The On-Demand Delivery Trilemma

In the crypto world there’s a popular maxim called the Blockchain Trilemma, which refers to the difficulty of simultaneously achieving three desirable properties in a blockchain network: security, scalability, and decentralization.

The trilemma states that it is impossible to have all three properties at their maximum level. As a blockchain network becomes more secure, it becomes less scalable; as it becomes more scalable, it becomes less secure; and as it becomes more decentralized, it becomes less secure and less scalable.

When designing a blockchain system, trade-offs must be made among the three parameters, leading to a balance that is acceptable for a specific use case.

There is a similar trilemma that exists for on-demand delivery in the convenience foods sector:

The ‘On-Demand Delivery Trilemma’.

The trilemma refers to the challenge of balancing three of the most important factors in the on-demand delivery of convenience foods: speed, profitability and affordability.

With the underlying model of delivering pre-ordered goods it is impossible to maximize all three factors at the same time.

For example, to achieve low delivery costs to pass on to consumers or in other words achieve affordability, the delivery system has to sacrifice either profitability or speed. To be profitable, the company needs to charge high fees making it unaffordable or improve its cost basis by being slow. To be fast, the company either needs to be profitable which means charging high fees, or forgo those fees and sacrifice profitability.

I’ll get into specific examples of each of the trade-offs and why they are not possible to get around while employing the existing underlying model of delivering pre-ordered goods.

Incumbents using this existing model of delivery include: delivery robots, route-based ice cream and food trucks, delivery apps and quick commerce.

Delivery Robots

Delivery robot companies like Starship, Nuro, Serve, Kiwi, and Coco, reduce costs by automating the driver. This reduction in costs helps them achieve profitability and affordability, yet at the cost of speed. It takes longer for a robot to make a delivery than a human, considerably longer in the cases of slow moving sidewalk bots.

Route-based trucks

Route-based ice cream and food trucks have existed for decades, but newer players like Scream Truck and Zing are trying to modernize them. However they still operate on the same underlying model of pre-ordering goods, and thus can achieve only profitability and affordability yet not speed. By virtue of being route-based, they park up in central locations and force consumers to walk to them, line up, and buy goods to either consume or haul back home. A considerably slow experience.

Delivery apps

On the right side of the trilemma we have incumbent delivery apps like Doordash and UberEats, that are fast and can be profitable. You may be asking then what the problem is. The issue is that they can only be profitable at an extremely high expense. They pass on the costs to the consumer in the form of delivery fees, service fees, and tips, all of which can easily be close to?$10 per order?in just fees.

Quick Commerce

Finally at the bottom of the trilemma you have services that attain speed and affordability like Q-Commerce apps Getir, Gorillas (part of Getir), GoPuff, and Flink, yet in their quest to deliver cheaply and quickly, they sacrifice profitability. Their average deliveries per hour?top out at 2.5–3, which means that no matter how fast or cheap they get they won’t be able to break-even.

There is however, an emerging model that addresses and solves the trilemma and gets around its challenges. This model completely reimagines delivery of convenience foods and changes the underlying model of delivering pre-ordered goods.

Roving Shops

Roving Shops are on-demand food trucks, snack shops, and ice cream shops that carry all their inventory within the vans. They do not require order pickups and are thus able to get around the trilemma and create a model that is fast, cheap and profitable.

A few of the players in this space include?Robomart, my own startup, serving up ice cream shops and snacks shops on-demand to consumers in as little as two minutes, and?Muncho, a pizza shop that parbakes pizzas and finishes them off in the van en route to the customer.

Both are able to attain over?5 orders/hour?to achieve profitability, because they don’t need to pick up orders, and are able to have their vans/shops roving around potential demand. This means that not only is the model profitable, but they are able to pass on the cost savings to consumers in terms of affordability and ensure the fastest speed of delivery possible.

Several other players have attempted this model in the past, like Zume Pizza and most recently Wonder. Both pivoted away from the model, because they spent a ton of capital trying to build a?vertically integrated approach, which added significant costs and reduced their potential for profitability.

This new crop of players delivering roving shops are taking an asset light approach, using off the shelf equipment and rented vehicles to keep costs low. Additionally algorithms that constantly match supply to demand ensure throughput of vehicles in any given operating neighborhood.

The ‘On-Demand Delivery Trilemma’ will continue to plague industry incumbents unless they are able to think beyond the existing model of delivering pre-ordered goods and find new compelling ways to get goods to consumers in the fastest, cheapest, and most profitable way possible.