The way we collect money has changed. In the era of automation, when big data and analytics are top talking points in strategy meetings around the corporate world, we have more information about our customers than we've ever had before. But does that mean we are smarter?
At the 2016 AICM National Conference I was invited to provide an overview of how the evolving use of analytics is influencing the way agencies are interacting with customers. The half-hour lunchtime session allowed for just that – an overview, an opportunity to scratch the surface of a topic that, by its very nature, provides increasing rewards for deeper analysis.
The presentation focused on how a few basic analytic techniques can be used to understand customers and optimise their engagement to enhance collection outcomes. As we know from managing our internal teams, the key to staff engagement is understanding what motivates an individual or group. Engaging customers is no different – the challenge is in filtering the extensive amount of data we have about our customers, into something meaningful and, critically, effective action. To do this, we need to discard our assumptions and embrace the theories of behavioural economics.
One key assumption we make about our customers – and ourselves – is that we are all rational decision makers. Behavioural economic theory tells us that we are not, and the evidence can be surprisingly compelling. When it comes to making decisions, most people favour the path of least resistance, even when it is clearly not the better choice. In essence, we are choosing to opt out of the decision making process, which seems simpler, but is still a decision by proxy. Understanding this behavioural tendency can enable companies to influence customer behaviour in very simple ways.
For example, a study conducted by Eric Johnson and Daniel Goldstein on Organ Donation concluded the decision to donate organs is influenced far less by advertising and direct marketing campaigns than by the inclusion of an opt-out tick box on driver's licence renewal. In fact, countries that appealed to their citizens' morals and sense of social duty underperformed those that appealed to their citizens' desire to avoid making a decision by 250%.
In another example, the Economist newspaper recently offered subscribers two levels of commitment: online only for $56, or online and print for $125. Naturally, the majority of customers opted for the cheaper, online only version, contrary to the publisher's preference for print. This led to a revision of strategy, and the subsequent introduction of a third option including both online and print for $125 that quickly became the most popular choice. In this case, understanding customers' unwillingness to choose one over the other, unless influenced heavily by price, allowed the publishers to subtly influence their behaviour.
Just as politicians crave the top spot on a ballot paper, companies can use this 'donkey vote' mentality to their advantage. Understanding this fundamental truth of customer behaviour enables us to 'frame the question' to our advantage.
At the conference, I asked a room full of credit managers who considered themselves rational thinkers, if they had ever purchased something they didn't need in order to get something for free. Online shoppers in the room identified with the scenario of purchasing additional items just to reach a free shipping threshold. Many identified with feeling less pain from using a credit card instead of cash, or undertaking a 10 kilometre round trip to use a four cent per litre discount docket.
None of these are rational decisions, but we immediately recognise them as fundamental and accepted consumer behaviour. We can recognise that smart companies are already exploiting this behaviour to their benefit, so what do we need to do to understand our customers and similarly prompt them to make the decision we prefer they make?
Firstly, we need to understand what triggers our customers to behave the way they do. Just because we may behave irrationally does not mean that our behaviour is random. As we have seen, with the right insights, irrationality can be quite predictable.
Let's look at how these theories can apply to credit management. In a recent review of collection strategy, we challenged 12 collection letters with varying content, colours and fonts. The message contained in each letter was essentially the same.
The existing, or 'champion' letter, had historically achieved an average engagement rate of 8.64%. We trialled the 12 'challenger' letters over eight weeks with 20,000 customers with varying outcomes, including one achieving an engagement rate of 11.64%, equating to an additional $300,000 in early collections for this client. A good result, but only the beginning of the process.
To fully assess the success of the various strategies, we needed to analyse the results across a number of customer profiles. At a high level, there were four basic categories of customers: willing and able to pay, likely and able to pay, likely and unable to pay, unlikely and unable to pay.
We looked at the trial results across each customer profile and found that while our engagement had significantly increased across the customer profiles with a high propensity to pay, we had actually decreased our engagement rate for customers with a low propensity to pay by around 10%. Further analysis revealed that each of the 12 letters performed differently across the four profiles. By looking one level deeper into the results from our trial and matching letters against profiles we have built over years of understanding customer behaviour, we increased our overall engagement rate by 6.1%, resulting in increased collections of more than $600,000 for our client.
Another strategy we ran this year focused on connecting with customers who had not engaged with us at all over the debt placement period. Using what we know about customer behaviour and general willingness to opt in where it is presented as an easy option, we were able to reach 7% of the group.
As with our previous example, we then applied a digital strategy across our customer groups, presenting slightly different options, attempting to ascertain an optimal offer for each one, noticing engagement for each profile dropped away at different points. To optimise the outcome, we tailored the strategy to each customer profile and maximised customer engagement by around 12% overall.
None of this would have been possible without a commitment to building and refining these customer profiles by distilling data from millions of contact attempts into meaningful and actionable insights. By viewing this analysis through the prism of behavioural economics, it is possible to predict and influence customer behaviour, even in the most emotional interactions that have the potential to drive even more irrational behaviour.
For us, that means constant analysis and challenging of our accepted processes must be the norm. As analytic tools become simpler to use and more widely available, successful credit managers must move away from a broad brush approach to analysing trends and closely monitor customer behaviour.
Understanding human nature – and our own irrationality – can actually be the key to success.
By Nicholas Harrak, Chief Operating Officer, recoveriescorp