In last-mile delivery time slots are good: carriers dealing with attended deliveries, especially e-grocery ones, know that all too well. This is even truer in Europe, where attended delivery is the norm, prompting a widespread use of time-slots in all sectors of Retail.
But time-slots are not all the same. At the basic level, they are a negotiated construct between the delivery company and the customer that takes into account convenience and willingness to pay on one side and the operational process and cost on the other.
The majority of delivery providers use fixed time-slots. With them, you get to choose when you prefer your delivery, but you have to fit your needs inside a given amount of time. They can be as narrow as one hour or as wide as four, but they always start and end at the same times of the day.
Example of Fixed Time-Slots
When implementing a fixed-time slots approach, one can either assume how many orders can be delivered in any given time-slot or try to optimize this projection further to build routing schedules based upon the known orders’ geography. In any case, by forcing the user to choose bounded time-slots, you are creating un-necessary logistics constraints that might not be needed to start with, as some of the users wouldn’t mind (especially if paying less) choosing a wider time-window (I.e. 8-12, or anytime before 12, etc.)
Flexible time-slots are those where the recipient can actually choose the window size. They can be as granular as the carrier can provide, like: from 4.15pm to 5pm, from 11am to 12.30pm and so on. For them to express their full potential as operational and customer-happiness boosters they have to be algorithmically linked to operational requirements, like Residual Capacity and Historical Density, and also with pricing. Flexible slots should in fact be dynamically adaptable for the incentivisation or de-incentivisation of certain day-hours, thus optimizing fleet saturation and driving customer behaviour.
Example of Milkman’s Flexible Time-Slots
This concept is called Revenue Management and it was born in the airline industry to manage demand. As written in the paper: Revenue management in last-mile delivery: state-of-the-art and future research directions (2019), by Andre Snoeck, Daniel Merchan & Matthias Winkenbacha, of the Massachusetts institute of Technology:
“Companies can influence customer behavior by choosing the lead-times or time-slots that are offered (capacity controls) and as well as their associated fees (pricing controls). These decisions ultimately seek to balance the capacity utilization and increase the profitability of the delivery operation. Not surprisingly, revenue management (RM) for Last Mile Delivery receives increasing attention in both literature and industry.
There are three significant differences compared to traditional RM:
- First, e-retailers sell both a physical product and a delivery service, influencing order acceptance decisions. High profit product orders should get priority and products with different sizes influence the capacity of the delivery service differently.
- Second, the location of a customer, and the other customers in the same delivery route, influences the cost of delivery (Campbell and Savelsbergh, 2005). (…)
- Third, the data acquired to calibrate customer choice models for Last Mile Delivery is generally of high quality. If customers purchase certain products online, it is only at the end of their visit that they are directed to the page where they choose a delivery option. This means that they already committed to buying. If they don’t not buy, it is highly likely that this is caused by the price or availability of the preferred delivery options”.
Mining our own Milkman Business Intelligence data we can see that during the Christmas period 2020, with an ongoing Covid-19 soft lockdown, the orders served through our platform in the grocery sector were 60% of “anytime”, meaning all day long. During the same-week in 2019 anytime deliveries were chosen only by 31% of the recipients.
The data speaks for itself: flexible time-slots have always been our go-to solution. Why should we force people to choose, say: a two-hour option? Even in a non-emergency state the data reveal that one out of three customers will be home anyway and might love to benefit from a discounted option.
The result is that the Carrier will have overall lesser constraints, lower cost to serve and a lower shipping fee that will bring higher sales to the Shipper. The Shipper is also meeting the expectations of more people, by enlarging the reach and distinguishing itself among the competition. This, of course, works both ways: there will be customers who want to pay more to have a very narrow time-slot, so they will be happy too.
Recently Amazon Fresh has started offering flex-slots to its U.S. Prime customers. They are not yet common and it’s very difficult to know when they will pop up in your Cart, but when they do they are cheaper than the usual 2-hour predetermined slots. And when push comes to shove: if Amazon does it, why shouldn’t you?
My Amazon Fresh Checkout: