Customer insights are essential to any marketing strategy. Do you know who your target audience is? Where do your customers live? What is their income? What kind of house do they live in (both age and value) and do they have children? What is their level of education? What are they buying and how do their demographics differ by product line? Answering these kinds of questions is made possible by demographic customer segmentation.
Knowing who your customers are is very important. If you know who your customers are (age, income, family composition), you know where you can find them. Once you know where to find them, you can develop a more targeted, data-driven approach to marketing to these audiences. Many different media platforms have the ability to target audiences at the campaign level. Demographic audience segmentation can help you reach new audiences, while behavioral-based audience segmentation has the advantage of being more prone to existing customers (shopping habits/buying patterns) and better suited for retargeting.
There are many ways to perform demographic audience segmentation. A common approach is to have your customers answer the survey response questions. Perhaps you could incentivize your customers through a customer loyalty program to complete surveys about their age, gender, income, or other areas of interest, in return for a discount. on your next purchase. Another option is to match existing customer address information with publicly available census data.
The advantage of this approach is that you only need the customer’s address. Getting existing customers to hand over information about their personal lives can be difficult, and in many cases businesses have a lot of information at hand via their billing and shipping addresses. waiting for Customer addresses can be easily traced to different geographic levels, from DMA/CBSA to Census Block Group. Census block groups are as close as possible to neighborhood level insights. For example, as of 2019, the United States can be divided into 217,740 block groups He and 41,692 zip codes her. This means that block groups are approximately five times the size of a postal code.
of American Community Survey (ACS) publishes data on each of these block groups annually. ACS provides many estimates of age, income, education, family composition, and more. By analyzing the composition and composition of the regions in which our customers live, we can infer the composition and composition of our customer base.
real example
To illustrate this approach, Kaggle sales product dataset, and geocode a sample of 15,000 unique purchase addresses to latitude and longitude. As you can see in the chart below, these customers live primarily in metropolitan areas in the United States.
If you have the latitude and longitude of your customers, you can identify the census-defined block group in which these customers live. In the example below, the census block groups residing in the latitude and longitude of Los Angeles, California are overlaid. of.
Using the ACS API, here are some of them: many Wrong population estimate You can get information about the block group in which you live. Some of these are listed below.
- median household income
- HH with presence under the age of 18
- Median HH value
- Population by age
- Household Income by Bucket
- HH value
- Owner Occupancy HH by Building Age
There are many ways to subdivide a demographic dataset. One seemingly obvious solution was to compare electronics type breakouts by metrics. The demo dataset includes two types of phones (Google Phone and iPhone), two types of laptops (ThinkPad laptop and MacBook Pro laptop), and three types of headphones (Bose SoundSport Headphones, Wired Headphones, and Apple Airpods Headphones). I have. ). An equally obvious question is whether there are statistically significant differences in any of the above demographic characteristics for electronic device breakouts.
To answer this question, one-way analysis of variance (ANOVA) can be used to test whether there are statistically significant differences in population means for different breakouts. If so, what insight can we gain from these comparisons? debt value and ap value.of debt Values are defined as:
F=MST/MSE
Where, debtis the ANOVA coefficient, MST is the sum of mean squares due to treatment, and MSE is the sum of mean squares due to error. Simply put, this formula can be rewritten as:
F = mean squared processing/mean squared error
This is how we determine the variability between samples. If there is no statistically significant difference between the two samples (also called the null hypothesis), debt The scores converge to 1. debt The higher the value, the greater the sample-to-sample variation.In response to the debt value, and the number of degrees of freedom, the corresponding p The value exists, so we finally know if the sample is statistically significant.a p A value less than .05 indicates that the population difference is statistically significant.
Using this approach, we can identify one statistically significant insight for phones and a few more for laptops. Some of these results are shown below. Note that the results are from the sample dataset. Our intention is not to extract a literal interpretation of the results for this particular sample dataset, but rather to present a technique that demonstrates how to mine customer addresses for demographic insights.
| indicator | debt | p | Occur | the subset with the largest mean |
| Population 65+ | 3.96 | .0467 | phone | iphone |
| median household income | 17.9 | 2.465e-5 | Laptop | ThinkPads Laptop |
| household income 200k+ | 9.11 | .00259 | Laptop | ThinkPad Laptop |
| Population 18-24 years old | 8.32 | .0039 | Laptop | ThinkPad Laptop |
| Population 35 – 49 | 9.86 | .0017 | Laptop | ThinkPad Laptop |
| Population 50 – 64 | 7.03 | .008 | Laptop | macbook pro |
| Population 65+ | 9.03 | .0027 | Laptop | macbook pro |
For a given index, debt When p Identify the subset of data with the highest mean, knowing that the difference is statistically significant. A literal interpretation of the first line would be “among people who bought a phone, those who bought an iPhone lived in the block group with the highest percentage of people over 65.” As I said before, one of his assumptions we make is that the 65+ population tends to prefer iPhones over other types of phones.
find the right audience
Findings like the one above can have a significant impact on your marketing strategy, both at the macro and micro level. If you know where your potential customers live based on overestimated populations, you can develop targeting strategies across both digital and traditional channels to acquire new customers. This is the biggest motivation for performing demographic audience segmentation. Famous words of 19th century retailer John Wanamaker: “Half the money I spend on advertising is wasted. The problem is I don’t know which half.In the 21st century, using a targeted, data-driven marketing approach using demographic audience segmentation can significantly reduce wasted marketing spend. We utilize demographic insights to help our clients target their optimal audiences.
