The Ultimate Customer Measure Effort
The Holy Grail of Effort measurement?
"Resolution of my query” or First Contact Resolution (FCR) ranks in the top three customer experience needs in all the customer research we have seen, but it has always been hard to measure. FCR is often seen as the “holy grail” of customer experience outcomes. This also aligns with the Customer Executive Board’s research (2010) which demonstrated the value of reducing customer effort. Resolving problems reduces effort and prevents the costs of lengthy repeat interactions. We are known for our opinion that no need for an interaction is the lowest effort of all, (see our 2004 book The Best Service is No Service) and that companies should see the overall rate of contact (or Contacts per X) as a critical measure. If a contact is necessary, then it’s obvious that customers want their problem sorted.
Proven mechanisms to measure resolution (FCR) now exist, that we have found to be better than all the previous techniques. They are both accurate and provide a granularity that make them valuable in improvements to processes and individual performance. In this paper we’ll describe why FCR is so valuable to organisations as well as customers. We’ll cover why many of the traditional resolution (and their flip side repeat) measurements are flawed and why it’s been such a big problem to get an accurate handle on this. Then we’ll describe the emerging techniques which produce improved measurement at useful levels of detail. Finally, we’ll explain how these new measures can be used to change many aspects of operations and processes and illustrate with a case study.
Why Resolution is So Critical
Where resolution is poor it results in additional costs from repeat contacts which we have shown always take longer to resolve as shown in the chart. Repeat contacts account for between 10-25% of contacts by volume (in the >75 clients where we have measured this) but they result in an even larger proportion of workload. We find that the handle
time of repeat contacts is up to 50% longer than the original contact and a third or fourth takes even longer. This is not explained by customers' behaviours such as anger and frustration, restating the problem, arguing about solutions and demanding to speak to others in authority. Up to one third of all service costs are caused by fumbled and poor customer experiences or lack of resolution. These unresolved problems may also escalate to more expensive outcomes such as complaints or in some situations cause customers to look elsewhere for the product or service. Non-resolution is therefore an expensive problem that generates repeat contacts and “failure demand”. Increasing resolution is often a major opportunity to improve the customer experience and reduce costs, a nirvana like outcome.
The Challenges in Resolution Measurement
Given its importance and value you would think that most organisations would have a tight handle on resolution in general and FCR, but we rarely find that to be true. Over half our past clients have had no resolution measures and most others use two common, but flawed, mechanisms. The first method is to build an FCR question into post interaction feedback surveys. It’s often a simple “yes/no” in response to a question like, “Was your query resolved?”. The problem with this mechanism is that customers often
complete the contact thinking something is sorted when it isn’t. A staff member may have said. “we’ll get that done for you”, or “I’ve made that request for you”, or even “I’ll look into that and get it sorted”, all of which make the customer think their problem is resolved. However, the reality is that this often results in hand offs and work requests to other areas or extended durations of work that lead to a further call. A fourth outcome is that the customer themselves has work to do which may then lead to a follow up contact. In the example shown, the rate of true resolution was only 40% and a further 45% resulted in actions but weren’t resolved at that point. In 15% of contacts no resolution occurred. Asking a customer a simple Y/N question in examples like this will rarely be accurate. We also have a problem with asking the customer if things were resolved, putting them to more effort, when the business should have mechanisms to know.
The second common measurement is to “estimate” FCR by creating a crude measure of repeat contact and deduce resolution. Several of our large clients tracked a second call by the customer to the same service queue within a fixed period (some have used seven days, others a month) as a measure of repeat contact and by implication non-resolution of the first contact. This is a crude guesstimate at best. Firstly, it may take different periods of time before a customer realises something wasn’t sorted e.g. perhaps not till they see their next bill, so a short period may not cover it. Another issue is that customers may call twice in a short period for very different reasons which gets picked up as non-resolution by this measurement when it isn’t. Thirdly, the customer may have needed to act, so the second call is in some way legitimate but instead it’s being counted as a problem. Lastly, a single overall measure doesn’t have the granularity to drive improvement as it doesn’t tell you about individual processes and provides inaccurate feedback on staff. What this type of mechanism lacks is understanding of why the customer called and knowing whether the next contact is related or not.
Companies are also beginning to grapple with Omni channel and multi-channel environments. It’s even more critical to put in place an accurate measure of resolution in a multi-channel world. Companies can seldom measure multiple customer contact across one channel, let alone whether the customer started a process in one channel and then repeated the contact in another. Few companies we have worked with were able to “relate” the contacts in different channels and that is why we think the new mechanisms that are emerging are so significant because they are able to track cross channel contact and resolution.
New mechanisms to measure first contact resolution
The new mechanisms bring together two key capabilities: effective and consistent analysis of contact drivers and the power of analytics to make reporting repeatable, and
consistent across channels. The bedrock of the analysis and measurement is correct classification of all contacts (calls, email, chat, etc) into the same customer contact categories. That means that the organisation can report “why” customers are making contact across channels. Our method is a two-stage process. First, we sample contact and use “humans” to come up with an initial contact categorisation. For example, listening to calls may highlight 30-80 common categories such as “where’s my bill?” or “I want to pay”. This will also cover the range of language that customers use that all mean, “I want to pay”. Once the categories are clear and language identified we can train the machine (or analysis tools) to look for those categories across all contacts.
That mechanism alone delivers effective “demand reporting” of the numerous reasons customers make contact. This reporting enables what we call demand management and have explained in other publications like, “Omni Channel Best Service is No Service”. It embraces all contacts and it shows other potential problems that are driving all types of contact without separating out or reporting repeats.
The power of analytics enables cross channel integration. We get all contacts across all channels (including calls) into a text form so that common analysis of all contacts is possible. That’s easy with chat, emails and the like which are already text based. Now
speech-to-text tools are common, it makes this possible with calls as well, to a high degree of accuracy. Once the machine has been “taught”, the AI tools can learn and expand the classifications and perform this analysis day by day.
The second thing we teach the machine is, how to spot “repeat” or non-resolved contacts. Once contacts have been classified into reasons, we can do things that make this accurate and valuable. This includes looking for the same contact type recurring within and across channels for that customer. We can fine tune the “period” for the repeat contact and make them specific to that contact type. Lastly, we can add other associated costs of contact e.g. sending a repair team or resending a bill, to make the reasons for change clearer.
These capabilities add far greater accuracy and sophistication to the analysis. For the first time the organisation can really understand whether contacts have been resolved and the cost associated with repeats. Better still, customers and staff don’t have to expend effort to enable that analysis. It does involve human intervention and business smarts to set up the analytics, but once we’ve trained the machine, the analytics do the work.
Impacts of resolution and contact type measures
The real power of internal FCR and analytics become apparent when added to operational execution. We often call repeat contacts “snowballs” meaning problems get bigger the more they repeat. Once an organisation has data on how many problems agents cause (snowballs created) versus how many they solve (snowballs melted) this can replace traditional quality sampling and measurement. This is a far better indicator of a quality outcome and if it is measured across all contacts then it is more accurate than small quality samples. This form of analytics and resolution reporting can then be tied into rewards for staff and coaching by team leaders. Instead of sampling 5-10 calls a month for quality, organisations can measure quality as defined by FCR across all contacts.
The second key outcome is that reporting resolution and contact rates at the contact type level, enables other types of process improvement. Now it becomes clear which contact categories have high repeat rates. That can lead to analysis of whether it is a process or a training problem. The analysis can also show how repeats and resolution are working across channels and highlights issues in apps or digital applications leading to other types of failure demand. A final benefit for customers (and costs) is that organisations can be more selective about what to survey customers about.
Case Study Example
So how does a company really focus internally on customer effort and use this analytics-
powered approach? The charts show a company that had already been convinced of the idea of focusing on call per customer as a key metric. Having subscribed to the “Best Service is No Service”, methodology they had an advanced programme of contact understanding and contact type reporting. However, the contacts per customer were not declining. They had a clear view of cost by contact but this view alone was not stimulating the organisational change required. Departments were fixed in their ways of operating and could not see a way to attack the problem of call volume.
The idea of FCR analytics matched to contact code was the crucial next step. They were able to construct a contact code library or taxonomy of key interactions.
They used algorithms to allocate all contact to the key codes and to understand the relationship of different contacts. The FCR rates were analysed in two fundamental ways; firstly, the repeat contact by code and repeat contact by agent. The repeat resolution by contact type meant that the organisation understood, for the first time, which types of contact were harder to resolve. This gave a very different view of cost to the organisation. They were able to target the 30% cost of non- resolution immediately and work on processes that had inherent complexities and training gaps for all staff.
FCR by agent could be measured accurately and simple charts used to highlight issues. One powerful view (shown above) mapped agents on two dimensions:
the rate of repeat contacts they caused versus the rate of repeats that they fixed. This wasn’t used as a stick but to consider training on specific transaction sets, better internal practice design and training specific users who needed help. The results, (shown right) were very impressive. Improved, FCR increased satisfaction, increased revenues and reduced the cost to serve. Needless to say, the board is delighted.
Conclusion - Measurement and Improvement Nirvana
We hope you agree with our conclusion that FCR is a critically important measure for customer experience improvement. We’ve also shown that accurate internal measures are now possible through granular analysis of contact types and repeats and that they can help drive process and individual improvement. Combined with measurement and reporting of contact reasons, it’s possible to reduce costs and improve the experience while reducing the need for surveys. The combination of analytics, reporting and focused continuous improvement produces financial outcomes and reduced customer effort.
If you are interested in this world of reduced contact and transformed measurement and coaching, please get in touch. If you would like more information email firstname.lastname@example.org or call 03 9499 3550. More details are at www.limebridge.com.au