How Many Calls Should Your QA Analyze?

I spoke a few weeks ago at the LOMA conference on Customer Service. LOMA is a great organization that caters to the insurance and financial services industry and my workshop was about Avoiding Common QA Pitfalls.” I’m always interested in what I learn from these conferences. You get a feeling for the hot issues in call centers.

The question that seemed to raise the most discussion at LOMA was “How many calls should I score and coach per person?” A book could probably be written on the subject, but let me give you a couple of thoughts based on our group’s experience.

Are you using QA results in performance management? If you are, then the question really needs to be, “do we have enough calls to be statistically valid and hold up to scrutiny?” If you are giving any kind of merit pay, incentives, bonuses or promotions based on their QA scores, then you’ll want a valid number. Assuming your QA scorecard has a valid methodology (which is a big assumption based on the fact that most QA scorecards we audit have major problems with their statistical validity), you’ll want at least 30 randomly selected calls. More is great, but there is sort of a rule in statistics that once you have more than 30 of anything, you’ve got enough that you know they can’t all be outliers. Let me say again, I’m talking minimums here.

The “Wait ’til Mom & Dad are Gone” Syndrome. Many call centers coach each agent religiously once a week. That’s fine from a feedback point-of-view. But like kids who wait until they see their parents pull out of the driveway to start the party, agents often know that they only have to watch their service until they’ve been coached for the week. After that, all bets are off. Sometimes a seemingly random coaching schedule that keeps agents guessing is a good thing.

It might depend on the agent. In our politically correct world we are conditioned to do the same thing for everybody. Yet, some agents need little feedback or coaching. Score the calls, make sure they’re still stellar, and then let them know their scores and give them their bonus.

Why waste time, energy and money coaching them? That’s like the guy who washes his car everyday whether it needs it or not (then parks it diagonally across two spots in the parking lot…I hate that guy!). Seriously, the number of coaching sessions is a separate issue from how many calls should you score to have a valid sample. Spend your coaching energy on agents who need it the most. It even becomes an incentive for some agents who dread the coaching sessions: “Keep your numbers up and you don’t have to be coached as much.”

From the discussion I had with some QA managers at the LOMA conference, there were several who – in my opinion – were coaching their people more than was necessary. We’ve seen agents greatly improve performance with quarterly and even semi-annual call coaching. Still, that’s not going to be enough for other agents.

There’s the challenge for you – finding out which agent is which and tailoring your QA process to meet each agent’s needs.

Technorati Tags:
, , ,

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: