Premature Scaling: Will Gorillas Go Puff?
Gorillas, the 10-minute delivery e-grocer, just announced that they are putting their much-anticipated expansion into the US on pause.
The announcement came with the information of a new fundraiser, albeit at a lower valuation than before, and news of layoffs.
Gorillas refused to provide a reason for postponing the expansion. Still, recent reports have noted that the startup has struggled to raise funds and has had issues with worker protests in its home country, Germany.
Some of the reasons for their decision to halt the US expansion may be related to their specific business model:
It’s possible that the model of hyper-fast delivery just doesn’t work, and is too expensive to execute. While customers want a 15-minute delivery, it’s impossible to deliver it at scale.
It’s possible that the model of hyper-fast delivery doesn’t work in the US. Compared to Europe, the US has a higher variety of groceries (after being spoiled by many years of big-box supermarkets) and a lower density of consumers. Both of which pose issues in speed deliveries.
It’s possible that the model of hyper-fast delivery works, but is not as profitable as it seems, and will take longer to scale. This has driven valuation down, and thus the firm has to invest in efficiency before it tries to scale.
It’s also possible that the model doesn’t generate a wide enough moat to deter competition. Even if the firm expands, it will be in constant competition on both the demand and supply side, eroding profitability. Once investors realize this, they will be less willing to burn cash and offer lucrative valuations.
Other reasons might be specific to the firm. Given the amount of competition, and the firm’s struggle to manage its workforce, it's possible that investors just put the brakes on and called their bets off.
All these reasons are possible and not mutually exclusive. But let's look at other news, on a US competitor.
Several articles reported chaos and waste at GoPuff:
"The manager at one midsized warehouse said he discarded at least $10,000 worth of food every week — sometimes two or three times that. He's told Gopuff's supply-chain managers he has no space, but the product still comes. Because his freezers are full, he said, new food often goes from the delivery truck directly into the dumpster."
What's causing this wasteful behavior? According to the article, some blame the supply chain’s inability to manage inventory or the understaffed warehouses that can't process or store all the incoming shipments. The article also outlines a few other, quite odd reasons for the waste; unpaid utility bills which were driven by the fact that warehouses were never properly registered—the main point: as the firm grew, so did the food waste.
These stories outline one of the critical tensions at the growth stage of any firm: the tradeoff between growth and efficiency.
In particular, is the situation we see at GoPuff just the price one has to pay, and maybe should pay, to achieve hyper-growth in a competitive market? And, is Gorillas just the outcome of taking this approach a bit too far?
The Life Stages of a Startup
In order to understand the key questions, it is crucial to understand the stages of a startup; the six stages of a firm:
Discovery: Identifying the customer problem the startup intends to solve, ensuring that the market size is big enough, and exploring whether the solution offered to “soothe this pain” can be monetized. In other words, finding a problem/solution fit is the most important concern. The main tools are ethnographic in nature, i.e., design thinking and customer interviews.
Validation: Achieving a product/market fit. Andy Rachlef, who coined the term, defines this as: "A value hypothesis is an attempt to articulate the key assumption that underlies why a customer is likely to use your product. Identifying a compelling value hypothesis is what I call finding product/market fit. A value hypothesis identifies the features you need to build, the audience that’s likely to care, and the business model required to entice a customer to buy your product." He also describes the process of this stage as: "First you need to define and test your value hypothesis and then only once proven do you move on to what’s known as a growth hypothesis."
Efficiency: Optimizing the business model so the startup can eventually become profitable. This ultimately involves developing a repeatable and scalable business model as efficiently as possible, given the urgency dictated by the market. Note that this is not only about operational efficiency but efficiency in any aspect of the business model, from customer acquisition to execution to retention.
Scale: Once the startup has successfully progressed through the first three stages, it’s time to start engaging in accelerated customer acquisition strategies, expanding, and scaling operations.
Sustainment: Focusing on profit maximization.
Conservation: Focusing already on renewal.
We will focus on the "seam" between efficiency and scale. The central tension here is between two key ideas: Blitzscaling and Premature scaling.
Scaling: Blitz vs. Premature
Blitzscaling is a term coined by Reid Hoffman, which he describes in his book.
The main idea of Blitzscaling is that in an age of network effects and uncertainty about technology, speed should be prioritized over efficiency. The book uses Reid Hoffman’s experience at PayPal and LinkedIn as the primary examples.
The counterargument is that if you skip the stage of efficiency, and go from validation to scale, more likely than not, you are going to hit premature scaling.
The term "premature scaling" was described by Jim Pitkow, a serial entrepreneur, as growing in anticipation of demand instead of growing in response to demand.
A similar definition is provided by serial entrepreneur Michael A. Jackson:
"Premature scaling is putting the cart before the proverbial horse, and in the case of startups, this can potentially relate to both engineering and operations. Getting venture money can be like putting a rocket engine on the back of a car. Scaling comes down to making sure the machine is ready to handle the speed before hitting the accelerator."
According to data collected by Startup Genome a few years ago, nearly 70% of companies scale too quickly. Further, researchers say that next to lack of Product/Market Fit, growing too quickly is one of the most common reasons that tech startups fail. Why? In an attempt to grow as fast as possible, everything else is sacrificed.
And there are many ways firms grow too quickly (or too early):
Hiring a large staff and middle managers too early or offering generous compensation before stable revenues are established;
Focusing on growing the wrong segment of the market or cultivating users who don’t significantly impact profitability;
Pursuing growth efforts prematurely, without truly understanding the market, including growing marketing spending without established objective metrics of success (but focusing on vanity metrics);
Attempting to increase revenue or capture additional customers at the expense of profit margins, and growing revenues without building the right operations to serve these customers efficiently.
Gorillas and GoPuff are illustrations of the tension between efficiency and scale. I am, once again, reluctant to use the term premature scaling, primarily because there is no simple cutoff. There’s no flag or sign that tells you that you are definitely in the post-efficiency stage and ready to scale. Firms continuously move between these stages.
Why is GoPuff so Wasteful?
Let's analyze GoPuff first.
The firm has been under a lot of pressure to grow fast, and has been quite successful in doing so. But there is no question that the issues reported in the article are outcomes of this exact pressure.
The firm spends money on food waste by over-forecasting. Developing good forecast accuracy is not hard. But it's nearly impossible if it's not a priority.
The firm spends money on food waste since it lacks a proper process to ensure utilities are paid and warehouses are running properly. Again, this may sound trivial for most firms, but adding processes requires attention and time. GoPuff has no time for attention to such details.
When there’s chaos in a warehouse, the solution is to add more people. People cost money. For a fast-growing firm, it's hard to take time to optimize the process through lean, continuous improvements. These processes take time and managerial capacity. GoPuff has neither.
Finally, as a fast-growing business, GoPuff doesn’t have the time to deal with edge cases, resulting in fraud of the type discussed in the article.
These are all common cases, and I have seen them in most firms I have worked with. The term I sometimes hear is "bend, but not break.” Having the operations as the constraint is a sign that customers want the product. The opposite case, well-built operations without customers, is clearly even less desirable.
But we are in an era where attracting more customers to your product or service is probably easier than ever. I don't mean it's easy, but given the number of marketing channels, from Google to Facebook to Instagram, and the amount of VC money available, it is easy to plow cash into acquiring customers.
And given that the valuations are tied (primarily) to top-line growth, the priority is to continue to grow revenues, even if the firm is not efficient in growing these revenues or serving its customers.
And all of that continues until the music stops, and you are forced to pause your growth plans.
Gorillas: The Writing on the Wall
A few months ago, The Information reported that:
"Gorillas told investors that its contribution margin for each warehouse—a measure of profitability that includes only the wholesale cost of goods it sells, driver pay, warehouse staff pay, warehouse rent, packaging, and credit-card fees—would be 1.1% of revenue assuming the company reached 1,100 orders from each warehouse every day."
This is a very thin margin. Gorillas told investors it roughly had only 260 daily orders from each warehouse in May and hoped to surpass 450 daily orders per warehouse by December. In other words. Even at scale, it will barely be profitable. And the scale needed is not trivial.
What do you do until you scale? You spend or waste money.
The problem for both of these firms is that the race to acquire customers by spending money (given the competitive nature of these industries) and then spending more money on serving them (given how inefficient and under-scaled they are) would only make sense if this were a winner-takes-all market.
But it’s not. I don’t see one firm winning all customers (or even the majority) for fast delivery even within a specific geographical area.
I continue to be reluctant to consider either firm in the premature scaling stage. Both are working in a challenging sector that requires making many hard choices.
But at this stage, the signs are not very promising for Gorillas.
Returning to where I started (trying to understand the main reason for Gorillas’ decision to stop its expansion), I think operating a hyper-fast delivery service in the US is more complicated than in Europe, primarily due to wider variety and lack of density. This requires spending even more money on the back end, compared to Europe.
But, I am still a big believer in the need of this 15-minute industry to solve the problem with e-grocers: they are just not convenient enough.
However, the path is much harder than it seems. Solving the issues discussed in this article, from food waste to fraud, requires scaling and learning. Whether this is Blitz or Premature is only a question of who is winning or losing. “History is written by the Victors” (and no, Churchill was not the one who coined this phrase).
Really interesting article Gad thank you! The ironic aspect in this story for me is that speed compromised speed. The stop, start nature of this will no doubt cost time and, potential market position. Even if they can reset, revaluate and get going again.
Am wondering whether the "seam" between efficiency and scale is an area that could do with more research leading to a more repeatable approach for scale-ups...
It makes sense that if the market is "a winner-takes-all market", a startup can take a speed-first strategy. However, if the moat is relatively shallow, how much impact does this one-time dominance bring to the company?