Optimize Local and Global Decisions with Snowflake’s Geospatial Support

This article is written by Andrew Meyendorff and originally appeared on the Snowflake Blog here: https://www.snowflake.com/blog/optimize-local-and-global-decisions-with-snowflakes-geospatial-support/

 

Even in a global economy, businesses need a deep understanding of local markets. For example, marketing campaigns designed to attract buyers in a large metropolitan area won’t necessarily attract small-town customers. Noticing that buying patterns in one area are extending into a larger regional or nationwide trend can lead to decisions that increase profits. But accessing and analyzing a broad spectrum of geospatial data has been difficult and expensive.

That is changing. Snowflake is excited to announce a public preview for geospatial data support in Snowflake Cloud Data Platform. It is one of many new features that we announced June 2. If you missed the announcement, you can register to watch the recording on-demand.

This blog post provides an overview of the new geospatial support capabilities, its ecosystem, and a brief summary of using the technology.

OVERVIEW

Geospatial data provides a powerful tool for organizations. The insights from analyzing geospatial data can increase profits, reduce fraud, and eliminate risks. Here are some examples:

  • A real estate investment firm wants to understand the concentration of its properties in different regions and analyze risk to those properties based on weather patterns, flood zones, and more. Knowing that information will help it decide where to invest its limited funds.
  • A food delivery company wants to understand the total addressable market in a new zone they’re considering entering. Geospatial data can reveal local trends, market saturation, and potential growth opportunities.
  • A company with a mobile app wants to understand how user behavior differs based on a user’s location in order to send notifications or place ads that are more relevant to the user.
  • A fraud detection department at a bank wants to check whether a customer making their third ATM withdrawal in a day could possibly have traveled between those ATM locations in that time frame.

All of these use cases and more can now be natively handled in Snowflake.

GETTING STARTED

You can start working with geospatial data in Snowflake right now. It’s easy with the new data type called GEOGRAPHY and a suite of functions that work with it.

A GEOGRAPHY value represents a point or set of points, made up of longitude/latitude pairs, on a spherical representation of the Earth. A GEOGRAPHY column can include any of the following types of geospatial objects:

  • Point
  • MultiPoint
  • LineString
  • MultiLineString
  • Polygon
  • MultiPolygon
  • GeometryCollection
  • Feature
  • FeatureCollection

Geospatial data comes in a wide range of formats. Snowflake currently supports the following formats as input and output:

  • GeoJSON (default output format)
  • Well-Known Text (WKT) / Extended Well-Known Text (EWKT)
  • Well-Known Binary (WKB) / Extended Well-Known Binary (EWKB)

Tip: If you need to load or unload data using other formats that aren’t natively supported, see the Ecosystem section of this post. 

By loading geospatial data into a GEOGRAPHY column, you can simplify writing your geospatial queries and take better advantage of built-in performance optimizations.

Snowflake makes it easy to load geospatial data by automatically detecting the source format when you load it into a GEOGRAPHY column. It is also easy to return query results using the output format you want using either the GEOGRAPHY_OUTPUT_FORMAT session parameter or a set of standard formatting functions.

Over thirty functions are currently supported for working with the GEOGRAPHY type, and we will continue adding to that list based on customer demand. To share feedback about functions you’d like to see, please use the survey link at the end of this post.

ECOSYSTEM

Snowflake’s geospatial innovations are part of a broader ecosystem that includes Snowflake Data Marketplace and Snowflake’s Partner Network.

Snowflake Partner Network

Snowflake has a strong ecosystem of technology partners that can help you process and analyze the wealth of geospatial data that is now available to you.

Safe Software is one example and we are pleased to welcome them to our list of partners. Safe Software’s product, Feature Manipulation Engine (FME), is the data integration platform with the best support for spatial data. FME makes it easy to build custom workflows that move data between other systems and Snowflake without writing any code. It understands all of your data’s formats and spatial reference systems, and it can perform any necessary transformations before loading the data to or while unloading data from Snowflake. Check out more details and demos of the integration and sign up for a free trial of FME here.

And that’s just the beginning. We will continue working with other Snowflake technology partners to integrate with our geospatial capabilities for visualization, data integration, and any other possibilities that arise. If you’re a partner who is interested in integrating with Snowflake, check out the Snowflake Partner Network.

Snowflake Data Marketplace

One of the keys to getting complete geospatial data is the new Snowflake Data Marketplace. This is an exciting new collaboration feature where you can combine your data with data from other Snowflake customers, accessing a wealth of information that wasn’t available to you before. Now you can get deeper insights into the regions that are important to your business and make decisions based on data rather than guesswork. Snowflake Data Marketplace is already available to Snowflake customers, so you can take advantage of it now.

If you are a data provider with interesting data sets to share or a Snowflake user wanting to learn more about the data sets that are already available, check out this link for more information about Snowflake Data Marketplace.