The DataHero Blog

Using Census Data in Marketing

May 20th, 2014


The US Census isn’t just useful for news outlets to tell you how many people live in an area and their demographics. It provides an incredible amount of demographic information that has real business applications, like income, population, ethnicity, etc. The problem with census data is that it isn’t combined with the information we care about in business; customer information. Thanks to DataHero, you can combine any dataset (like a Census dataset) with another dataset (uploaded Excel or CSV files to SaaS data). Thus, we can leverage the information in the census with business specific data, to allow us to better target our marketing efforts to the right segments and increase your ROI. In this post, I’ll walk you through how DataHero enables you to combine US Census data with real business datasets.

In this example I’ll show how to combine census data with customer purchase data, a great resource for customer information. There are many places where you can download census data in CSV or Excel format. In this example I’ll be using this dataset. Also note, customer purchase data can be substituted for any dataset that contains relevant customer information, like email marketing data or customer surveys.

Combining two data sets:

1. To begin, import the census data.

2. Navigate to the specific file then click “Combine Your Data” in the left column. Click “Combine With Another Dataset”.

3.  DataHero then suggests keys that you can combine on, in this case select zip code.

4. Select the next dataset. Click on Your Data in the left column, then select the payment data that you’ve uploaded from your desktop.

5. Drag the common key (zip code) onto the drag and drop area.

6. Check the total records in each dataset and verify the combined number is what you would expect.

7. Click Combine and retitle the new dataset.

Now the groundwork is laid for data analysis with the correct reports, and we can start making the charts that leverage the census data. The charting process is very similar to the charting process with an un-combined dataset. You can choose to use one of DataHero’s suggested charts, or create a new chart.

In this example we’ll create a new chart so I can outline exactly which attributes to add to each chart to discover customer demographics by location.

Where are customers spending the most money?

From the Combined dataset page click “Create New Chart”

Drag on the following attributes:

  • Zip code
  • Purchase amount

Once you drag these two attributes onto the chart canvas, DataHero will automatically create the following map:

In the chart above we can see how purchase amounts are distributed by zip code. It seems that California, New York, Pennsylvania and Florida are the top states for purchases of this product.  Now we can start taking a look at demographics of each state to compare how our users match up with generalized demographics.

How does income compare across zip codes?

From the Combined dataset page click “Create New Chart”

Drag on the following attributes:

  • Zip Code
  • Adjusted Gross Income

DataHero then creates the map below, highlighting average income by zip code (or state).

It appears that the top locations for purchases align with top locations for income. To double check this, we can further investigate the relationship between purchase amount and income with a scatter plot.

How is overall income by zip code related to purchase amount by zip code?

Drag on the following attributes:

  • Zip Code
  • Purchase Amount

Click “Type” and change the chart type to scatter plot

You’ll see the scatter plot below:

There is a positive correlation between purchase amount and adjusted income. This means that as income increases, purchase amount generally increases as well. Thus, we can look at the zip codes where the purchase amounts are highest and discover more about the demographics in these areas.

What are the demographics of the areas with the highest purchase amounts? 

Drag on the following attributes:

  • Zip Code (check only the top 4 zip codes by purchase amount)
  • Hispanic/Latino Population
  • White Population
  • African American Population
  • American Indian/Alaska Native Population
  • Asian Population
  • Native Hawaiian/Pacific Islander Population

DataHero will then display a bar graph by the top four zip codes for purchase amount:

We now have more insight into the demographics of our top-selling zip codes. This means we can target our marketing better here, and perhaps use similar tactics in zip codes that have not reached their sales potential.

By combining data from multiple sources in DataHero, we can now make better data-driven decisions by enhancing the business’ specific data with US Census data.   Get the answers you need  today, by merging datasets more easily than ever before. Sign up for a DataHero account and give it a try with your data today.

Create My DataHero Account

By Kelli Simpson

Create my Free DataHero Account

Get the fastest, easiest way to understand your data today.