In a world full of on-demand services, customer expectations are at an all-time high. Regardless of the industry you serve, you’re competing with companies like Uber, Amazon, and Airbnb that offer their customers instantaneous service and real-time visibility into delivery.
So how can organizations that deal with complex tasks and constant disruptions, keep up with customer demands and meet business goals? Successful execution requires demand forecasting and planning, creating a flexible schedule, connecting with the field in real time, and measuring the outcome to continually improve the process.
In the last Decoding the Service Chain, we introduced the term Service Chain Optimization (SCO), a decision-making process that considers the full life-cycle of service delivery and works best with automation and artificial intelligence. In this week’s post, we further dissect each link in the chain—planning, execution, and analysis—and discuss the tasks involved.
Whenever you’re undertaking a complex task—whether it be a building a house or throwing a party—it usually starts with some sort of plan. The same goes for field service scheduling. You must first determine the scope of demand and then evaluate and distribute resources accordingly.
Building a schedule should rarely be done on the fly (emergency jobs are an exception) and won’t be successful without careful preparation—especially when there’s so much to consider. For example:
- How many jobs will there be on a given day?
- How many unplanned jobs do we expect?
- How many technicians should be on hand to meet demand?
- Will there be a lot of traffic? Or a big storm?
- What tools will be needed for a particular job?
- Where are all the jobs located, and which makes sense to go to first?
Before you can come up with a plan, you need to know how many resources (technicians, skills, tools, equipment, parts) you’ll need on a given day or timeframe. Demand forecasting is the process of predicting the number of tasks that will arise during a particular time period, and the amount of resources and time it will take to complete them all.
You might be wondering how, without psychic abilities, you are supposed to forecast demand. Look at your historical data to help make predictions (artificial intelligence comes in handy here). At this time last year, how many jobs were there? What was the weather like, and did it cause delays in the schedule? Were there a lot of storms or potential for emergencies? How much time and how many resources were spent on a particular job type in the past?
Once you can calculate demand for resources, it’s time to allocate these resources appropriately.
Capacity planning is the art and science of ensuring you have enough field resources to do both the scheduled and unplanned work on any given day. It’s easier said than done.
For instance, you could have all your field techs on call just in case demand peaks. But that means you’re wasting money on resources that you likely won’t use. Or you could schedule just enough techs for the planned jobs, but when an emergency job comes up you are forced to cancel on a customer.
The best way to cut costs while meeting customer demands and SLAs is to find the optimal amount of resources you’ll need through forecasting and planning. You’ll need to know how many techs will be needed, which skills will be required to complete jobs, and which parts and equipment will be needed. From there you can distribute the resources accordingly.
Once all the necessary planning is complete, you’re ready for execution. This includes appointment booking, scheduling and dispatch, and all work carried out in the field. Let’s take a closer look at some key elements.
Appointment booking is the process of agreeing to a service window with a customer. Though part of the execution stage, this usually occurs days or weeks in advance.
Many organizations (typically those who use manual processes) offer appointments based on a number of pre-determined slots for a geographic area or time window, until all slots are filled. This way it’s a simple first-come, first-served approach. However, this approach doesn’t consider the location of previous jobs or the true availability of workers and their skills.
Think about it. A tech is assigned a job in the west side of town and his next job is far east. Later in the day he’ll have another job on the west side. Does it really make sense to send him back and forth when both money and time could be saved if he stayed in one area, completed multiple jobs, and then went to the other side of town?
Scheduling is the most critical link in the service chain. Up until now you’ve been forecasting and planning with a degree of certainty, but now it’s time to match real jobs to specific times and mobile workers.
It can also be one of the most difficult tasks, because jobs aren’t scheduled in isolation. You need to consider the next job, it’s location and requirements, the next job, and so on. There will also be in-day disruptions such as emergency jobs or customer cancellations that you’ll have to account for. As a result, you’ll likely be moving things around quite a bit throughout the day. Manually reshuffling is very difficult to do effectively, so it helps to have a degree of artificial intelligence and automation on your side.
Mobile Field Execution
Managing and keeping track of a mobile workforce can be difficult because everyone is in different places at once. Being said, it helps to maintain a degree of real-time communication with both your workforce and customers. This could mean updating the customer when the tech is on the way, or sending them a reminder the night before. It could also mean keeping in touch with techs via mobile devices or tracking their location to make sure they’re both safe and on schedule.
The final link in the chain is where data from the day of service is collected and analyzed. Though it’s the last stage of the process, it doesn’t really end because the results are fed back into the forecast as fuel for continuous improvement.
Service organizations have been collecting data on all aspects of the chain, and applying it to measure key task types and KPIs. They are increasingly incorporating more data sources like weather, traffic, and customer demographics, and applying artificial intelligence and predictive analytics to get even more precise service delivery. Let’s discuss some examples:
Customer Service History and Demographics
Data about your customer’s history and demographics can help you predict the likelihood of a customer cancellation or no show. For instance, if the customer has a history of cancellations when they have an appointment slot in the middle of the day, you customer service team will be prompted to offer them an appointment time that falls outside of normal working hours.
Traffic and Weather Patterns
Tracking traffic and weather patterns enables optimal routing and travel decisions to save both time and fuel costs. For instance, if you know what time traffic has historically been the heaviest you can wait to send technicians into a particular area until traffic has died down. You can also keep track of how weather has affected travel times and traffic in the past. For instance, heavy snowstorms or torrential rain will likely delay travel so you can account for that ahead of time when scheduling.
Before you can ensure near-flawless service you must understand the service chain and how it works. Once you know what’s required for optimization you’re ready to take action. In the next Decoding the Service Chain we’ll discuss some methods to get you there.
For more field service trends and best practices, subscribe to the Field Service Matters Blog.