Top excuses in marketing for not using value based customer segmentation
IT is doing a major redesign project and they cannot help
BI / Analytics has no ressources to support us
Everybody agrees it is important, but now is not the right time
Isn’t it suprising how easily your most valuable customers turn into a second priority?
Here is a quick checklist on the maturity of your customer segmentation
- Do you have a monetary value (ideally lifetime value) attached to each customer in your customer database?
- Do you have an expectation on how often each customer usually transacts with you?
- Are you aware which customers have transacted recently with you or how it has been since the last purchase?
- Do you update and report these values on a regulars basis?
- Do have processes and automated workflow, which depend on these values or changes in these values?
Do not feel bad, if you do not check on all these questions. Most companies will not. However, once your most valuable customers start churning, you will be in deep trouble! There take
Ready to try value based customer segmentation? Here is how…
Upload you transactions data
Extract your transaction data from you database and reformat it to fit an uploadable CSV file. We limit the model to transactional data on purpose, as it is the easiest for most people to get a hold off.
Configure the RFM and CLV models
In a first step, define the number of clusters (we recommend 5 in case you have not other information). Then set the number of years for your customer lifetime valuation computation. If you have no other assumption, set the value to 5 years.
Third, set the weights for monetary, frequency, recency via the slider. Why? Our model segments your customers according to three dimensions: 1. How recent have they bought from you, 2. How frequently have they bought from you? 3. How much have they spend? By adjusting the slider you can assign how much you value each dimension. By default, we allocated 50% on recency, 25% on frequency, and 25% on monetary.
Explore the results and export data
Once the results are shown, try to get a feel for how the cluster look like, how much CLV value is tied up in each cluster, how many customers are assigned to each cluster. Export the results, match the results in your customer database, and cross match them with your current audiences in your marketing campaigns.
Our suggestion would be to fill in the form and get free access to the app and the sample data. Then you can explore yourself. Then show and discuss the use case & dashboard with your colleagues.
It is important to understand that this use case is just one example. There are many variations of this model, e.g. recency could also be replaced or complemented with interaction. This would mean however that you would need to include non-transaction data such as website visits or social data in your model.
Please feedback to us your comments and suggestions.