Have you ever stopped to consider how much it costs when you lose a customer? Does your company place a value on customer retention? How does your company stay ahead of the curve and what actions are you taking to keep customers longer and make them more profitable? In future installments, we’ll explore in much more detail how machine learning and artificial intelligence can help you get and stay ahead of the curve, but before we delve into that, we should first talk about the real cost of customer attrition. Without understanding the economics for your company, the journey to improve is more ambiguous and the goal of improved customer retention is harder to accomplish.
So, you may be asking, “What ARE the costs?!?!?!” Well, let’s start with the more obvious metrics most organizations use to measure the impact of attrition. Initially, a company will look at the loss of revenue, whether in recurring or non-recurring revenue models, to indicate whether trouble lies ahead. Because should that number increase without a subsequent increase in new business, there will be some even tougher questions to answer later. The first step in solving any customer churn problem is to understand the signals in your data and to do so, you have to determine which datasets are most important. Taking a data-driven approach is the best way to solve the problem. But this is just the top of the funnel when it comes to measuring the impact. There are several more layers to uncover to truly understand the significance. According to research conducted by Bain & Company, increasing customer retention rates by a mere five percent could increase profits by 25 to 95 percent, depending on the industry.
Ok, now we can see how much revenue we’re losing when customers leave (why are they leaving is a completely different topic), but that’s not all that’s at stake here. Next, we need to look at the loss of renewal and expansion revenue which can represent a significant number. Think about how your company operates, and how much of your business model depends on renewals… the answer will vary. Then think about how much revenue your organization receives from expanding those relationships. Expansion revenue represents the additional income a company receives from cross or upselling to its existing base. But if that base is diminishing, so is our ability to expand. Are your company’s analytics designed to recognize these trends? And more importantly, what actions are being driven off of those insights? If you keep this statistic in mind, you will start to shift your focus to uncover and act on valuable data insights. The probability of selling to an existing customer is 60-70 % while the probability of selling to a new prospect is 5-20 percent.”
Now before you run off to start figuring out how much attrition is really costing, keep reading to make sure you have factored in everything. This leads us to the other ‘obvious’ metric we should be considering, the cost to acquisition. Depending on your business model, this can be a tricky calculation but it is critical to truly understanding what’s at stake. Acquiring a new customer is anywhere from 5 to 25 times more expensive than retaining an existing one. This is another opportunity to improve your bottom line with machine learning. Your sales teams yearn to provide more solutions to your prospective customers, but what tools do they have to realize that goal? Do they start their day with an insights-driven dashboard? And are your processes set up to take advantage of those insights? If the correct action framework is implemented/adopted, the sales team will not only benefit from but they can also start feeding valuable data back into your models. As they take action and record those activities, you begin to see the model continue to improve.
Enough of the obvious. Let’s jump into some more subtle costs when a customer decides to take their business elsewhere. How much have you considered the negative brand impact on your existing customer base? Word of mouth and networking go a long way in any business model, and if that network is not seeing positive reviews, you can bet that will start to eat away at your bottom line because referrals will decrease. This is another area in which ML can be implemented to uncover and ultimately be used to solve the issue. Listening for customer sentiment is nothing new, but taking that data and creating an actionable strategy is what will set you apart from your competition. Another indicator of a larger problem is how much you are spending to defend your online reputation. Not only could you lose revenue from lost referrals, but you are also adding costs to combat the negative effect on your network. And if you are a brand leader, this fallout could be more severe, impacting marketing dollars as well. Most companies will look to marketing data, along with customer service metrics and insights to see why the brand is taking a hit. And while this can be beneficial, it would be better if they had a dashboard that highlights potential brand issues and offers the potential next best action for multiple personas. After building a relationship, customer spend grows alongside trust. Eventually, loyal customers spend 67% more than new ones. (Bain)
Second Order Opportunity
This one might be the hardest to quantify, but understanding it and improving customer engagement are the first steps to lessening the blow. Loss of second order revenue can be viewed from two perspectives with the most glaring being lost customers not renewing or purchasing more. In this scenario it’s simple, a customer leaves and will no longer order/renew your product/service. However, consider the customer who purchased from you and had a bad experience, then they leave their organization and join another. Now that your former customer is in control of the purchasing decisions for their new company, how likely do you think it is that they will reach out to you in the RFP selection process? That was a rhetorical question! This cost is one that would be hard to calculate, but being aware of its impact will help you demand better customer engagement.
So Now What
Now that we’ve added up the costs of customer churn, how can you use your data to reverse these potential losses? Step one is in understanding what your data is telling you, and to do this you must first set up your data on a solid foundation. This is where most companies get a little apprehensive because it seems like a daunting endeavor, but it doesn’t have to be. Atrium can help you organize and unlock your data to not only build a model to identify potential customer attrition but to also turn those insights into action.
One of the more challenging parts of any AI/ML implementation is determining what actions to take and how they can be consumed by the users. This necessitates the involvement of your marketing, sales, and service teams in the creation of the action framework. Loosely defined, an action framework is a set of recommendations that will decrease or increase the likelihood of a particular outcome based upon one or more models. An action framework for customer churn could be a model that recommends the best time and method to contact your customers to reduce the chance of churn. For example, it could recommend that due to the behaviors and specifics of this particular customer, your retention team will have more success if they follow-up in the evening. These insights will better allow your customer retention agents to plan out their day and have higher confidence in reaching your customers in a way that will increase engagement and satisfaction.
Your employees would love to know how they can improve customer engagement and retention. Let us show them how. If you want to reverse the effects of churn on your business and realize the full lifetime value of your customer, let Atrium be your guide!