Churn management for SMB without magic and data mining. Part I

Andy Dovgan's picture
Churn Management

Churn management practice is strongly associated with industries that usually work with large numbers of clients. Try searching in Google, Slideshare as well as YouTube and you will see descriptions, presentations and even books for telecoms (mostly mobile), insurance, banking etc. Nevertheless churn prevention is a significant issue for every single organization. You cannot ignore or set low priority for it no matter how big or small you are. I’m not going to describe all facts why it is so important and cite examples here from researches. Instead, let me summarize the most significant points:

  • Company has already invested in the existing customers;
  • Customer acquisition is usually more expensive than nurturing the current ones;

It reminded me a recent TFM&A event in London where lead nurturing tools were at the limelight. Is the battle for a new customer really tough? Unfortunately it seems that this fight will only become tougher in future.

So, without doubt existing customers are very important assets. While bearing in mind and appreciating the value of managing and preventing a churn, medium businesses experience lack of tools to design and execute such campaigns. Traditionally, enterprises have used complicated data mining solutions with prediction functionality (based on mathematical models). For lots of smaller companies this method is not affordable and takes too long time. Medium businesses expect to deliver actions rather than digging into mathematics. In fact as their customer base is not so large, they have more options to build personal relations with customers. Certainly tools should be different. So let’s have a look at how to simplify it.

Basically there are 2 ways how to ensure churn prevention. First is to deliver unique and exceptional customer experience (which by the way is #1 trend in today’s CRM environment). However if you are not Starbucks, Zappos or South-West Airlines you ought to figure out symptoms as well as internal factors which could cause a churn and act effectively at the right time with the right message leveraging all benefits of individual approach. Second option is to design and execute churn management campaigns.

Take a look at a high level model of such campaigns.

Churn Management Campaign Plan

A start point is to assess the current state. Firstly, attention should be paid to defining customer value rate. One question that is so easy and so sophisticated at the same time: what constitutes the real value of a customer to a particular organization? Frankly speaking, a lot of stuff. However during the early stages, companies could use simpler models like the traditional Customer Lifetime Value (measured with calculator). Their value could be ‘existing’ as well as potential (a customer with strong potential for future cooperation however with small current profit). My suggestion is to map all components and place into a simple formula.

Additional components should also be taken into consideration (ad hoc example: company’s client pays a little money although provides a lot of references being brand advocate).

Having done the preliminary work, it’s time to move on to asking customers theirs opinion. This is perhaps the most important stage. Idea is to match traditional Voice of Customer programs (technique to collect insight from customers) and surveys with analysing what will make an impact on the decision to shift to competitor. The question of formulating right questions and regularly circulating the questionnaire is out of scope of this post. Anyway, there is plenty of advice on the web (for example here).

Let’s emphasize some general issues here.

  1. The most common mistake is to use general questions and not specify them.
  2. It’s risky to ask directly what can cause a churn. So it is important to be careful in gathering answers using indirect questions.
  3. It is not an option to do it just once. Quality Assurance practice should be run on regular basis according to value of the customer. All historical events should be collected and analysed dynamically.

I think this information is more than enough for a start. Next time we will talk about churn factors and symptoms, and see some examples of how to sense the symptom and initiate countermeasures.

Want to know more? Want to visit webinar according to following topic?

Subscribe