Top

Call Center and Soccer: Two of a kind

Thomas Wieberneit | Jun 3, 2017 48 views No Comments

Share on LinkedIn

One of the eternal problems in a call center is getting an enquiry routed to the right agent. This is doubly true in a mobile world that demands conversational support in near real time.

Add to this the fact that most customers seeking support already failed to find an answer to their inquiry using an FAQ, a community, or other self-services. In this situation customers expect an answer within few minutes, if not seconds.

On top of this, the ability to make engagements with the company easy, efficient, and ideally joyful, is becoming more and more a distinguishing factor for companies. Customer experience is the result of engagements, and for humans the experience gained from the most recent engagement tends to have a higher influence than older ones. Consequently, a positive customer experience matters, not only during marketing- and sales, but even more so in situations that require active help of the company that sold the product or service. So, getting a solution to an issue must be as easy and as human as possible.

The challenge is that every support organization needs to live and work with limited resources – human as well as technical ones.

It is Like a Good Game of Soccer

11 players and a ball. There is a goal keeper, are defenders, midfielders and attackers who play as a team against their opposition, trained by their coach and guided by the captain. There is a core team, and some players may be assigned to different roles, even within one game. Depending on the opposition team, the coach and the captain change player assignments, tactics and roles.

Together they play the ball with the objective of scoring.

Like a soccer team call center agents are organized around their strengths and like soccer players the agents are – or should be – carefully trained and assigned to roles on their field: the queues with which incidents are managed. Like the soccer players, agents can be assigned to different positions on the field: different queues.



The incident is the ball and the scoring is resolving the incident.

With all its intrinsics, soccer is a simple game. It has a few rules that dictate the do’s and don’ts. Following these rules, considering available players and their strengths, is the table stakes. Yet, this doesn’t make for a game that the spectators may find exciting – nor might it lead to scoring. There is an art to it, the art of how and to which player to move the ball in any given situation.

This is the same in a call center environment.

Yet, in some call centers, when receiving the incident, the team stops and asks the spectator where to start. The user is asked to qualify it and to route it into the right queue. This is a task (s)he is obviously ill prepared for.

Doing so would lead to a heavily disrupted game that no one would like to watch, not even talking of enjoying. Scores would be rare and far in between.

In a call center environment, this leads to unnecessary load. Load that is caused by re-routing the incident, delays in handovers and resolution, and ultimately frustration on both ends: The end user and the service agent.

In brief, it is causing a poor customer experience, following another poor experience.

There must be a better way.

And there is.

Back to our in-app support world.

Why not using the incident itself to automatically identifying the right queue and smartly routing it there? This is like the soccer players using the ball’s momentum and the given situation to proceed to score.

There are two basic possibilities to achieve this smart routing. One I would name the classic approach, the other one the text mining approach.

The Classic Approach

Automatic routing to the right queue can easily be achieved by using metadata that is provided by the app as part of the incident report. Incidents are tagged based on this metadata. The tags are used to route the incident to an appropriate queue or agent. A configurable automation does both, tagging and routing, without explicit intervention of the app user, who is already in distress enough as there is an issue that he or she couldn’t resolve.

This is a proven approach that requires an elaborate design of metadata and its collection within the app and the ability to build rule sets on top of this metadata in the service back end.

With sufficient data and advanced analytics tools this approach can also be used to offer pre-emptive support, which further improves the customer experience by avoiding the negative experience in the first instance.

This approach, however, finds its limits in the ability of instrumenting an app so that enough situations can be identified from the metadata for appropriate tagging. Not in the least because instrumenting has an adverse effect on the app.

The Text Mining Approach

With more and more service center software running in the cloud, massive improvements in AI, Natural Language Understanding (NLU) and machine learning technologies, additional information can be used: The description that gets submitted by the user as part of the incident report.

Albeit the users do not always provide accurate information, this is regularly enough to augment and improve the tagging, therefore getting a better routing and ultimately faster resolution.

Given that in many systems the first line service agents can resolve incidents by themselves consulting a knowledge base the AI might be able to resolve the issue before escalating it to an agent.

Smart Routing is the Future

This paves the way. Currently leading systems smartly queue and route incidents based upon an elaborate architecture of metadata and tags that are automatically given based upon the metadata. This already tremendously helps companies in providing good service.

The next step is including the descriptions given by the users into the mix and to use it to suggest solutions to agents until confidence is high enough to have the software independently suggest solutions to the users or to route the incidents. NLU in combination with machine learning / deep learning are nearly at this stage.

In either case, smart routing is reducing the friction in the support process and therefore improving customer- and service agent experience. The result of this concentration on positive experience is a measurable benefit for the business – because of adding value for the customer.

Print Friendly, PDF & Email

Republished with author's permission from original post.


Recent Editor's Picks:


Categories: BlogContact CenterService and Support

48 views

No responses yet, why not leave yours?

Add Your Comment (All comments are reviewed by moderator, no spam permitted!)