As a professor of logistics, and as a citizen interested in, and concerned about, sustainability of business models, I am watching the meteoric rise of rapid grocery delivery services like US-based GoPuff and Germany-based Gorillas with some skepticism. During times when sustainability and the protection of our climate has reached the mainstream and has become a main pillar of politics, as well as a main talking point at conferences and on business websites, grocery delivery services have experienced dramatic growth and have reached billion dollar evaluations by investors. These companies are featured in business magazines and newspapers, their success stories are becoming the subject of case studies in business schools – and few seem to think about the sustainability of their business models.
At the core of models of rapid grocery delivery services, but also “less rapid” delivery services like Instacart, lies the concept of crowd-sourced logistics, or crowd logistics.
In this short article, I will present four theses related to the sustainability of crowd logistics. I discussed these theses with my students in a recent lecture of my course Sustainability & Sustainable Logistics at FHWS, and I thought there might be some value in writing them up. You are welcome to comment, to disagree, or to add missing points.
Aggregation of loads is one of the fundamental pillars of efficient logistics. There is a reason why both, parcel delivery companies like DHL and UPS, and regular grocery delivery companies like Picnic (Netherlands, Germany), Oda (Norway, Finland, Germany), or Ocado (UK) deliver through carefully planned milk runs, often in hub-and-spoke networks. They reduce the total distance driven and thus the negative externalities related to driving, such as fuel consumption and CO2 emission, noise emission, risk of accidents, and wear of infrastructure. And, of course, they save unnecessary costs for the operator.
Almost by definition, crowd resources provide low transportation capacity (and thus bad ratio of vehicle weight to payload) and will not carry out carefully planned milk runs, nor are they likely to integrate smoothly in a well-planned distribution network due to limitations in timing (delivery within 30 min?), as well as matching of loads with capacity available on transportation routes.
The optimization objective is lead time, not efficiency. Is the delivery of a bag of chips and a liter of Coke within 30 minutes important enough to dismiss any efficiency, and thus sustainability considerations?
With all of the above said, it is important to point out that crowd logistics does have the potential to improve sustainability of transportation of goods and persons. One of the early ideas associated with crowd logistics is that on a planned transportation route spare capacity can be offered to those in demand of transportation. If I plan to drive from Würzburg to Berlin, why not offer a seat to somebody who needs to go to Berlin, too, and why not offer to carry parcels in my trunk? It makes sense economically (I would get paid), it makes sense ecologically (more efficient transportation), and it makes sense socially (if you enjoy company, and less traffic is a benefit for society, too).
But that’s true only if I don’t decide to drive to Berlin because I want to transport one passenger, or because I want to transport parcels, the reason being that there are much more efficient means of transportation available to carry out these jobs than me with a passenger car.
The problem with crowd logistics as it is applied by rapid grocery delivery companies is that every single transportation flow is triggered by an individual order. There is no match of spare capacity with transportation demands. And moreover, if Thesis #1 holds, then these additional flows are highly inefficient, too.
By reverse logistics, I specifically refer to the return of parcels from consumers to e-com companies; I’m not talking about waste logistics here, which is another topic within the broader domain of reverse logistics. Just to give you some context, and to illustrate the order of magnitude of this issue: In Germany, 280 million (!) e-com packages were returned in 2018, containing 490 million individual items.
One means of increasing efficiency of reverse logistics (parcel returns) is the combination of delivery with pick-up of returns. In that case, delivery vehicles will carry out “double-cycles”, removing the necessity for a separate transportation flow from consumers to a drop-off point for returns. In fact, the most efficient reverse logistics network might be one of combined scheduled deliveries and pick-up of returns.
Currently, pick up of returns is not part of the business model of rapid grocery delivery services. At the same time, they are capturing part of the market share of “regular” grocery delivery services, many which do offer pick up of returns.
Does this have to be so? No, it does not. In fact, a sole purpose for the existence of a business could be that they pick up returns from consumers in an organized and efficient manner, thereby sparing consumers a separate, inefficient drive with their cars to a return point. And while they are picking up your Amazon return, they could as well deliver the bottle of Coke that you failed to add to your shopping list for yesterday’s visit to the grocery store.
So, this thesis can be turned around 180° if rapid grocery delivery services offer to pick up returns from a wide range of parcel delivery companies. Let’s hope this is going to happen!
This is a point about the social aspect of sustainability.
In order to be able to offer economically attractive delivery services in spite of all logistical inefficiency, it is necessary that workers engaging in crowd logistics activities be paid very badly compared to professional service providers (already paying their drivers low wages). Also, companies offering crowd logistics services based on self-employed “gig workers” have no means nor legal responsibility to ensure healthy and safe working standards, self-exploitation of workers being one potential consequence. This risk is exacerbated by rigorous “quality” tracking, based on sometimes mindless feedback by customers of these services, leading to dismissal of workers on short notice.
If Thesis 4 holds and delivering potato chips under time pressure is not the job a large number of people was dreaming of, then it follows that increased activity in a given economy, and thus increased competition for labor, leading to higher prices for labor, will reduce the potential workforce available to crowd logistics services and can reduce their reliability.
If underpaid gig workers can find a more permanent job that allows them to make a living, it seems unlikely that they will continue to be available for rapid grocery delivery services.
I expect that the valuation of rapid grocery delivery services will continue to increase for some time, that they will continue to expand, and that some of them are here to stay. That said, I do believe in the rationality of markets in the long term while expecting a certain degree of speculation, even insanity, in the short term. And there is reason to believe that we are witnessing an epic “new tech” and “gig economy” investment bubble. The points I make in this text, however, do not only concern economic viability, but social and economic sustainability. I am convinced that rapid grocery delivery services are unsustainable in all three dimensions of sustainability.
Frankly, I don’t believe there is such a thing as “sustainable logistics” – for reasons that I will address in a separate article (and never mind that I’m teaching “Sustainable Logistics”: you may consider it academic click bait). As logistics engineers and planners, as entrepreneurs and investors, there are things we can do to make logistics less unsustainable, however. When it comes to rapid grocery delivery services, it seems to me that many people are mindlessly accepting another convenience, and that investors know too well that mindlessness of consumers coupled with the demand for convenience are a good foundation to build on.
I do hope that somebody will challenge my assumptions, will change my mind, and that my theses will be refuted eventually. And should this not be the case I will hope that more people will get their act together and add their Saturday night junk food to their Friday afternoon shopping list, so that no 30 minutes delivery of a bottle of Coke will be necessary. Because the way to make logistics sustainable – the only way to make logistics sustainable indeed – is to avoid it in the first place.