Experimentation

Debt collectors have long conducted informal experiments with the debtors they are dealing with. As I explore in Chapter 3, in the early days of consumer credit collection, in which individual collectors often dealing with individual debtors (in the US in particular in the 1940s and 1950s; in the UK much later), this involved practices like trying to tailor letters to the person you were trying to collect from, or trying to gauge the different success rates of a new letter design sent to the full range of debtors you were tasked with dealing with.

However, as collections work became concerned with far larger numbers of debtors, and as automation gradually began to make it possible to quickly and easily customise letters sent to debtors en masse, experimentation became a practice that collectors took far more seriously. I look, for example, at a 1960s experiments in the US with debtors that explicitly looked to successfully manage debtors’ emotional states. As I’ve written elsewhere, debt collectors can then very much be seen as one of the ancestors of the recent and controversial Facebook experiment.

Nowadays, experimentation has become a routine practice in collections work, sometimes referred to as ‘Champion Challenger‘. As a collector, the idea is that your ‘champion’ — that is, your existing collections routine, which might involve and telephone calls and particular sets of letters, or even text messages and emails, combined in a certain sequences, with prompts coming at certain pre-defined times — is then tested against against a new way of doing collections: the challenger. If the challenger is more successful — that is, it raises more income, always accounting of course for any additional costs it might generate — then, in theory, the new approach should be rolled out across the particular set of debtors that are being worked with.

This, in turn, might be combined with the analysis of data that is held by the collector about a debtor or a particular set of debtors. This might involve the modelling of various potential outcomes, based on what is already known about how debtors respond to certain prompts or in certain situations. If champion challenge is a form of ‘in vivo’ experimenation, then this is experimentation ‘in vitro’. These processes are explored in some detail in Chapter 4.

In the book I examine some of these technologies, their history, their proliferation, their contemporary use, as well as looking at both their sometimes counter-intuitive effects and the political questions that arise.

Related posts: