Twitter: to SLA or not to SLA?

Recently I’ve been thinking about the impact Twitter has had on how time is viewed from a customer service perspective.

Companies continually strive to achieve greater levels of efficiencies, extracting what insightful nuggets they can through the detailed analysis of call volumes, first time fixes, abandonment rates, average handling times and more. Customers on the other hand, armed with increasing ubiquitous smartphones and a rich plethora of tools for self-expression, are becoming ever more vocal and more demanding.

In relation to improved customer service, time is of course critical to both parties. Companies use time in the form of Service Level Agreement’s (SLAs) to manage and control customer expectations. Customers instead turn to Twitter in the hope that something might happen faster, ahead of time, and exceed their expectations.

A resulting tension from these two extremes manifests itself between the company’s need for constancy, consistency, uniformity and structure, and the customer’s arbitrary, subjective, movable and unpredictable demands. In this age of convenience, the idea of ‘now’ is all pervasive; Twitter has simply exacerbated the divide.

I used Twitter myself to provide customer service at The Carphone Warehouse (@GuyatCarphone), where I was Customer Knowledge Manager. I quickly realised that Twitter was not only a great way to identify problems or complaints rapidly but also to be seen to acknowledge or resolve them. However, once we adopted a more formal approach to Twitter customer service and set up a team of agents to provide this service, the need for structure, consistency of approach and constancy of service soon became evident.

A few months ago I posted the following question on a LinkedIn group ‘where social media meets customer service’: “Can Twitter have a meaningful SLA? Or is Twitter, because of its nature, simply the wrong type of platform to try to apply an SLA to?”

One of the contributors to the discussion was Esteban Kolsky, a well respected thinker in the field of social CRM. He wrote: “Why treat Twitter any differently than you did when you went to email, or chat, or even the phone? Would you consider using email for customer service without an SLA? Why would Twitter be any different? Just asking the questions means that Twitter is either not taken seriously or the possibility of doing customer service via Twitter is not seriously considered.”

I believe Kolsky makes an interesting point and wonder how long it took before an email SLA came into effect, or a telephone one, and what actually precipitated it? I do believe that asking the original question actually implies seriousness in considering Twitter as a viable customer service channel. The difference, however, between email and telephone is that these are resolution channels, whereas Twitter is more suited to being an identification channel, so what is a satisfactory identification SLA?

James Gurd, an eCommerce and marketing manager wrote: “If you are using Twitter as a customer service function (however you define its usage) then I think you do need an SLA but the SLA does not have to be based on specific metrics like “time”. It could simply be a measurable statement of intent…”

Certainly time plays a great role in our expectations, and those expectations result from our cumulative experiences. The use of Twitter as a customer service channel offers an unprecedented immediacy of response, whilst the mainstreaming of smartphones frees individuals from the constraints of geography. I can communicate with anyone, from anywhere, at anytime.

What sets each experience apart is how that expectation is unexpectedly exceeded, challenged or questioned. That expectation is also movable, with each instance determining what the expectation will be and whether it will be met, not met or exceeded.

This article was written as part of the April newsletter