broad base of attributes including demographic, lifestyle, spending, interests, and behavioral traits.  From my perspective, the individual attributes themselves are not so important here (although, the API supports querying/analyzing these individual attributes in other tasks).  The value and “wow” factor behind Tapestry Segmentation is the fact that, the segments effectively describe  households which share a common set of characteristics (almost like a “cluster”).

Why is this amazingly powerful?  Well, let’s go back to the example of charitable organizations.  What if, based on prior knowledge or records, they have a list of areas, locations, or addresses which represent the greatest donor activity?  Using Tapestry Segmentation, the charity can take this list of their “best and most reliable” donors and classify them into one of the 65 Tapestry segments.  This classified list may contain several different Tapestry Segments at which point, the charity can take the “dominant” tapestry (the Tapestry Segment which describes most of the analyzed areas, locations, or addresses) or the “top” tapestries (which often is described as the top three) and have a good indication of who these households are and………….where they can find more.

But, before I discuss finding more of these households which represent your charity’s lifeblood, I should note that you can now look at some information associated with the dominant tapestries of your donors which may give you an general idea of who they are and what are some traits, behaviors, and characteristics which are common between them.  To do this, you can leverage these resources:

OK. Now, you want to be able to find more of these households—the households which have, according to your experiences, been the most generous in helping to sustain your organization’s charity work.

There are many options here and one that I commonly come across is to, once again, leverage the Tapestry Segmentation data to identify areas for marketing or outreach.  This may be done on a large scale like cities, Designated Market Areas (DMAs)—which are media markets, Congressional Districts, and Metro Areas; or at a smaller or more “local” scale like ZIP Codes, Census Tracts, and Census Block  Groups.

Great!  So, say you have the tiniest of outreach budgets and you want to micro target Census Block Groups (which may have as few as a few hundred households) which share the same dominant Tapestry Segment as the majority of your best donors.  This would be better than making a large investment in ”blindly blanketing” a bunch of random households throughout a wider area with solicitations for donations.  With the  location intelligence obtained through Tapestry Segmentation, you can significantly increase the chances of reaching your donor base (or market, constituency, etc.) .

Esri Tapestry Segmentation – Through the Community Analyst API

Well, conceptually, the two popular ways to leverage Esri Tapestry Segmentation through the API are through reporting and through “data enrichment” of  study areas.

(A1). Through Reporting, Tapestry Segmentation summarizes information about study areas in reports.  With the latest data release, here are the newest Tapestry Segmentation “Reports”:

(A2). Through “Data Enrichment,” Tapestry Segmentation allows developers to append or associate additional attributes with input study areas.  With the latest data release, here are some of the analysis variables that can be specified to enrich an existing dataset:

These study areas can consist of the following:

Today, I’m just going to go over the “data enrichment” option in our charitable organization use case.  Here’s a quick overview and summary of what I’m going to do for them:

(B1). The charity has provided me with a list of their  current, most generous, and sustaining donors based on their contribution history.  They believe in the charity’s  work and they feel  passionately about its cause.   This list contains hundreds of contributors, maybe more and their addresses (The more “observations” you have —the better the view of reality!).  I’m going to analyze this list by leveraging Tapestry Segmentation.  I’m going to determine the top three Tapestry Segments that these prime donors fall into after tabulating their dominant Tapestry Segments associated with their household addresses.

(B2). I have also been provided a unreasonably large list of potential donors and their addresses.  I need to further qualify and “filter” this list because the charity doesn’t have the budget or resources to send out brochures and solicitations to each of them.  I need to help increase the charity’s chances of successfully reaching their donor base while preserving their existing donors’ contributions! 

(B3). The charity has decided to only send out mailings to ZIP Codes which have a dominant Tapestry Segment that matches one of the top three Tapestry Segments of the current list of the most generous donors.

And, here’s the “recipe” using the REST API (Remember, there are many ways to leverage Tapestry Segmentation data.  This is only one of them):

(C1). Geocode the addresses to get coordinates for each record in your list.

You can use this Geocoding service on ArcGIS Online .

Here’s a sample of geocoding a single  address:,Redlands,CA,92373&f=json

(C2). Generate a “token” with your Community Analyst API subscription with the Get Token service .

This token authenticates and validates your subscription credentials.  To leverage Tapestry Segmentation, you need a Community Analyst Standard or “better” subscription.<YOUR_USERNAME>&password=<YOUR_PASSWORD>&f=json

{"results": {"token": "pWFsqkpedjcCOL8Pn_EjIfDv-m6kfyZPGAx_M4T2KGeO-ZSF2K4lmA2yyjVm-FwG10U="}}

(C3). Determine the dominant Tapestry Segment of each location using the coordinates with the Smart Map Facts (AKA Thematic Query) service of the API.

Here’s a sample of determining the Tapestry Segments of 18 locations (We’re actually querying the Tapestry Segment IDs of the Census Block Groups that are associated with these locations.  Census Block Groups can have as few as a couple of hundred households—sometimes even lower than that so this is fairly “high resolution data.”)

(The token value in the sample will need to be replaced as it is expired.)

In the partial example response below, the IDs represent the Census Block Group FIPS code while the DOMTAP represents the dominant Tapestry Segment code.,ID,DOMTAP&geometry={“points”:[[-97.0990969,32.766428],[-96.8154520,32.8203959],[-96.7720944,32.8460005],[-96.7687214,32.950617],[-96.8034808,32.908863],[-96.7520755,32.8156843],[-96.7951148,33.0292412],[-96.6584883,33.130008],[-96.616677,33.2159190],[-97.7431824,30.3885401],[-98.4797976,29.4931635],[-95.6224025,29.5992683],[-95.5421063,29.6419595],[-95.5726391,29.7365877],[-95.4986994,29.754823],[-95.4416734,29.7068180],[-95.4181112,29.7391306],[-97.7527695,30.2707094]],”spatialReference”:{“wkid”:4326}}&geometryType=esriGeometryMultipoint&Token=ZxxJnedxCb49t7QZBW7cOBvt6w6gATimMDhcbd4JwGkc5SrVrG3L682hx6Y2&ActiveDatasetID=USACensus2010

(C4). Tally and sort the Tapestry Segments to determine the top three segments for all of your households.

Here are actual ranked results from the previous step :

5 households – Segment 27:“Metro Renters”  (Click to view PDF Description)

3 households – Segment 8:“Laptops & Lattes”  (Click to view PDF Description)

2 households – Segment 1:“Top Rung” (Click to view PDF Description)

2 households – Segment 39:“Young & Restless”
(This is actually “tied” with Segment 1.  In case of tie, take the lower segment number when ranking.)

1 household – Segment 4:”Boomburbs”

1 household – Segment 13:”In Style”

1 household – Segment 16:”Enterprising Professionals”

1 household – Segment 19:”Milk & Cookies”

1 household  – Segment 22:”Metropolitans”

1 household  -Segment 38:”Industrious Urban Fringe”

In general, we can conclude from the top three Tapestry Segment descriptions that the areas contain upscale singles who are fairly affluent, educated, and stylish mostly in their 30s and 40s without the financial burden of a mortgage or family.  There are also some very wealthy married couple households in some of the areas.

I now have the Tapestry Segments of the charity’s  most generous and reliable donors.  Let’s go find more of them! (By the way, I will let you in on  a secret about the locations I used in this sample in the next post.)

Today, I determined the top three Tapestry Segments of a charity’s most generous donors in order to understand who they are.  Next time, armed with this intelligence, I’m going  to show  you how to find where they are so the charity can coordinate and optimize their outreach strategy to find more of them.  Continue to Part II


“features” : [
“attributes” : {
“ID” : “484391131111″,
“NAME” : “484391131.111″,
“DOMTAP” : 39
“attributes” : {
“ID” : “481130006033″,
“NAME” : “481130006.033″,
“DOMTAP” : 27
“attributes” : {
“ID” : “481130133001″,
“NAME” : “481130133.001″,
“DOMTAP” : 1
“attributes” : {
“ID” : “481130001002″,
“NAME” : “481130001.002″,
“DOMTAP” : 22

About the author

Tony Howser

Tony is a product manager on the Data & Location Services team and is focused on Esri's GeoEnrichment and Places offerings. His primary goal is to empower users and developers with valuable location-based context to support their mapping, analysis, and decision support needs.

Next Article

Engaging Volunteers for a Cause

Read this article