Hey Boss! Look at all the Stuff I’m Measuring!

Calls offered by half hour
Calls answered by half hour
Abandons by half hour
Busies by half hour
Calls blocked by half hour
Average talk time by half hour
Average not-ready time by half hour
Average handle time by half hour
Service level by half hour
% of calls answered in 10-second increments by half hour
Average speed to answer or delay time by half hour
Call volumes to DNIS

STOP THE MADNESS!

Hey, I got jokes.  At least I think I do.  But this isn’t one of them.  This is an actual list of measures I saw recently on a contact center management dashboard. Oh, and this was not the complete list.  There was more.  But guess what wasn’t there?  Look at the list again.  Not a single indicator of how the customer viewed the experience or how well customer service was doing actually reducing demand for service through the elimination of broken upstream process that drive unnecessary service contact volume.

The indexes above do have have value for workload and resource planning.  You need to insure you have enough people in the seats to handle contacts or that, in turn, will impact the customer experience as well.  But beyond that, they have no place on a customer service scorecard.  Now in fairness, this company does survey its customers.  It is attempting to measure first call resolution.  And, it’s even trying to understand how to extract value from New Promoter queries.  But, none of these indicators appear on the contact center management dashboard.  Nor, are they metrics to which management compensation is directly tied.  

The other issue I have with this list?  Its just too darn long.  One of the first tenants of a performance management system is to develop a set of KPIs that you can actively manage and work towards improving.  Too many numbers leads to diluted effort which usually means nothing (or very little) actually improves in the operation. 

A better approach?

First, focus your management effort on the things customers actually care about.  I’ve yet to see a statistical study that correlates a customer’s willingness to recommend with AHT.

Second, for any organization, contact center or otherwise, there is opportunity to improve understanding of and implementation of a high functioning performance management system.  The first step is to understand the difference between a Measure and a Metric

As you think through the design of a performance management scorecard, spend enough time considering whether each number is a metric that you can proactively manage to or simply an interesting fact.
 
Richard Snow from Ventana Research, in his recent blog post Have Service Level Stats Outlived their Sell-By Date? conducted a study of the most common metrics use in contact centers.

 
Why are metrics least relevant to the customer experience the most common?  As Richard accurately concludes: because they’re easy.
Don’t do easy.  Do what makes a difference.

Comments

  1. Some of these stats are still necessary and useful to track. Why? … Number of sales are necessary to track for financial reasons, as example.
    As Manager of Call Center I want to know how long each agent is on phone – however not for usual reasons. I look at the whole picture, as many long years ago, I learned that the differences in talk time with each customer has many reasons. So we have to then look deeper.
    In this case I discovered that the agent with longer talk times had higher first call success rates and customers reported higher levels of satisfaction with this agent – in fact requested him when calling in to the Help Desk.
    How did we track – this was in 1996, so we had not fancy CRM tools, or even an Access database. We tracked on paper, and I contacted customers directly to get feedback, as well as reviewed CSR Notes :) Very archaic in today's standards.
    We didn't have IVR's or call tracking software, and probably had a better handle on what we were doing and how our customers felt than now with all the analytics. Why? Because we talked to our customers.

    I believe the lesson to be learned is that analytics can be useful in many ways and for many reasons, but don't forget the people factor and to look beyond the numbers.

  2. When I was still on the phones taking customer calls I rarely met the call time targets called for.
    I did however do everything that needed to be done for that customer and for the company to make sure all questions were answered, even the ones they didn't know to ask, and therefor not having to call back.
    That being said, even with my longer call times I still had the highest ranking in the center and merited the highest pay grade.
    I am now in a supervisory position and make sure the new hires know that short call times are not the be all and end all to customer calls.

Speak Your Mind

*