Newsletters

Customer Support:   (972) 395-3225

Home

Articles, News, Announcements - click Main News Page
Previous Story       Next Story
    
FADE -- How to Ensure Plan Success

by Ric Kosiba, Ph.D., Vice-President, Interactive Intelligence's Decisions Group - January 7, 2015

FADE -- How to Ensure Plan Success

Ric Kosiba, Ph.D.

Vice President, Interactive Intelligence’s Decisions Group

A long, long time ago, in fact in another millennium, I was a civil engineer. When working on a project—a bridge, for instance—we would try and assess the risk associated with the project succeeding (being on time and under budget) as well as the risk of the product succeeding (the bridge not collapsing). The process for both assessments was pretty much the same; the critical elements were listed and the probability of each element failing was estimated. Determining the overall probability of failure then became a straightforward calculation.

The elements of a bridge were the physical parts of the bridge—each abutment, each major joint, each beam, etc… Evaluating the risk of the overall failure of the bridge involved determining the probability of each critical component failing under different load scenarios (i.e. an earthquake versus a heavy snowstorm).

Similarly, evaluating a civil engineering project included evaluating the probability of success of each step of the project and their risks (mistakes in design and rework, weather holding up the project, etc…).

I was in a meeting with a large power company when the contact center executive outlined the very same thing, but instead this engineering logic was applied to the success of a contact center plan. It was a very interesting insight—the success of contact center management has four unique elements that can be evaluated for risk—and failure of any component would lead to a failure of the plan.

What are the main possible points of failure?

Facts: Correct, Clean, Accessible, and Useful Data

The first step to any plan is to develop a database of time series data, like volumes, handle times, sick time, attrition, probability of right party contact (outbound), etc. This data must be at the appropriate level of detail (hourly? daily? by center?), and care needs to be given to ensure that the queues are mapped appropriately to the right contact type and work hours are allocated correctly to each staff group. These are the planning facts; they reflect much of what we know about our contact center.

These facts can be used to generate other related facts—for instance, staffing arrival patterns compared to volume arrival patterns can give you a very good indication of how well your operation meets your customer demand. Different permutations of shrinkage compared to phone time, for instance, can derive a relative efficiency metric that demonstrates the amount of paid time to agent worked time. These historical metrics and derived metrics become the basis for the rest of the plan.

But it should be noted that gathering data—facts—is also our first point of possible plan failure. If we allocate agent hours wrong, if we map contact queues incorrectly, or if we draw out our multi-skill network incorrectly, we may use this incorrect data to develop inefficient plans or plans that do not provide the appropriate level of service.

Assumptions: Forecasts, Possibilities, and What-ifs

The next insightful point of failure the executive outlined was the failure of the plan's assumptions. A point to remember: our forecasts, even with the most mathematical rigor, are still (educated) assumptions. When developing forecasts, we are attempting to convert our facts (contact center historical data) into well thought through assumptions about our expected volumes, handle times, probabilities of right party contact, sick time, agent attrition, etc… It is important to know that the fanciest mathematical formula does not know the simplest stuff, like when marketing campaigns will hit. A smidge of information will always beat oceans of math.

But forecasts are not the only assumptions. Assumptions around acceptable service standards, new agent learning curves, center hours, marketing and sales efforts, and much more need to be considered, … and need to be correct.

If our business assumptions are wrong, our plans by extension will also be wrong.

Decisions: Resourcing the Center Network

The purpose of every contact center plan boils down to one major item: making sure that we have the contact center resources required to deliver an appropriate service standard. But associated with this is a series of decisions. When should we offer vacation or training? When should we hire each class? How many agents should we hire in which centers? Should we offer overtime at our seasonal peak? What is the best balance between hiring and overtime? Should we outsource? Should we transfer agents between staff groups? How should we design contact overflows? Each of these questions represents decisions that can change as new data is acquired, and hence new assumptions are made.

So how do our plans fail when making decisions? The obvious—making the wrong resourcing decisions—will lead to either too many or too few agents available to service the customers, which leads to either higher cost of service, or lower service.

There is also a real cost to not making decisions in a timely manner. When contact center facts change and forecasts are beginning to show a shift in expected assumptions, simply delaying a decision can result in a situation that guarantees service failures or cost overruns. Resourcing takes time. Inaction is a decision, even if it feels like it is not.

Execution: Adhering to the Plan

I am sometimes surprised by how companies decide to execute their plans. For instance, we've seen companies whose expensive planning teams spend a great deal of time and effort to develop well thought through plans, only to have them ignored when put into practice. A center manager may have very compelling local reasons to not follow a centrally developed staff plan, but decisions made in a local vacuum may lead to global service failures. Simply delaying a new agent hiring class can lead to significant shortages for the enterprise.

The best plans will, for instance, lay out the expected amount of undertime, weeks in advance. If the managers and supervisors are not held to their execution of these items of the budget, they will likely be ignored. The planning process will be for naught.

The execution of the plan is the responsibility of all involved, and if an organization makes one party responsible for the plan, but not responsible for its execution, then there is a risk the plan will not succeed.

It is always good to understand how any process is at risk of failure. For contact center planning, evaluating the Facts, Assumptions, Decisions, and Execution of the planning process is a good place to start.

Ric Kosiba, Ph.D. is vice president and founder of the Bay Bridge Decisions Group at Interactive Intelligence. He can be reached at Ric.Kosiba@InIn.com or (410) 224-9883.

 
Return to main news page