Improving scheduling at your shops starts with something you should have at your fingerprints: historical data. Unless you’ve radically overhauled your production system in some way, data from your last year or two will show you the amount of work your shop actually produces, rather than what you just THINK it produces.
For each month of your historical data, add up how many jobs (or “units”) you actually completed. Also divide this number by the number of work days in that particular month. Run these numbers for at least the last 12 months, and to get a good basic average of how many units you actually produce per-day, per-week and per-month.
The other number you should pull from your historical data: How many of those jobs on average each month were unexpected tow-ins or drop-offs.
Running these numbers for more than one year also can help you spot seasonal differences. If you’re in the Northeast, for example, you know you’re likely going to have more tow-ins in December than you do in August.
That’s really all you need to get started. Now take out a calendar page for next month. You know how many jobs on average your shop can produce for each week of that month, so that’s how many jobs you should schedule in for each week.
But you also know how many tow-ins or drop-offs to expect, on average. Will this vary a bit, week-by-week? Of course. But to not allow space on the schedule for those, when you have historical evidence of how many to expect, is just foolish. If you’ve typically seen an average of eight such jobs a month in the past, put two a week on your calendar, and schedule your drivable vehicles in around those. If those tow-ins show up, they won’t create chaos and missed deadlines. If they don’t show up, you can always call a customer scheduled for later to see if they want to bring their car in early. Few customers would ever complain about that.
I’m sometimes asked if looking at just average “units per month” is sufficient for good scheduling, or if shops should look at the number of labor hours on those vehicles, or the type of hit (light, medium or heavy). I’m a believer in just sticking with car-count. Unless a vehicle has been fully blueprinted, the number of labor hours on a preliminary estimate isn’t likely accurate enough to help with scheduling. An average number of units per month, if derived over a year or more, will include a mix of small, medium and heavy hits. It’s a simple system that likely will get you a lot closer to efficient scheduling that most shops currently are.
But here’s the other important aspect of better scheduling. Stop bringing everything in on Monday morning. Think about the stress that puts on your administrative staff, having all those cars arrive at nearly the same time. Meanwhile, your paint department may be staring at the four walls of an empty paint booth – until later in the week when that whole glut of vehicles is ready for paint.
Instead, use the historical average daily number of units your shop produces, and work to schedule in a consistent number of those each day of the week. I know what you’re thinking: Insurance companies won’t let us schedule work in on a Thursday or Friday. That’s just not true. Some cars are going to carry over a weekend regardless of when you bring them in. I can assure you some shops are scheduling work in later in the week. Not with every car and every insurer, but enough to help improve scheduling and flow.
It also may help to get some professional sales training to help your frontline staff have effective conversations that enable them to determine if you risk losing a customer by scheduling them further in the future, or if a particular customer would be fine dropping off on a Thursday or Friday.
The beauty of all this is you are bringing into the shop the right amount of work for what your system can handle rather than bringing in vehicles that just sit. Once you have this dialed in, you and your employees can then focus on making other improvements to your processes that will allow you to schedule in and produce even more jobs per week and month.
