Customer surveys are a brilliant idea, no? No, not always. When you try to measure superior service and the emotional connection a customer has with you or your product, it can be difficult to get data that really helps pinpoint where and how to improve. In this article I will highlight some pitfalls in satisfaction surveys and measures and suggest some simpler approaches to measurement that will help drive the right kind of change.

There’s an excellent cartoon on marketer Tom Fishburne’s blog which shows how those of us in customer-facing organisations frequently mislead ourselves that we know and understand the emotional connection with the customer. Organisations can spend a vast amount of time designing the experience customers will have but most of the time customers don’t want to have an experience with you in the first place. This is particularly true for organisations whose product is seen as a utility – in these cases when a customer initiates contact it’s often because a product or service which is often taken for granted suddenly doesn’t work.

Pitfall 1: over-complex measurement

The above point doesn’t mean that you shouldn’t measure customer service since, when the customer reluctantly calls, you need to know they are getting great service, but it does mean that you should exercise caution in the number of measures you have or, more specifically, the ones you apply.

The great thing about any operational unit is that there are a lot of things you can measure about it: wait time, throughput, productivity, defect rate, waste and so on. These are all excellent things to measure but before too long you can build up a small army of people generating reports with custom-built measures on them. In addition to adding to an organisation’s overhead, this creates complexity and, more seriously, can lead to a lack of alignment amongst management about what is important and how well the customer is being served.

Let’s say you have a customer service organisation with the following units: Sales, Faults and Complaints. You measure the customers’ satisfaction with each one and their satisfaction with the whole organisation. If Sales is full of polite, friendly people they may get a good satisfaction rating but if the bright, friendly people sell the wrong product, Complaints may see an increase in traffic. Complaints may be full of deeply empathic individuals but because the customer had to complain in the first place they are on a hiding to nothing as satisfaction with the complaints unit will be coloured by the customer’s overall dissatisfaction. Similarly, satisfaction with the Faults unit may be influenced by the ability or otherwise of the service engineers to rectify the fault. The solution organisations can often take is to have all the relevant measures available to managers. This seems laudable but means that it’s very difficult to see the connection between different units’ satisfaction scores.

Pitfall 2: asking the wrong questions

You might expect that the obvious way to find out if a customer has had a positive emotional connection with your organisation is to ask them how they feel. Fine, in theory – if your organisation has some form of psychic probe technology at its disposal – but in practice you will be getting the data some time after the event, when it may be difficult for the customer to pinpoint exactly how they felt at the time. Having said that asking ‘how did you feel about X’ is a better question than ‘how satisfied were you with X (on a scale from extremely satisfied to extremely dissatisfied)’ because satisfaction is not the same as emotional connection. For example I was quite satisfied with my last trip to a petrol station to fill up my car but I only felt neutral about it at best.

Scale data to measure satisfaction is superficially useful because it can be used to plot nice distribution graphs which might tell you that the majority of customers are slightly more satisfied than dissatisfied but this is a bit like having your head in the freezer and your feet in the oven and saying that on average you are quite comfortable! Verbatim quotes for those at the extremes will get you closer to how customers actually feel. However having said that, it’s not the be-all and end-all of measurement and below I suggest some ideas on how to get focused on useful measurement.

Idea 1: only measure real stuff

I have often been amazed by the ability of organisations to shy away from the simplicity of measuring real things like delivery time, fulfilment levels and so on. Very often this direct approach, which can easily drive process improvement, gets confused with customer satisfaction measures.

Before I run the risk of contradicting what I said above it is important to understand how customers feel about your service, but if they don’t get the product or service – on time, in full – they are very unlikely to feel good about it. My recommended approach would be to measure – in a statistically sound way – whatever things indicate process failure and to use sampling techniques – focus groups for example – to tell you how much your customers love you.

Idea 2: ask front line staff

One great resource that service organisations have that is often under-exploited is the front-line service staff. Getting their feedback on the quality of their interactions is one very good source of measurement. The trick here is, obviously, to decouple this feedback from any form of reward mechanism.

Idea 3: ask non-front line staff

This is more likely to test how well middle managers with less direct customer contact have a handle on customer behaviour and feelings, and should be carried out independently of any exercise involving front-line staff. It has the potential to highlight how the overall processes impact customer experience, depending on the level of visibility the managers have.

Idea 4: use complaints and praise to drive improvement

I have said earlier that the extremely satisfied and dissatisfied customers are important to focus on and, to reduce the effort expended on measuring everyone’s satisfaction, why not use complaint data or direct customer feedback to highlight where a service may have failed? (I have a live example with my broadband service which may well make it to the blog next week.)

In organisations driven by customer satisfaction it can be easy to fall into the trap of measuring satisfaction with complaint handling rather than the substance of the complaint itself and, like feedback from front-line staff this squanders a very valuable resource.

The topic of measurement is a large and often controversial one and this post can only scratch the surface. The main point is that, contrary to the title, superior customer service organisations should care about how customers feel but should be very focused in how they measure it.