SCO Principals

Call Center System Capacity Optimization (SCO)

Home/SCO Principals

Truth, Relevance & Reliability

SCO elevates any commercial or home grown WFM solution to a position of unprecedented integrity. It does this by removing the long standing barriers that are intrinsic to forecasting with call counts.

Naturally, you would like:

  • Truthful data
  • Forecasts and schedules that are relevant to your objectives
  • Reliable measurements of success
  • Reliable indicators of any need to alter course.

SCO removes three key barriers that hinder any WFM practice that relies on traditional interval base metrics. Most will recognize that every conventional WFM solution measures historical demand using 15 minute call counts and 15 minute average talk times. These are interval metrics.

So what is the factual basis behind stating that every interval based WFM solution is on the wrong side of three critical barriers:

  • Truth
  • Relevance
  • Reliability

These WFM limits are surprisingly easy to understand!

The Truth Barrier

For decades, virtually every WFM practice has forecasted using “answered and abandoned calls”.  The practice was first popularized by books like “Call Centers at the Speed of Light”, authored by Brad Cleveland. To be generous, the author might be credited with indirectly recognizing that forecasting with true offered call counts is not feasible.

However, the key word is indirectly.  Readers were never actually taught about the perils of true offered call counts, nor were they given a working alternative.

Instead, they were just told to forecast with the answered plus abandoned call counts that are collected each fifteen minute interval.  Readers were told that answered plus abandoned calls is equal to offered calls.

Unfortunately, the 15 minute counts of answered and abandoned calls are never equal to offered calls.

This is the essence of the truth barrier.  Every WFM practitioner who has been using a spreadsheet or commercial WFM software to forecast, has been forecasting with statistics that are never true.

It’s a big problem. The worst part is that the knowledge of the problem has been omitted from what planners are taught.

So lets get you caught up on why the industry has asked you to ignore the true offered call counts (that don’t work at all) in favor of statistic that are always false.

True offered call counts expose the full instability of interval data.  Interval data can only count each call in a single interval. Yet even with very short calls, 30 to 60 percent of calls will arrive in one interval and carry over into the next. The ignorance of that carry over effect is doubly problematic.

Work that belongs to the later interval is striped away and incorrectly associated with the interval of arrival. One interval is robbed of estimated workload while another is stuffed with grossly exaggerated workload estimates. To forecast with true offered calls is impossible. The large arbitrary misplacement of anticipated workload results in a forecast that jumps erratically above and below appropriate staffing levels.  It’s a terrible forecast to schedule to.  It does not work at all.

Answered plus abandoned call counts are not a solution.  It’s just a convenient but primitive mechanism for blurring interval data into a pattern that traces the prevailing staffing levels.

When your forecast tells you to staff pretty much the same as you did the last time no one needs to stick their neck out to defend the forecast.  The forecast looks plausible so it slips under the radar of proper scrutiny.

The difference between “true offered call counts” and “answered plus abandoned data” is obvious.

True offered call counts reflect the volatile rate at which calls arrive in the queue. Answered and abandoned call counts reflect the regulated rate at which callers leave the queue.

To the extent that arriving and leaving are opposites, “offered calls” are the opposite of “answered plus abandoned”.

Answered Plus abandoned data does not measure demand. They only measure the prevailing capacity to answered calls plus the relatively small rate at which callers abandon.

So what ever your demand is, answered and abandoned call counts will reflect your current call answering capacity plus or minus 2%.

So take a moment to come to terms with several contrasts between what you were taught and what you are learning today:

In the past, you were taught that WFM was preparing offered call forecasts. Instead, it has been giving you what is essentially an answered call forecast. WFM forecasts have never prepared your call center to address demand. WFM has only prepared your call center to repeat the last staffing pattern—no matter how poorly that pattern addressed the evolving needs of callers.

So while you would like to forecast using some semblance of the truth, answered plus abandoned calls is never the truth.

While you would like to forecast in a manner that helps you to keep pace with changing demand, you have actually been forecasting with a method that inextricably ties you to old outdated staffing patterns.

What you want to forecast with is the truth.   Interval forecasting can never give you the truth. That’s the truth barrier.

SCO gives you a truth based forecast overnight. That truth has nothing to do with any type of call count. The overnight improvement will surpass you wildest expectations for how much you thought your business could be improved.

The Relevance Barrier

When you forecast with answered plus abandoned call counts, the forecasts loose all perception of wait time.  Such forecasts have no relevance whatsoever towards improving wait times.

Answered plus abandoned data pegs every call to the interval when the caller’s wait is over. This in no way prepares you to answer call arrivals. It only prepares you to answer calls at the same historical rate that you have always answered calls.

If there are just a few intervals where you customers typically wait too long, the WFM forecast knows nothing of it. Those intervals will never improve.

If customers wait too long in every interval, the WFM forecast will ignore it and perpetuate the negative business impacts of under servicing the customer base.

How does SCO make a difference? That’s easy to understand. First, imagine planning without caring at all, about how long your customers have been waiting. That’s what WFM has always done for you.

Now, imagine planning for tomorrow with perfect attention to the wait time optimization of each conversation.  That’s what SCO does. The difference is black and white. In practical terms is can mean cutting you wait times by 64% without hiring any more agents than you have today.

The Reliability Barrier

Call center managers rely on metrics to determine if their call centers are on course to satisfy the needs of customers and to support the growth of the overall business.

The popularized WFM metrics are unreliable. More to the point, the metrics are misleading and untruthful.

Daily forecast accuracy is commonly relied upon to validate that the forecasts were sound. One would think that unacceptably long wait times would lower the score something called “forecast accuracy”.  Sadly, this is not true.  Daily forecast accuracy is just the percent difference between the daily totals of the forecasted calls and the actual calls.  If the forecast is for 4000 calls and the total calls offered turns out to be 3600, the daily forecast accuracy is 90%.  No matter how long callers wait, its 90%.  Even if staff were timed so poorly that every caller waited for hours, the daily forecast accuracy would still be 90%. Daily forecast accuracy lacks relevance. The purpose of the forecast is to minimize wait times yet forecast accuracy as a metric is completely insensitive to wait time problems.

Interval forecast accuracy is equally misleading.  If you prepare to answer 300 calls in a given interval, you will inevitably answer roughly 300 calls.  It’s the maximum that the forecast has prepared you for.  If the forecast predicted only a fraction of the actual demand, you’d think that the forecast accuracy score for the interval would reflect the inadequacy of the planned staffing levels.  Not so.  You will answer roughly the 300 that you were prepared for.  Answered plus abandoned calls will be within 2% of the forecast. Thus an exceptionally underprepared interval easily generates forecast accuracy of 98%.

How can managers manage effectively when WFM always says it was accurate no matter how inadequate the plan actually was?  This is the WFM reliability barrier.  The metrics are made up to look good no matter what. That will never help a call center to improve.

SCO eliminates the reliability barrier with the best technology available—the truth.  SCO intricately interprets the second by second truths of each call center’s historical demand. It’s a truth that takes any call center to the pinnacle of its productivity and efficiency. And why shouldn’t it.  The truth works so much better than just making stuff up.

Accessing Planning Integrity

The forecast is the foundation for everything that follows. If the forecast is untruthful, irrelevant and unreliable, those deficits carry through to the schedule. Poor forecasts and schedules cause long wait times and low labor productivity.

The good news is that any WFM practice can break through all three WFM barriers just by switching to an SCO forecast. Those forecasts convey integrity into the entire planning cycle. You don’t need to discard your WFM. Just enable it with an SCO forecast.

The SCO forecasts guides your WFM to produce schedules that serve your truthful business needs. You get consistently low wait times and high labor productivity. Any need for increases or decreases in labor become transparent. Any SCO recommended hiring serves profitability. Any SCO recommended staff reductions carry no wait time penalty.


Beyond WFM Barriers

With SCO, the forecast teaches your schedule how to time staff for real results. There are no illusions. The same number of staff that you have today are moved into more effective patterns. Consistently low wait times allow call center staff to focus on satisfying customers. Satisfied customers are less expensive to service and they help your business grow faster. SCO understands that growth so your future forecasts and schedules keep pace with the truthful needs of your business.



Leave a Comment