Off Center
If the key call center metrics were to form a rock band, Forecast Accuracy would most likely be the bass player – less flashy and famous than its fellow members like C-Sat, FCR and Service Level, but no less critical for an effective performance.

Forecast Accuracy is sometimes referred to as “forecasted contact load vs. actual contact load”, but only by managers who like to make things more painful than necessary. The metric shows the percent variance between the number of calls (or chats) predicted to arrive during a given period and the number of contacts that the call center actually receives during that time. Most managers consider a 5% variance to be acceptable, though they naturally shoot for better (a lower %) than that. Those that regularly achieve a 15% variance or worse are sent directly to workforce management prison.

Missed it by That Much

So how exactly does one go about tracking Forecast Accuracy?

I’m glad I asked.

Call centers can retrieve data on forecasted contact load from whatever system or tool they use for forecasting (e.g., the center’s WFM system or Excel spreadsheets), then compare that to the data on the actual contact load received, which comes from the center’s ACD, email/chat management system as well as other report sources. The best call centers report forecast accuracy at the half-hour or hour interval level, rather than across days, weeks or months, as interval-level tracking gives a much clearer view of how horribly you botched the forecast.  

Accurate forecasting is paramount in any call center that gives a darn about customers, agents and cost efficiency. Without a measure in place to gauge the effectiveness of the center’s forecast, under-staffing can often occur, causing queues to fill with furious callers, furious callers to verbally eviscerate innocent agents, and innocent agents to throw fists through expensive equipment. Of course, all of this adds expensive seconds and minutes to wait and handle times, causing irritated executives to cut budgets and rescind their promise to add a window in the call center. 

Inaccurate forecasting may result in costly over-staffing, as well. And while this may make customers happy, it will certainly ire senior management -- as well as give agents too much free time between calls to think and figure out that they could probably earn more money making balloon animals for children in the park.

If you don’t have anything nice to say, share it in the comment box below.

Nice comments are also accepted – after careful review.

The queue is finally empty
The customers are calm
The reps no longer gasp for breath or cry out for their mom

The ACD did not explode
As many thought it might
It routed all 10,000 calls to agents out of spite

The supervisors now can coach
And work on raising scores
Though might as well forget the reps who’ve sprinted toward the doors

The manager’s exhausted
But has yet another task
To calculate the C-Sat rate – that’s why he has a flask

The call center is quiet now
It’s rare to have such peace
It’s the kind of calm that comes when reps just press “release”

So close your eyes, oh weary ones
You day is now complete
Worry not of calls and chats – you’ll soon be screwed by tweets

In clinical trials, this poem successfully lulled 87% of call center insomniacs to sleep. If, by chance, you continue to struggle with insomnia even after reading this poem, talk to a licensed physician about medications and supplements that can aid sleep. Or, if you need IMMEDIATE relief, simply have a conversation with somebody from Accounting or Legal.  

Sending a new agent straight onto the phones following just a couple weeks of classroom training is the equivalent of sending an aspiring boxer to fight Mike Tyson following just a couple weeks with a punching bag.

In both cases, the rookie is going to get knocked out, and their ear chewed off.

Nevertheless, many call centers continue to throw their new-hires into the customer contact ring well before the reps are ready to do battle – then act surprised that their average rep mortality rate is less than two months. These centers have fooled themselves into thinking that a week or two of lectures, role-plays and e-learning exercises is enough to prepare new agents for the unique demands and challenges that come with the frontline territory.

In contrast, most world-class call centers I’ve worked with have built a “transition training” component into their new-hire training program, thus enabling rookie reps to ease into a life of chaos and panic on the phones rather than diving straight into things.

What Is “Transition Training”?

Transition training entails having trainees handle basic calls in a controlled environment after they have woken up from classroom and other types of didactic training. In some centers, new-hires may enter the transition training bay after one week in the classroom; in other centers, they may not enter the bay until after they’ve completed two or three weeks of classroom instruction. Once agents enter the bay, they are routed a small number of calls that – based on the number dialed and/or the IVR menu option selected – should be relatively easy to handle.

Smart call centers ensure that there are plenty of supervisors or team leads on hand in the transition training bay in case a call turns out to be complex or a trainee turns out to be terrified. Where a typical agent-to-supervisor ratio on the official phone floor of a call center is 15:1 or 20:1, the agent-to-supervisor ratio in an effective transition training bay should be in the 3:1 to 5:1 range. Many small call centers that don’t have the luxury of a large number of supervisory staff to assist trainees often turn to their top agents to lend a hand in the transition training bay. Such an approach is great for building peer camaraderie, and for helping experienced agents learn to be bossy.

In most call centers, trainees return to the classroom following the first transition training period (which may last anywhere from a few days to a few weeks). This enables them to close performance gaps uncovered while in action, and to learn new skills and information that will allow them handle more complex call types – which they will get to do during the second transition training period. By the time they complete the second period of hands-on practice, most agents will be ready to graduate to the official phone floor or, if they have shown a particularly high level of talent, to be stolen from the call center by Sales or Marketing.   

It’s a Win-Win-W… It’s a LOT of Wins

With a carefully implemented transition training program in place, everybody wins: New agents’ gain more confidence and lose fewer lunches; veteran agents (who assist in the training bay) enjoy a vast sense of empowerment and superiority; and the call center as a whole saves a ton of money by reducing early turnover and the number of body bags needed on the phone floor.

If your call center uses a transition training component as part of its new-hire training process (or if you’ve ever helped implement such an initiative), feel free to share some of your experiences in the comment box below. If your call center does NOT do transition training, feel free to share some photos of your trainees crying their first day on the phones.

Just because your call center surveys customers and occasionally even looks at the feedback they provide doesn’t mean you have a “Voice of the Customer” initiative in place. A true VOC program entails continuously and carefully analyzing customer ratings and sentiment, identifying trouble spots and trends, and taking decisive action before your customer base starts to hate you as much as your agents do.

If your call center is as serious about the customer experience as it is about low wages and bad lighting, then you need to make sure that your VOC initiative includes the following special components:   

Tools that report whether the customer was using their “inside voice” or their “outside voice.” Naturally, you want to pay attention to any customer who provides negative comments about a recent interaction, but for prioritization purposes it’s important to distinguish between customers who are merely a little frustrated and those who are considering hiring a hit man. By investing in speech analytics tools that detect customers’ emotion/volume levels during calls and survey responses, it becomes easier to determine which customers to ignore, which ones to call back within the week, and which ones to kidnap immediately before they ruin your brand via Twitter.
“Fist of the Customer” (FOC) software. Sometimes customers don’t verbalize exactly what they are feeling, thus it’s important to have tools in place that can dig deeper and uncover hidden sentiment. While still very much in the testing phase, FOC technology measures how forcefully frustrated customers throw their phones or punch their computers when interacting with an agent or IVR. Equipped with special motion-detection software that I’m too stupid to understand or explain, a typical FOC solution can be programmed to send an instant alert to the call center’s recovery team whenever a customer’s punch reaches a “Mike Tyson” or “Jerry Springer guest” level of force.

A “Last Word” option for agents.  To avoid having your customers’ negative and abusive comments adversely affect agent retention and morale, it’s important to incorporate a VOA (Voice of the Agent) component into your VOC program. After receiving a scathing rating or comment from a customer, agents will likely want to retaliate and get the last word in after they stop crying. Let them do so by providing them with what they think is the customer’s phone number but is really the number to a crisis hotline where operators are used to enduring profanity-laden diatribes from complete strangers.   

NOTE: If you found Greg’s “Voice of the Customer” recommendations to be insightful and valuable, you should consider seeking help from a licensed mental health professional. Contact Greg for referrals.