Good maps lie at the heart of any location application. When analyzing data on maps, proper design can be the difference between interesting and insightful.
But effective design is easier said than done.
Choosing which attributes to symbolize, how many attributes to include, and what kind of thematic map type to use can be challenging. Add in data distributions, classification methods, and appropriate methods of symbology and things get even more confusing.
You might find yourself asking questions that only expert cartographers or GIS professionals could answer, like:
- Is my classification method right?
- When should I use a diverging color scheme vs a sequential one?
- Is there enough distinction between the colors on my map?
- What is the best way to symbolize multiple attributes?
All of these decisions impact the usability of the map.
Our research team at CARTO has been experimenting with ways to solve this challenge and help non-GIS experts optimize map design to improve their analysis. We’re excited to announce a new capability inside CARTO Builder called Auto-Style.
Auto-Style analyzes your data to determine and apply the best visualization type by default. It takes the guesswork out of thematic map design and turns one map into many. By default, it provides the optimal visualization automatically.
To illustrate the impact of proper data-driven map design, let’s look at a common map type - a choropleth map of US Census county-level data from our Data Observatory - to understand how median income varies between demographics.
The map below symbolizes all counties in the US equally. We have added CARTO Dashboard Widgets for each of our variables of interest. Interacting with these widgets updates the visualization to summarize the current view.
Dynamically explore variables
As a first step, Auto-Style can be used to explore each variable of interest. Doing this with traditional workflows requires going back to the drawing board for each view.
Behind the scenes, CARTO analyzes the distribution of the data in the current map view, classifies the data using the best method to detect the patterns, and then applies an appropriate color scheme (we call these CARTOColors).
For example, if we want to visualize the percent of population that is Hispanic or Latino in each county, we can specify the Percent Hispanic or Latino widget. The map automatically updates to a choropleth map and the bars of the histogram widget update to function as the legend.
Filter, re-analyze, and visualize
What if we want to compare the Hispanic or Latino populations only in Texas? If we filter the data to only show Texas, Auto-Style recalculates the distribution for those counties and re-applies the color scheme only as it relates to your selected criteria. This not only adjusts the map visually, but also quantifies it proportionally in the interface.
By filtering counties with at least 50% Hispanic or Latino population and visualizing the filtered results with median income, we begin to uncover patterns of income and demographics in Texas.
By combining multiple filters with Auto-Style, anyone can quickly understand and act on visual analysis hidden in their location data.
Try filtering various criteria to take a deeper look into demographic and economic distributions to discover new patterns and insight. For example, combining demographic filters and Auto-Style by median income exposes patterns that can be used in everything from retail planning to government service distribution.
SEE THE POWER OF LOCATION INTELLIGENCERequest a live demo
The Change for Map Users
As location data (and maps) becomes more pervasive in analytics for organizations of all kinds, proper visualization techniques become even more important in how data is understood and acted on.
Features like Auto-Style are reducing skill barriers and enabling a broader set of users to unlock the power of their location data.