by Ben Savage, Chief Technology Officer

 

How order pickup data helps you avoid the hazards of manual takeout processes

How big is your blind spot when it comes to order pickup data for your off-premises orders?

 

How long is that order sitting on the shelf before it’s picked up?

 

Automated pickup solutions are a lot like the blindspot monitors new cars have …. But better because we can tie their order pickup data into analytics systems that help you see the pattern.

 

Why is order pickup data like a car’s blindspot detection system?

A lot like a modern car, an intelligent food locker can give you notifications of when something has changed state that you can’t see without leveraging its sensors. If you think of an older car with no sensors, all the burden of work is placed on the driver to check the blindspot and make sure no one has moved when you get ready to change lanes.

 

But in newer cars that have blindspot monitoring sensors, you can get an alert as soon as you flip your turn signal on or as soon as you start to change lanes that something is in your blind spot. This allows you the driver to spend more time planning what you’d like to do.

 

With an intelligent cubby, you can free your staff from having to manually keep any eye on the counter/shelf with your pickup orders to ensure they aren’t sitting for too long and creating potential quality issues. The intelligent cubby system can watch those orders for you by monitoring its sensors to track Dwell Time. For this conversation, Dwell Time is how long it takes from when an order is assembled for the customer until the order is actually picked up by the customer/DSP.

 

Based on a brand’s criteria, different alerts can be configured to visually alert the brand’s associates that an order has been sitting too long. Then they can remake the order or perform service recovery as needed.

 

The power of an integrated tech ecosystem

Let’s zoom out a level now and imagine that instead of a single driver in a single car, we can network the cars together. Specifically, the data from the different cars. So that now when the driver goes to change lanes the other cars receive this data in real time so that they can change their speed and allow for a seamless lane transition with no interruption from the cars around the driver. Pretty cool right? Especially if you’ve ever tried to change lanes in NYC at rush hour.

 

What does this mean for automated pickup solutions? Well, we can connect them to other parts of a brand’s ecosystem so that those parts can receive real-time Dwell Time information. This will let us do things like send reminders to the customer if they haven’t picked their order up yet, send more accurate notifications out to customers about when they should pick their order up so that it includes not just the kitchen’s capacity but customer pickup behaviors.

 

A special variation of this would be for communicating with Delivery Service Providers (DSPs) so that they could use it in their route planning and only bring drivers into the location for pickup instead of having them arrive early and clogging the location due to lack of order status visibility.

 

An example use case would be communicating to the DSP that the order is received with an estimated ready time. Then, assuming it’s a digital kitchen, they could send a notification once prep is actually started, and send a final notification to the DSP once the order is loaded in the intelligent food locker. The final notification would be the trigger for the driver to enter the brand’s location and retrieve the order.

 

Order pickup data gives you new operational insights

Let’s take a step back from real-time events and look at how we might use the data over a broad geography and/or time. To keep with the automotive analogy, think of it as “I want to understand the traffic patterns in NYC vs London” or “How do the traffic patterns of London vary over a 30-day period?” If we make the assumption that all of a Brand’s locations have intelligent food lockers, we can start to identify patterns.

 

From an operational perspective we can look at all of our deployed locations to see how dwell time differs. We can see not just how it differs by location but also by customer vs DSP (note you must populate the right data in the order to achieve this). If you utilize multiple DSPs, wouldn’t it be nice to know which ones pick orders up right away and which ones are letting them sit? If we want to enrich the data what if we also add in line-item data…. Does customer behavior change based on the contents of their order or the size?

 

Let’s move away from dwell time to more operational insights we can gain. Another piece of data intelligent cubby solutions can generate is around orders that customers pick up versus orders that are retrieved from the device by associates (reclaimed). We can look at locations across the brand and track the number of reclaimed orders to compare that against orders picked up by customers. This can be used to identify best practices and training gaps.

 

For example, you have a new associate who thinks that the right thing to do is hand the customer their order if they come to the POS, even if the order is in an intelligent cubby solution. Tracking the change in behavior of increased reclaimed orders would allow a manager at the store or another level to identify the behavior and retrain.

 

Using data to achieve efficiencies

Back to the analogy of cars with sensors. What if we wanted to plan a trip from Atlanta to Alpharetta in the fastest way possible but using the minimum amount of energy to get there. We would want to share our route plan with all of the other vehicles that will be on the roads so that they have a general plan of where our vehicle will be. We will want to be able to share real time status changes (braking, acceleration etc.) with them and our current location so that other vehicles can automatically update their routes as we move, and traffic conditions change. The goal is for all for all of the moving vehicles to communicate and allow our vehicle to move at the optimum speed for the energy usage required in order to reach its destination.

 

If we think about food, many brands have menus that have a mix of hot, cold and ambient items. Think of the scenario where a family of four orders a burger, a salad, chicken nuggets, and a grilled cheese. Then they add drinks including a coffee, milkshake, two diet cokes and a water. We could create a customer experience where the orders aren’t prepared until the customer walks in the door, but now they are frustrated because they have to wait. We could use a food locker with just heat, but we get a warm burger, melted milkshake and a wilted salad. We could use a system with 3 temperature zones, but how many partially forgotten orders are we comfortable with?

 

Or we can use an Ambient intelligent food locker super-powered by data. We can take the Dwell Time data the solution generates and use that to power our customer communications. That means we are automatically updating the suggested arrival time window we communicate to the customer based on Dwell Time and kitchen backlog. We can send them a second update that their order is in the kitchen (similar to an Uber notification that your driver is approaching) and then a final notification for when their order is loaded into the intelligent food locker.

 

Finally, the customer enters the brand’s location, they scan their pickup code and retrieve their order from a single compartment to minimize the chances of any items being left behind.

 

Your operation needs to be on super-cruise, cranking out orders at 60 miles an hour. The last thing you want is customer road rage. With an Apex intelligent food locker supercharged by order pickup data, you can avoid blindspots like customers grabbing the wrong order, orders sitting too long resulting in poor food quality or staff being too busy to hand a customer an order that is ready.