Ken Turbitt Blog
Aidan Lawes Blog
Paul Gostick Blog
Dr Jenny Dugmore Blog
Shirley Lacy Blog
Alim Ozcan Blog
Juan Jimenez Blog
Ian Clayton Blog
Nas Ozcan Blog
Aidan Mills Blog

Digital Health Technology Vision 2018
Health organizations are embracing intelligent technologies but must do more to prepare for societal impact...

Ten Strategic Technology Trends for Government
Technologies that enable new service models for digital government must be at the top of the list for government organizations as they prioritize technology investments...


The Robots are Coming: Are CEOs Ready for the Era of Automation?
CEOs agree that robotics is going to make their companies more efficient, with 94% of those who've already adopted robotics saying that it's increased productivity in their business...


The 2015 Chief Digital Officer Study
More companies are appointing a Chief Digital Officer to join their C-suite - but are they doing it quickly enough?...


18th Annual Global CEO Survey
The United States has overtaken China as top target for growth for the first time in five years...

31 July 2009 | Dr Jenny Dugmore Blog
Send to a colleague | Add to MY ITP

Effective Service Level Agreements (SLAs)
This week Jenny looks at why SLAs are more effective if the supplier understands what matters to the customer in order to tune their processes and procedures...

  Dr Jenny Dugmore

I was recently discussing service targets with a client and was reminded of some work I did last year.  I was asked by an organisation to establish what their customers thought of their service, a very interesting mixture of customer satisfaction measurement and market research. 

The supplier had a series of detailed SLAs which had taken a lot of time and angst to develop and get agreed.  They had developed complex and detailed service reporting against the SLA targets.  They held regular service reviews, although they had problems getting senior customers to attend them.  All good practices, and now they wanted a benchmark before doing more improvement work. This was also because they had expected the number of complaints to decrease (and they were getting just as many as before), and that just maybe they would get a few accolades (and they were not). 

I’ve looked at this situation before and I always find it interesting, so I was very pleased to be asked to do this.  I found a few basic things which I’d seen it with other organisations. 

The first thing that emerged was the differences in terminology used by the customers and supplier.  To start with, the supplier thought one of their key services was ‘desktop support’.  That is what they called it in the SLAs; it was even what part of their support organisation was called.  None of their customers used this term when asked to describe ‘in their own words’ what services they used.  Most of the senior managers described their service as ‘the PC I use for email’, or even just ‘email’.  Some also made a reference to ‘the network. An even smaller number added that the supplier ‘fixed PCs and printers when they went wrong’.  This was a relatively complex set of services based on a sophisticated infrastructure, reduced in the customers’ minds to not much more than ‘email’! 

Some of the more junior staff had a slightly different view – they usually described it using terms that related to their role, ‘I use it to so XYZ’, although ‘the help desk’ also got some mentions.  The supplier had something called ‘the service desk’.

Nobody used exactly the same terms of their supplier (something I have found over many similar assignments).  The closest anyone came to using the terms in the SLAs was a manager who had been closely and rather anxiously involved as one of ‘customer representatives’ during drafting of the SLAs.  Even that person did not use exactly the same terms as the supplier. 

The one common theme was that each person interviewed at least knew that there was ‘some sort of agreement’ in place, some even referred to it as an SLA.  They usually knew that it defined ‘how quickly things would be fixed when they went wrong’.  A few people remembered that ‘important things would be fixed in four hours’.  Nobody referred to this being ‘95% in four hours’ or was sure what was meant by ‘important’ It was rare for anyone to know what else was in the SLAs. 

Over the years, doing this sort of work has changed my view of the strengths and weaknesses of SLAs. I have long since ceased to think they are the method of setting end user expectations on what services and service levels they should expect from a supplier.  Even if key targets are extensively publicised few people will remember them and will only care when they have a problem – and then they just want it fixed quickly.

I still think SLAs are an excellent idea, even though they do very little to directly set expectations.  Because many people are reluctant to sign an SLA until they are convinced it is correct to do so, a supplier has to give serious thought to understanding what matters to their customers, and a good SLA is based on that much better level of understanding.  So far so good.

I believe that the SLAs are effective only if the supplier then uses their much improved understanding of what matters to the customer to tune processes and procedures, such as to how priorities are set for many of their activities, including ‘fixing things’.  It is even more important to take that knowledge to set better goals for their own support staff.  The right set of goals can make a huge difference; the wrong goals can be a disaster and generate many bad practices as people aim for the wrong goal. Most of the benefits of SLAs are only visible to the customers only as a consistent and appropriate response to a request for help.

And in this case what were the next steps?  The supplier realised they needed to tune their procedures and they also changed most of the goals they set their support staff.  The result?  A gradual improvement in customer satisfaction as well as real improvements in service levels, but no changes to the SLAs and the customers and supplier still use different terminology.

Any feedback and comments are always welcome!

Dr Jenny Dugmore Email to a colleague | Add to MY ITP

terms & conditions