Mar 20, 2019

"Conservation" Walkthrough

Lead Data Scientist, Jack Schroder, gives an in-depth breakdown of the "Conservation" segment. Understand each aspect of the data and correlations as he walks through the segment section by section.





Audio Transcript

Hey everyone, I’m Jack Schroder, the lead data scientist here at Spatial AI. We take location-based social media and create customer segmentation. Today, we’ve got a dual holiday: the Spring Equinox and International Earth Day so we’re going to take you through segment EH05, aka “Conservation.”



Here’s the summary page for our “Conservation” segment. This segment is driven by the kinds of people who advocate for recycling and hate wasting anything. Looking at the types of social media posts behind this segment, you see topics like recycling, solar, sustainability, and many more. Perhaps you do this yourself or have a friend engaged in this segment.


Getting a quick view of this segment across the country, it’s nice to see this segment appearing strongest in places like California and Maine, places known for their pro-environmental spirit.


Download Sample Data


Characteristics of High-Scoring Neighborhoods

Let’s move on to characteristics of high scoring neighborhoods.


We can see that this segment tends to be found in areas where people are typically younger and is especially prevalent in areas with lots of college-aged people. Income tends to be higher than the national median and areas don’t tend to have high diversity.


Continuing on, this segment is well educated, and residents of these areas often work in Health Care and Education. This segment is slightly more common in urban areas than rural ones, home ownership matches the national rate, and voters from these areas showed strong liberal support in the 2016 election.


One of the coolest things about these Geosocial segments is the story you can tell with them, and how that story makes a lot of sense. All of the characteristics we just heard, I think, make a lot of sense to the average person and we’ve been able to quantify that these are statistically significant relationships. We’ve also got these “additional insights.” It’s interesting that the “conservation” segment shows a strong tendency to not be religious and also is correlated with newer buildings.



In the time section, you can get a sense what time during the day, week, and year this segment has peaks and valleys. Unsurprisingly, this segment scores higher in the summer and, perhaps surprisingly, this segment is much more prevalent Monday through Thursday than it is on the weekend.


Retail & Restaurant Impact

Finally, we come to the retail and restaurant impact section, where we can see some interesting retail correlations here with clear business implications.


If you are a southern food, fast food brand or department store you will want to steer clear of these areas. The demographics might give you a thumbs up but the community will not readily accept these concepts.


You can look at this from the other way as well. Let’s say you are a community planner looking to create an environmentally friendly community. You should consider leaving open space for parks, schools, libraries, and incentivise non-profits.


That’s it for our “Conservation” segment. But that’s just one of 72 Geosocial segments that you can explore in our taxonomy. Check it out at and download some sample data yourself! Thanks for watching, and Happy International Earth Day!  


Download Sample Data

Get the latest from in your inbox
A weekly summary of news, product updates, and special offers.