Customer Segmentation approach, your thoughts please


I was given a task as part of interview for a Data Analyst, i really appreciate your thoughts. I never worked as a Data Analyst before, just started to learn

The dataset (customer_seg.csv) contains the record ID and zip code for ~2000 UK
customers. Using this information, segment the dataset into the following tiers

  1. Airport
  2. T1 e.g. London
  3. T2 e.g. Birmingham, Manchester
  4. T3 e.g. Liverpool
  5. Urban area
  6. Rural area
    The segmentation is hierarchical, i.e. a zip code that is both rural and airport will be designated
    as airport.

Task: your approach to real-world segment the dataset

The data set contains only 2 columns

  1. RecordID (1,2,…)
  2. Postcode / Zip code (G1 1HE, BS13 9HZ, …)


You basically have to build the data set using the Zip code. Try to find a reference file where the information you want is available(like govt data on govt website) and you simply have to join the files and create the necessary output.


Thanks Vivek. Here i need my views on how i can approach for this kind of problem. Your tip will help me to enhance my ideas


You can use this as a reference file. Then build for each zip code the variables you need. The way you build the data is the way you can summarize.

Also google your way to understand what qualifies as those segments and build a logic.

Your goal is to classify each zip code under a category. You will have to know the logic to classify them.

Post that its a matter of simple pivot and graphs


Thanks Vivkek for your help. Unfortunately, they were not happy with our approach. However, i really appreciate your thoughts.