VRM: Exit, Voice and Loyalty

Feb 08, 2009

This post is part of a series:

  1. Vendor Relationship Management
  2. VRM: Rent, don’t buy
  3. Data Egalitarianism
You may find it helpful to read (or at least skim) these before continuing.

Exit, Voice and Loyalty is the title of a 1970 book by the economist Albert O. Hirschman in which he examines the failures of organisations and the influences that can help to rescue them from failure. His thesis (by no means unique) is that organisations tend, naturally, to have a considerable amount of ‘slack’ or under-performance in their activities and, absent any pressures to the contrary, this slack will increase over time as the organisation becomes less efficient. Without feedback to regulate performance, organisations will simply drift and, crucially, this results in poorer service for their customers ('customers’ is here used as a catch-all for consumers, citizens, patients etc.).

Hirschman identifies two primary means by which customers of an organisation can pressure that organisation to improve: exit and voice.

Exit is simply the choice to sever the existing relationship - no longer buying from a particular shop or a particular brand, choosing to be treated in a different hospital, or emigrating to another country. If enough people do this, it should send a clear warning signal to the organisation that they are failing to meet the needs of a significant number of their customers.

Voice is exercised when the customer chooses to complain, campaign or attempt some kind of communication with the organisation without exiting (though the implied threat of exit remains in many cases). Examples would include writing a letter, organising a petition or making a view known on a blog. Voice allows the speaker to request specific changes to how the organisation operates.

A third concept, loyalty, is covered in the book - loyalty affects the likelihood of choosing exit or voice, as well as the amount of decline in performance a customer is willing to put up with before making the choice.

Astute observers may notice that exit is the system primarily associated with markets in which customers can choose from a wide number of providers and there is little cost to switching, and voice is the system primarily associated with democratic government, in which government claims a monopoly on the provision of certain services and exit is a costly option, but elections and public pressure can mean that voice cannot simply be ignored. Of course, this distinction is not a clean one, but it does reflect certain tendencies. Hirschmann himself notes that voice is the preferred concept for political scientists, whilst exit is the preferred concept for economists.

This quote is illuminating:

The customer who, dissatisfied with the product of one firm, shifts to that of another, uses the market to defend his welfare or to improve his position; and he also sets in motion market forces which may induce recovery on the part of the firm that has declined in comparative performance. This is the sort of mechanism economics thrives on. It is neat - one either exists or one does not; it is impersonal - any face-to-face confrontation between customer and firm with its imponderable and unpredictable elements is avoided and success and failure of the organisation are communicated to it by a set of statistics; and it is indirect - any recovery on the part of the declining firm comes by courtesy of the Invisible Hand, as an unintended by-product of the customer’s decision to shift. In all these respects, voice is just the opposite of exit. It is a far more “messy” concept because it can be graduated, all the way from faint grumbling to violent protest; it implies articulation of one’s critical opinions rather than a private, “secret” vote in the anonymity of a supermarket; and finally, it is direct and straightforward rather than roundabout.

Hirschman goes on to point out that economists are apt to regard voice simply as too confusing to be useful. How can you measure performance based on the analysis of communication with customers, when the sample of customers who engage in this communication is self-selecting? And these messages are written in natural language, which can mean different things to different people. How do we know that something which causes great dissatisfaction to one person will have the same effect on others? In contrast, the “bottom line” is definitive: if sales went up, we know people like the product.

This is the logic which, albeit often poorly applied, lies behind many “market-oriented” reforms in government services in recent years. And it’s certainly true that exit has a unique ability to focus the minds of the people running those organisations whose customers are leaving in droves. But voice should not be so lightly dismissed. Voice can be much more specific about the reasons for dissatisfaction, in a way that exit cannot. Many businesses might not know why they lose customers; they are reliant on the data that they have gathered about those customers to identify what they may have in common, which in turn may give some hint as to why they are no longer customers.

VRM is a much more customer-driven approach to customer-vendor relations than typical business practices. Instead of vendors managing their customers, customers manage a choice of possible vendors. In doing so, they will likely be heavily swayed by the voices of other customers - in blogs, reviews, Twitter messages and so forth. And there’s always offline communications! Already, one of the world’s largest experiments in VRM, eBay, makes prominent use of customer reviews of vendors. Having many bad reviews will greatly harm a vendor, whilst positive feedback is extremely valuable. What this means is that bad reviews are not bad per se, but unaddressed bad reviews are potentially catastrophic; if you’re doing something wrong and you don’t fix it, you can expect a long list of complaints that will undermine your business. Experiences like this will become common as businesses attempt to clean up the damage to their reputations.

However, VRM should allow for businesses to be much more reactive to voice. By encouraging customers to come forward, even with complaints, the business can react sooner and deliver a better experience in future. This picks up a point I made in an earlier post, that VRM allows the business model of a small firm with personal relationships with its customers to scale up to a mid-size business and beyond. When Hirschman wrote - in 1970 - that the primary motivating factor for businesses was almost always fear of customer exit rather than voice, it was with an implicit assumption that any corporation over a certain size would be “faceless” and incapable of understanding the complexity of their customer’s opinions. The first exceptions to that rule were created by the major multinational corporations who have enough customer data to gain statistical insights; VRM opens up that possibility for smaller businesses. And so, voice is, at least as far as the eBay example is concerned, providing a useful alternative and complement to exit as a measurement of success.