In order to establish a rationale for deploying any workforce automation system a four part thought process should be considered:
All efforts around assigning and managing work and work orders must be focused on meeting the minimum SLA of a customer, to be there when promised.
With Enough Time.
Placing the burden on the technician to manage the day will result in rushing to the next job. Work must be managed to ensure that enough time is given to do a job correctly.
With the right skill.
Sending an unskilled worker to a trouble call or install, guarantees failure in the form of an incomplete install or TC, a repeat or service after install trouble call, as well as calls into customer care, etc.
With the right equipment.
Algorithms that are in place for assigning work take many things into consideration, but rarely take into account the CPE that resides on a technician’s vehicle.
In the next few posts, we will explore a number of business operations elements to automated workforce management solutions. We will look at what steps you need to take to make sure you are maximizing the potential of your WFM platform and driving superior ROI performance.
There are three main pillars that lead to operational excellence; people, processes and technology. We will look at the common pitfalls company’s face in each of these areas and will take a look at Total Cost of Ownership as a means to understand those struggles. We then explore the right metrics for trend analysis that every operator should look to in order to create a continuously improving service delivery environment.
In spending time with telecom operators I’ve seen an interesting phenomenon related to continual changes in the industry, along with the constant need to reduce cost and be more efficient. That is, that whatever we have now isn’t good enough, so we must buy something new to replace it. Yes, there may be certain legacy issues that keep a software from being as fast as a new version, or even valid functionality gaps, but the alarming observation I have is that operators don’t fully utilize the systems they have invested in to achieve the benefit they were looking for when they bought them.
There are many complex applications in today’s modern enterprise, all intended to drive efficiency and profitability through automation in the company’s key functions including construction, marketing, sales, field installation and repair, network management, and customer care. The cost of replacing even one of these systems can be staggering. And then too often we are back in the same boat, looking for something new to solve the problem.
Near and dear to my heart, for example is the automated workforce management (WFM) application. A typical scenario that regularly plays out among operators is to look at newer solutions for complex workforce management systems because the installed one isn’t doing what it promised. In many cases, there’s a reason (or reasons) that the company may not have taken into consideration.
Five years ago, maybe 10, when the WFM application was installed, it was thought of as a panacea for all service related issues, on par with solving world peace or traveling to Mars. There were great expectations and for the most part many of the benefits were realized. This is related to the focus and attention given by the leadership team who bought it. They were on the hook for operational improvements and made certain that every user was trained properly to get the most benefit for the company.
Over time, priorities shift or leadership changes occur (leaders move on to new roles, new companies, maybe even out of the industry), and the focus may be lost. Sometimes the application is acquired when companies merge and its not understood. The next generation of leaders and users move in, but may not have the same interest as the original team. Training may move from formal to casual, in that the next generation of users are trained by sitting next to an experienced user and learning by observing. In this case, they learn only the elements that the “trainer” shows them and may miss other areas that can help them to do their job in a more efficient and automated manner.
As time goes on, the powerful capabilities of automation are eroded. I’ve observed companies using spreadsheets and tick marks to track things that are all automated, if only those features were used properly. The end result of all this is that the application that was once the solution is now the problem! The next logical step therefore is to replace it with a new one. A new one that does exactly what the existing one does. It will have a different user interface, but it does the same thing as the one that exists.
Has any business decision maker ever encountered a vendor who had a solution that “increases OPEX?” I’ve never met one. When the new product is presented, somewhere in the presentation there is always a promise to reduce OPEX. The new application that is replacing the existing one will promise to do so. But can it really reduce operational costs more than the existing product if it does exactly what the current application does? Is the cost of disruption, customer impact, and re--‐training on new tools accurately estimated? Are the expectations of the new system realistic? How quickly will the new solution become the old solution in need of replacement? In my experience helping operators work through these issues, costs are often underestimated and the expected benefits are inflated.
Re-Investment for success
I have a different thought. In many cases it can make more sense to re--‐invest in the employees who use it. Pay for experts to train and re--‐train the users to be absolute in--‐house experts. A few weeks of business process analytics from the vendor will uncover all the areas where simple re--‐ invigoration will once again allow the operator to realize the results that were intended. The choice for the operator may be to spend $100,000 for a well managed re--‐start, vs. $1,000,000 to replace the tool and achieve exactly what was already possible. That is money better spent.
In order to make such tools culturally significant, they must be used as they were designed. For example, with WFM tools techs, dispatchers and managers must be thoroughly trained and intimately familiar with proper use. The best way to ensure that is through compliance management. Measure usage by each individual to verify that they are taking advantage of all it has to offer. Measure tasks, feature usage, and reductions of unnecessary activities (like a tech calling to provision a device that can be done using the tool).
Finally, all these things are intended to save time. They don’t. We cannot save time, only spend it. Looking at how the workforce is spending its time may be an indicator that a tune--‐up is needed.
Taking a look at completion percentages, including cancellations and reschedules, will yield another view
into a company's ability to keep the workforce productive. Many cable operators report average
cancellation percentages in the 20 percent to 30 percent range. With a few weeks of tracking to verify, one
could determine that overbooking by a similar percentage will let a dispatcher fill the "hole" created by a
cancellation so that a tech is not left unproductive. Minimally, ensuring that all available quota is used (96
points per tech) will minimize this issue, but a dispatcher should be continually looking for jobs in the
schedule pool to bring forward should a tech have an opening because of a cancellation.
Putting all these data points on a tech scorecard is a great way to give feedback to the workforce and make
smart decisions to improve productivity, plan workforce schedules, and be better tied to the individuals
who may require coaching. Figure 1 is a sample of a tech productivity scorecard.
On a day-by-day basis, and then by weekly average, a supervisor can now manage better and know how
well the field workforce is using its time. Metrics to watch include:
1. Average service time: Based on the start and stop times the technician reports and entered in military
(24-hour) time by a dispatcher
2. Average drive time: Time between end of one job and beginning of the next
3. Lunch duration: Largest gap
4. Start time: Time of arrival at the first scheduled job of the day
5. Finish time: Time the last job was completed each day
6. Points per day: Points allocated to the tasks assigned to each tech
7. Off dispatch: Large gaps that cannot be explained, like long drive times; usually indicative of reporting
issues between dispatch and field
8. Data quality: A check to make sure that times are input in military fashion
9. Work time: Actual hours being productive (not an HR reporting number)
10. Points per hour: Total points divided by productive hours
11. Percent hourly efficiency: Using the full hour in scheduled activity, including drive time
12. Daily efficiency: How well the whole work day was utilized; 88 percent efficiency would be a pretty
By putting this discipline in place and by viewing your work staff with good data, you will be well positioned
to go the next step and automate those best practices.
Another method to gauge work force activity, availability and productivity is analyzing much time spent
driving, in a customer home, at lunch and so on. This can be done by looking at the time stamps in the
billing system. Occasionally, this requires some process improvement in the tech/dispatch relationship.
Ensure that techs update instantly and a dispatcher inputs the correct time stamps. If a tech leaves a job at
9:15 a.m. and starts the next job at 9:30, it is logical to assume 15 minutes of drive time between those jobs.
By looking at these numbers, an average drive time can be determined, per tech. It is fair to assume that the
largest gap between jobs is lunch time.
By looking at when installers arrive at their first job each day, a manager can make decisions on how to
ensure that the field force is on the road and working as soon as possible each day. Seeing that in many
cases, the time of arrival at the first job is up to one and a half hours after the start of the shift caused one
MSO to analyze a couple of things:
1. The customer premises equipment (CPE) issuance process: Many techs spent an inordinate amount of
time standing in line to turn in or receive CPE for each day's work. By creating and using a locker system,
or other efficiencies, this backlog can be eliminated.
2. The morning meeting: It is generally desirable to meet with the workforce each morning for a daily
briefing, but these meetings can sometimes take on a life of their own. By limiting meetings to once or
twice a week, techs can be expected to be on the road sooner.
Supervisors should not be put in the position of having to look all around the office to make sure techs are
on their way to the first job. Rather, by requiring techs to report to dispatch that they are en route, dispatch
can report a tech who is not on the road when expected, allowing the supervisor to use data to coach the
Points are used to allocate time to a task in the field. There are a number of ideas for the value of a point,
which can be set up in the billing system. Consider that a point should be worth 5 minutes. Markets that set
a point value of greater than that can "strand time." If a point is worth 15 minutes and a task takes 20, a
scheduler will assign 2 points to the job, and then 10 minutes of time are stranded.
Because not all jobs are created equal, points become the best way to measure productivity in the
workforce. In an eight-hour day, 96 points of time would be available to schedule a tech or installer. The
first metric to evaluate is points of work per day, per tech. This is the first measure to consider whether you
do a good job of scheduling, routing and assigning. It is also a good means to understand your capacity in
workforce for planning.
Next, after points per day, a simple calculation will show how productive a tech's day actually is. By taking
the total points per tech per day and dividing that number by the hours worked, a "points per hour" metric
can be determined. Why is this important? Say a tech performs 96 points in a day. On the surface that looks
great, but what if it took the tech 12 hours to complete those 96 points? Not too efficient after all, is it?
Conversely, a tech performing 48 points in 8 hours is only 50 percent "efficient" for the day.
These two metrics will allow a supervisor or manager to understand how well work is being assigned and
how well the field force is being utilized
Automated workforce management systems have matured to the point where there are great proven results
in productivity improvements. They are not, however, a panacea that will cure all the ills of supervising a
field force. As a matter of fact, they may only serve to automate bad habits.
Here's an analogy: If I went out to the golf store and bought a $500 driver, expecting to go on the pro tour, I
would be sadly disillusioned. The truth is, my swing is really bad. I should be interested in buying that
driver only after I've got everything else working right and when my desire is to shave two or three strokes
off my game. So it is with automated systems. They will certainly "shave strokes" off your game and make
for a much improved operation, but only if your operation is already sound. Consider a few things that can
be done in a non-automated fashion to improve performance and productivity.