Alison Bond: Forget management by numbers! 3 tips for real service management

Ten years ago it became clear to me that customer satisfaction measurement wasn’t working. I felt like I had been part of the problem. I had written pieces and made speeches quoting lines like “if you can’t count it, you can’t control it” and I really believed it, having worked in airlines for many years where service was a differentiator which really mattered. 

 

I had spent many hours sat in meetings where we tried to come up with ways of “getting the customer facing staff to deliver a consistent product”.  The answer we came up with, 18 years ago was pretty innovative and it certainly worked. It got the airline back in to profit in just over a year, and produced us our first ever airline service award. So why doesn’t that way work anymore? 

The simple answer is that it might, but the reality is that to make it work many organisations have designed a strict code of command and control which they enforce with some rigour and that has the effect of beating the personality out of the service. The problem with this is that if you lose the personality from service then you are also in danger of losing your key differentiator and quite likely your best people as well. This is because I have yet to find someone who knows they are doing a good job for the customer, being told to stop doing it like that and do it a new way designed miles away from the customer, but which suits a mechanised process.

Why is this the case now when it wasn’t 18 years ago? Crucially way back then it was much tougher to measure performance; IT and software were not as well developed or used and the staff were used to working freestyle in service professions. We were one of the first big consumer businesses to automate our reservations systems with our direct marketing and our service measurement. It was a big job and I had to specify the programmes and manage a number of providers to generate the system I was after to make it work. 

A number of the free styling staff were horrified about what the capabilities of the system which had been developed were and complained bitterly. I vividly remember a Senior Captain at the airline coming into my office and berating myself and my very sweet PhD assistant about the horrors and the dangers of the monster we were creating. Now I realise that he was - to a very large extent - right. However, that monster saved the airline from bankruptcy and kept several thousand people in jobs during the torrid time of the early 90s – yup, we have had recessions before.

So with the benefit of 20:20 hindsight what have I learnt about managing customers and service delivery? And if I was having the same conversation with that Captain now, what safeguards and freedoms would I put in place to ensure that we achieved real service management rather than management by numbers? The answer is three things.

Command and control using numbers has become something which we accept as part of big organisation life. The trouble is that it makes people unhappy. Staff don’t like working in that sort of environment and customers like to be treated like a human being rather than a number. There is a big question mark in the idea of taking out the targets; when I talk to those who run the organisations who use numbers as their main way of controlling their people they worry taking them out will mean that the people in organisations become inefficient, or that they, personally, will lose control. 

I think the answer to this is to have numbers which motivate and create freedom for staff and teams. This means less transactional measurement, that is the busy things which people get up to, and more 'benefits' measures, that is more measurement in the space where staff are asked to co-create the service offered in a way which gives motivation to deliver more benefits to all the stakeholders in the relationship, be they staff, management, shareholders, customers, even suppliers. 

The way I would describe these to the Captain would be that yes we are still collecting the transactional numbers, as they are important for us in the management team as signposts to show us that we are going in the right direction, but that bonuses and bouquets are being given out on the basis of performance against the benefits targets.

If I knew then what I know now I would not have published the league tables of which area of the business was performing better than another. I would have shared where there appeared to be excellence that went against the norm, but would not have made coming top of the league in any way important, in fact I would have made the measures broad enough that it would have been very difficult for anyone to come top of every measure. 

This is not because I am anti-competitive; I love to win at games as much as the next person but in services there is very little which is finite. In fact because they are delivered by people the permutations in services are infinite. Therefore, any league based around services is frankly nonsense. We had so many variations in our service mix; geography, nationality, size of airport, size of plane, type of business mix and it meant that the leagues were impossible to 'win' fairly. What we found was that if the mix an airport base worked with was less likely to create high scores they tended to 'absent' themselves from the conversation and became less engaged than they were before we started our measurement. 

We fluked, and I really do mean fluked, a result which meant that those most likely to win the league were also bringing in the most profit, so we were lucky. You in your organisation may not be so lucky, and you can burn through a lot of money, a big chunk of staff morale and lose your best customers if you try to win the league when that fluke of numbers doesn’t run your way. Just look at the public sector to see several thousand examples of this working against service providers.

I built that monster at the airline, and I had him tamed. He stayed on a lead which was very firmly tied to my desk. I shared all his uttering’s with everyone at the airline, making sure that the people at the front line heard them first. This way they were as much in charge of the monster as I was. I also appointed monster babysitters who were very adept at interpreting his words and insights which meant that he stayed a friendly and benign monster. But I left and started to help other people build monsters. Although of course then I didn’t realise they were monsters. 

One monster, I helped build was used to sell the company I helped build it for £500 million several years ago. The lucky company, who bought it, didn’t understand about the monster and its behaviour and a couple of year’s later they turned the lights off and wrote off the majority of that £500 million. So, Captain, what do I do now to ensure that I don’t create monsters? 

I dissected every part of the monster which I had helped build and looked at the things which made them dangerous instead of helpful. How could I design in 'bite' when it was needed but make it helpful and generous to those organisations that needed their help? What I found was that if we measured in the space where people were being the best that they could, when they were delivering to those who needed the service, in a way which worked for every stakeholder then the monster became a pussy cat. Something which was there to help, to lower stress levels to show the way forward, to find and reward good behaviour. 

The key was to properly understand the organisation and build a system which worked with them, and motivated as it went. It was not something which needed controlling, because it was designed with the understanding that most people want to do the best they can for the organisation. In fact the systems are much simpler than the 'monsters' more fun and a very great deal cheaper. Doesn’t necessarily make me popular with other research companies but they work and they work because the people in those organisations want them to work.

So does it mean that all customer satisfaction measurement is a bad idea? Well, yes and no. It is not good if it conforms to any of the three issues I have talked about. But if you can make your measurement work on the principal that if you focus everyone in the organisation on positive, outward-facing benefits, and which support your organisation's best way of working, which the people you have can all buy in to and demonstrate everyday, then it should. 

If you are able to do this with your measurement then the processes will look after themselves and the satisfied customers, and the profit will drop out of the bottom of the organisation. Getting it right will take away the need to have lots of targets which take people's minds off the serious business of delivering benefits, and it is benefits which people buy - so it's benefits which need delivering. Simple really, the equation is: Measure transactions = Busy; Measure benefits = Better.