Key features used
- Demand Forecasting
- Automated Scheduling
- Capacity Planning
- 95% forecasting accuracy
- Improved demand and supply fit
- Happier employees
Roadchef is one of Britain’s leading roadside service area operators with 30 locations across the country, hosting several brands including McDonalds, WHSmith, Costa Coffee and Leon. Operations Planning Manager Iain Thomson knew they needed to harness the power of AI to make highly granular forecasts on brand level, automate their scheduling and plan well ahead of time to make both short and long term strategic business decisions.
We spoke with Iain Thomson, Operations Planning Manager at Roadchef, about their demand planning challenges and harnessing the power of AI to make highly granular forecasts on brand level, automate their scheduling and plan well ahead of time to make both short and long term strategic business decisions.
“Roadchef is not a rinse and repeat business. There are definitely trends that you’ll be able to see year on year, week on week. However, it can be a complex challenge to cater to high arrivals of customers wanting to use our services and brands without prior notice. Being aware of those nuances throughout the day and the week helps us to plan labour and resources to meet that demand at that specific time prior to that arrival rate” explains Iain.
One of the key aims for Roadchef is to make sure that when customers arrive, they are sent on their way feeling a lot happier than when they arrived which, according to Iain, is not always a given. Roadchef has to deal with highly fluctuating demand levels and make sure that their services and environment are completely ready for arrival. Demand can range from high volumes of customers using different brands in very short time windows to very low demand during certain hours – especially for 24/7 stores. Having different demand levels also means having different implications for schedules in terms of shifts and breaks. But it isn’t just about labour, they also have to make sure that fridges are full for the different restaurants and cafes, shelves are stocked in retail and grocery stores, and large seating areas are available.
The problem was that the details of the forecasts didn’t go as far down as they wanted to nor were they differentiable per brand. This lack of granularity refrained managers and senior teams from planning on that level. On top of that, they didn’t want managers to spend 100% of their week looking at forecasts and making schedules but instead focus on the customer experience and operations.
Building up further, Iain notes: “As much as we want to deliver on customer service, we also have to make sure we have the right deployment of costs and keep the profitable element of the business. One of the big things we’re looking at is the cost control across all our sites. We want to give our site managers the financial ownership of the business.” And with customer service and cost control in mind, Roadchef selected Quinyx to help with their forecasting and workforce planning.
Brand-specific forecasts and schedules
When customers arrive at a Roadchef site, there are a number of brands and franchises to choose from. From McDonald’s to Costa Coffee and Leon to WHSmith, each and every brand has its own operating methods and focus on what they want to deliver for their customers. Thus, every brand needs a different approach to forecasting and scheduling.
Iain explains Quinyx’s role in this: “Quinyx was very well aware of the fact that we’re not a standard business and a one-size-fits-all solution wouldn’t work for us. Using their hyperlocal forecasting functionality, we are able to configure all key labour rules and labour metrics that are different across our brands. Doing this allows us to make very accurate sales, transactions, and labour deployment forecasts for each individual business down to 15-minute intervals.”
This level of forecasting granularity and accuracy is important to Roadchef as it gives the managers more autonomy on scheduling within agreed labour costs, budgets, and forecasts. Since a lot can happen on a day to day basis, having that labour forecast and schedule a month in advance really gives the managers the confidence to implement a degree of flexibility within their team and have employees plan their life better.
However, It’s not only the managers or the short term planning that benefit from Quinyx forecasting. “The data that has been delivered back to us was in a format we’re comfortable with. We’ve been able to digest, analyse, and turn that around and share that with key stakeholders like the teams, executives, and investors. It allows us to talk about upcoming sales trends, looking at last year’s performance, trends, and run rates across different businesses which is all rolled up into a high-level sales forecast that we use for the financial planning and budgeting for the next year. “ Iain explains.
95% forecasting accuracy
Roadchef has consistently received forecasts above 95% accuracy for the next 7 days to 5 weeks out. When it comes to forecasting more weeks out, however, it becomes a bit more tricky. Despite that, Roadchef has never had the accuracy drop so far that would trigger an additional headcount or labour deployment.
Iain recalls one example where the demand forecast fine granularity showed its value: “All the experience and all the knowledge that we had, told us that the highest demand peak in the day has always been around 12 pm but in reality, we saw the demand start to significantly increase from around 11:15am – just as the forecasts showed. Seeing those nuances gives the site managers the confidence to actually plan and meet that demand.”
Improved demand and supply fit
The very nature of Roadchef’s motorway network is that they are often located between final destinations. There’s no guarantee that a site is close to a city or town, which can cause recruitment challenges in these areas as people feel reluctant to travel 20 to 30 kilometers for work. Roadchef used the granular forecasts to fill these gaps in unexpected ways.
“We introduced a subsidised mini-bus transport service to collect people from those remote areas. Having those forecasts really helped us to schedule those hard to recruit and hard to schedule shifts to meet demand. Now we know that the shifts we need on a certain day are earlier or later and a certain length. That certainty and the fact that we don’t have to pick up people at 7 in the morning and let them stay until 6 pm makes the job more attractive and easier to schedule employees and transport. It wasn’t one of the benefits we foresaw, but it just shows that there are more elements that we’ve been able to improve.”
The question always begs: “Are we able to deploy our labour better without increasing costs?” Another upside of the new approach is that Roadchef is now able to plan more efficiently while staying cost-neutral.
That cost-neutrality has also helped cater to stakeholders in another way. When asked about the acceptance of schedules, Iain explains that the new schedules have been well received. “Employees don’t have standard 9 to 5 jobs here, but we can make schedules that incorporate certain wishes and still meet demand well. Just an example is that we’ve been able to offer shorter and longer shifts than your typical 7 to 8 hour day in some areas, which allowed employees to combine their schedules with their lives better. In all honesty, the general feedback of employees has been hugely positive”