Customer Lifecycle Management

What is customer lifecycle management?

For us, this is the most exciting opportunity for data analytics. Every business has customers, and every business either has lots of data about their customers, or has the ability to collect that data! Using proven statistical techniques we can acquire customers more cheaply, keep them longer, and make them more profitable. We do this by

customer acquisition - identifying where our desired prospects are, and targeting them precisely with the right message

customer profitability - we learn what existing customers want, and we offer it to them

customer retention - predicting what actions a customer is likely to take in the future, and targeting them precisely with the right message to get them to take the action we want them to take

We've achieved great results by following this tried and tested process. 

How can it help us as a business?

By applying data analytics to our interactions with prospects and customers, we can increase revenues and cut the costs of our marketing efforts - sometimes drastically.

We can identify the prospects that are most likely to buy. We can identify the prospects that, if they do buy, are likely to bring the most revenue and profit. Furthermore, we can identify the best channels on which to reach these prospects. We can also identify the price points and offers that are most likely to appeal to them and cause them to buy.

We can offer a better, more personalised customer experience to all existing customers, thereby increasing customer satisfaction and loyalty.

Statistical analysis gives us the power to pinpoint cross-selling and up-selling opportunities with existing customers.

Finally, we can predict when an existing customer is likely to become dissatisfied and leave us, and reach out to them at this exact moment with the precise offer that is most likely to cause them to stay.

How do we perform it?

So how do we make the right offer, at the right time, to the right people, via the right channel? Firstly, we segment existing and potential customers into different groups which have different characteristics and exhibit different behaviour. We can make intelligent inferences about what these groups think and feel. For example, you may group customers and prospects by some combination of demographic data, existing customer data, and data of prior interactions between that individual and your business.

Acquiring Customers

We can use our knowledge of our business domain, and knowledge of our existing customers to decide what customer group we want to pursue. Relatively simple analysis will tell us which customer groups are most likely to buy and which will provide the greatest revenue and profit over time. We can then use existing customer data to determine the best channel on which to approach them (phone, emails, mobile, targeted ads on internet or social media, etc). If we can't derive that from existing customer data we can simply experiment by carrying out A/B tests.

Optimise Customer Experience

By using our data on past and current customers we can optimise our customer service. We can predict our customer's behaviour, and thereby answer their questions and fulfil their needs before they are even aware of them! In the same way we can use data on past customer purchases to learn which products or services are typically purchased together. This informs us of opportunities for cross-selling and up-selling.

It is much more cost efficient to sell to existing customers when compared to acquiring new customers. Therefore activities like cross-selling and up-selling are hugely beneficial to your bottom line.

Reducing Churn

We can also use data analytics to reduce customer churn. In order to do this we use data of existing and past customers to map attributes and behaviours to outcomes.

E.g. if some of our past customers churned after X number of months, we can predict that very similar future customers will also churn after X months. We can dive into our data to see what offers were made to those past customers when they were about to churn, and what were the success rates of different offers. We can then choose the best offer to make to the future customer. If we don't have such offer data then we can simply experiment with different offers, and by measuring our success rates we can use that data to inform offers made to other future customers.

Conclusion

I hope that this has given you an idea of how data analytics can help you to optimise all stages of the customer's journey. After creating groupings of customers we can learn which group is the best target. We can then use statistics and A/B tests to learn which channel and message/offer are most likely to convert the target. Once the target becomes a customer, we can use the same methods to predict their behaviours and needs, and thus make them a loyal and profitable customer.