CLS Customer Segmentation Models
Brands and Product Offers need to be positioned to resonate with their target audiences.
South Africa is a very broad base to extract insights from and therefore needs to be segmented into clusters with "dimensions" that enable actionable insights to support strategic and tactical business requirements.
The COVID pandemic has altered the economic landscape significantly along with POPIA constraining electronic messaging for marketing purposes.
Previous learning's, internal or external, may no longer hold true in today’s economy and therefore a national framework underpinning new learning's and results are critical to enable competitive Advertising and Marketing where efficiency improvements are core to being successful.
The core challenges facing many Insurance Companies is worsening Retention rates and lower Acquisition Leads volumes which lead to lower revenue and profits.
Buying in leads is expensive. Simply taking any prospect lead that is not on your customer database and pushing them to your Call Centre sounds worthwhile - but not all potential customers are equal and many result in churn.
Losing customers is expensive. Are there patterns in your customer data that when overlaid on external data models will produce actionable information and help alleviate attrition and inform acquisition strategies? Yes - but do you have the data to identify the key patterns and variables?
The CLS Solution
The Customer Life Stage (CLS) solution to help solve the above three challenges is premised on 3 value adding areas:
Improving Customer Data Quality structure and accuracy - using incomplete/inaccurate data will bias the results
Segmenting the customer data statistically into clusters that match external prospects to internal customer profiles
Identifying and using the correct channels for marketing and advertising to engage with the consumers.
Overview of our EI CLS
A Customer Life Stage model is used to determine how many groups of people have similar characteristics that can be grouped together - or not. Once you have grouped "like" people together you can then quantify and analyse them to determine the characteristics of your most profitable customers and then analyse and quantify the groups to see where the most likely opportunities lie.
Essentially we overlay a clients customer data on the EI national data set after applying data enhancement and then using statistical methods and our Machine Learning algorithms, we can then build your CLS view and advise and assist with whatever objectives you are seeking to achieve.
The difference between Machine Learning (ML) and Artificial Intelligence (AI), is essentially that ML learns and predicts based on passive observations, whereas AI implies an agent interacting (online as an example in a chatbot) with the environment to learn and take actions that maximize its chance of successfully achieving its goals. Our EI Machine Learning algorithms takes customer data and turn it into actionable insights through accurate prediction of customer behaviour which enables specific targeted marketing to individuals. Machine Learning becomes more powerful with larger detailed data sets to learn from and at EI we have built very large robust training data sets for the ML optimisations. This is a big differentiator to other companies and contributes to our success.
We have used these statistical models very successfully on many types of customer data - Bank services, Credit cards, different types of Insurance, Loans, etc.
Once the Customer data has been through all the transformation/update processes then we are ready to apply the CLS base cluster model and all pertinent algorithms. As can be seen from the "1 - CLS - Base" slide below, this enables us to classify all the customer data into appropriate Age and Financial Status groups as the base level. This can then be viewed against the national base to see where your customers are predominant and where you are better represented or worse represented. This forms the first base level to assess overall life stages across the total SA landscape.
Once the overall position has been assessed then we can drill down to deeper more granular levels. This level enables us to establish analytically a hierarchy based on assets ownership and Income along with other variables. This is further enhanced with Persona to aid matching customers to advertising media and locations. These two enable one to asses the potential product take-up by group in order to focus on the low hanging fruit as well as identify new opportunities.
The next level, the Technology Adoption layer is used to indicate a number of aspects, depending on the product/offer being assessed. It is also a good indicator of communication channel preference.
We also can model your data from scratch to build custom cluster models if you have an internal model built on a qualitative model with a sample universe that you wish to roll-out, but don't have the necessary variables to categorize the rest of your database.
There is a lot more that we can discuss in terms of the impact that a CLS model can offer that will assist with making your advertising and marketing more efficient, more effective and assist with growing your customer base and revenue.
However we need to understand your products and objective and then can suggest the optimum way forward.
Finally - we also can model your data from scratch to build custom cluster models if you have an internal model built on a qualitative model with a sample universe that you wish to roll-out, but don't have the necessary variables to categorize the rest of your database.
We do have deep experience in all statistical methods and using our data along with yours can combine and use the most appropriate methods to deliver the optimum modelling results delivering your objectives.