Two companies needed a new growth vector. One was great at branding alcoholic beverages. The other was great at making home appliances that customers could use to create self-serve beverages. A match made in heaven! Or maybe not…
When companies make a product that has more to do with their own needs than with solving a real customer problem, the results are seldom pretty. One of the latest casualties? Drinkworks, the product of a joint venture between ABInBev and Keurig. Both of the parent companies have been grappling with strategic growth issues. People aren’t drinking as much beer as they used to (or at least as much of the kind that Anheuser Busch sells) and Keurig has long lusted after finding markets beyond coffee for automatic home preparation.
The history
Drinkworks was founded in 2017, funded by AB InBev’s venture fund, ZX, as part of a joint venture between the highly successful coffee pod maker Keurig Dr. Pepper and Anheuser Busch. The technology behind the at-home drink maker was underway at both companies, reportedly, for over ten years. While I don’t have any insider information, it sure looks as though when the two companies hit the “go” button, the recipe for landing in my “flops” file was in place. Lots of up-front funding, a big team, assumptions taken as fact, leaders totally bought in to the concept without the potential to change direction…and so it goes.
The product launched in St. Louis in November of 2018. The offering was a machine that one would buy for $299, plus a set of pods which contained all the ingredients necessary to create a great cocktail. The presumed customer?
As their CEO said at the time of launch: “The core person who really loves this idea is somebody who steps up above the ordinary: the person who hosts a party and makes up a cocktail or sets things up so that everybody can make something unique on their own. We can provide not only that but also a huge variety of drinks. And we give them back the time they would otherwise spend.”
The technology came mostly from the Keurig side of the house, possibly benefitting from an earlier failed venture, that time with Coca Cola, on a project called “Keurig Kold”. That product was deemed to make “frighteningly expensive fizzy drinks.” To one reviewer, at least, the product did not taste “the same or as good” as the store-bought variety. The machine was expensive – $370 for a huge machine that weighed 23.7 pounds. Making the drinks was also time-consuming. Unlike the Keurig coffee maker that delivers a cup in about 30 seconds, the soda machine (after the initial setup of 2-hours to get it into chill mode) took 90 seconds, as the reviewer reported, “about 75 seconds longer than it would take you to reach into the refrigerator and open up a can of soda.”
But perhaps the biggest problem with the Keurig Kold was the economics – as the Cnet reviewer noted in 2015, “I picked up an eight-pack of 12 ounce classic Coca-Cola bottles for $5 here in Louisville. With a total volume of 96 ounces this comes to 19.2 ounces per $1, or just over 5 cents per ounce. The same equation works out to just 6.4 ounces of beverage for every $1 dollar you spend (or more than 15.5 cents per ounce) through Kold pods— a whopping three times the cost. Frankly, that’s outrageous — especially for a solution that doesn’t solve a customer problem as far as I can tell.”
And that, to me, is the same issue that the Drinkworks product faced – do enough customers have a big enough problem making cocktails to merit the undoubtedly big investment the firms made in creating the solution?
It sure sounds good
Visiting the Drinkworks web site (as of January 30, 2022), you might think you were witnessing a revolutionary startup out to change the world. It’s pretty impressive. Nathaniel Davis, President and CEO, proclaims that the company’s mission is nothing less than reinventing the entire drinking experience. As they say, “From purchase, to creation, to enjoyment – striving to make it all more remarkable through innovation.” The company values are amazing, too – “We deliver remarkable” “we make the user the hero” “we act as owners” “we collaborate for success” “we act thoughtfully in everything we do”.
Because what we really all need is a push-button way of creating a craft cocktail at home, at a price per cocktail of about $5 each.
We should maybe test that assumption?
Technology first, real testing with customers later…
At the time of the St. Louis launch, spokespeople for the company described with great pride the enormous investment the joint venture had made. As their CEO at the time said, “We’ve got scientists, chemists, process engineers, and so on to figure out how to do that. There’s lots of research and development, lots of technology.” As of 2019, the company continued its fairly aggressive development, announcing a partnership with Brown-Forman corporation leading to the ability to make even more branded types of cocktails. The company went nationwide in 2020, hoping to capitalize on a resurgence of interest in all things at-home during the pandemic. If you look at the video on their website, you’ll see big offices, manufacturing plants and a whole lot of other stuff that must have been really expensive.
One of the core assumptions underlying the Drinkworks value proposition was that what would attract customers would be the huge amount of variety that could be created with interchangeable drink pods. As Nathaniel Lewis, their CEO said, in a July, 2021 interview (that’s about 6 months ago, folks) “When people buy into the system, especially if they’re buying the machine, there needs to be a huge set of choices and the expectation that there’s always going to be new things around the corner. We said, “Well, how can we help our users navigate a complex field of choices?” We landed on using the bar menu because that’s what people understand as the natural way to navigate cocktails. We’ve got collections that are the clusters that you’d see in a really comprehensive bar menu. There are 40 cocktails, by the way, in our system.”
Lewis came up through the technology side of things, and is clearly a master of the complex inventions that are necessary to produce the output of the machine at all. And the investment would be necessary if you truly believed that a huge variety of cocktails was essential to delivering the consumer value proposition. By all accounts, he’s a great leader, but even that won’t help you if you’re not solving a big enough customer problem.
Back to the wisdom of “The Right It”
Now we come back to a fundamental issue that Alberto Savoia has so wisely talked about in his terrific book “The Right It.” An example from the book to me eerily represents what could have been the issue with Drinkworks. And it reflects what happens when enthusiastic people live in what Alberto calls “thoughtland,” that magical place where our beliefs and assumptions are not questioned, all ideas are good ones and there is no such thing as the potential for missing the mark.
In his book, Alberto talks about a fictional invention that makes tortellini on demand. The “Tortell-o-matic” is a $250 automated tortellini maker for the home. Fill it with flour, water, eggs, salt and your choice of filling and two minutes later the machine pops out perfectly shaped, freshly made tortellini.
In thoughtland, this sounds like a GREAT idea. Who wouldn’t want to enjoy freshly made tortellini without all the bother of making them yourself or heading to the store? We might ask, who wouldn’t want to enjoy freshly made cocktails without the bother of making them yourself or heading to the store (or a bar, or wherever). But not so fast. As Alberto says,
“I can use one of my favorite pretotyping techniques, the Mechanical Turk, to set up an experiment and collect real data about real customer reactions. A dummy Tortell-o-matic sits on a table. Under the table and hidden by a tablecloth, an accomplice holds a bucket of premade tortellini. When I push a button on the nonfunctioning machine, the accomplice plays a recording of mechanical noises and begins to push tortellini through the machine.”
As he goes on to explain, you can use this kind of mockup to assign what he calls “skin in the game” points to the invention. Using this pretotype, you can work your way through a series of questions about how viable this idea is likely to be – with real “points” given to a validated email address with the understanding that it will be used for product updates, a validated phone number, a commitment of time (such as attending a product demonstration), a cash deposit and placing an order.
You can watch Alberto’s mini lesson on this technique here.
As he points out, you could get a lot of evidence on real customer demand before building any technology whatsoever. Rather than building out all that expensive infrastructure, supply chain, partnerships and manufacturing, get real evidence that your offering solves a compelling customer job-to-be-done first!
The discovery driven discipline as well – market benchmarking
Finally, customer reaction aside, let’s ask the question from discovery driven planning of what success would have to look like for this venture to be deemed a winner, from the point of view of its parent companies. As of last year, Anheuser Busch had revenue of about $15.59 billion. Keurig Dr. Pepper’s revenue was $11.62 billion. Together, that adds up to $27.21 billion. Here we run into the classic problem of large companies, which is that for any internal venture to be considered successful, it has to add a meaningful chunk of revenue to the parents’ bottom line.
Let’s say that the Drinkworks initiative, to be material to the two parent firms, has to generate an increment of 5 percent to their top lines. That translates into $1.36 billion. If we assume that 50% of that amount has to come from machines sales every year, at $299 per machine, that implies sales of 2.3 million machines each year. If we further assume that the other half of revenue has to come from pod sales, at a price of $17.99 for a 4-pack, that means sales of 37.8 million packs, again, every year.
A November 2021 article reported that the “cocktail platform” was selling 250,000 cocktails a month as of that time. That means 62,500 4-packs a month, or 750,000 per year. For the platform to deliver to my hypothetical targets, they would have had to see a glide path that could increase sales 30-40 fold!
Anyway, someone must have finally done the math, and after 4 years and I can only imagine how much money, the companies announced the immediate cessation of the business on December 1, 2021.
Could an ecosystem approach have been better?
As Ron Adner and I discussed in our Friday Fireside Chat, every business today has an ecosystem dimension to it. What Drinkworks was trying to do was become a go-to at-home bartender in much the same way Keurig’s popular coffee machines are used by 27% of all Americans. One could conceive of such a machine that would be a platform through which multiple producers could sell their products. One could even conceive of the popular razor-and-blades model. In that model, companies make one part of the solution inexpensive or even free, and make their money on the renewables that are used with it. Gillette of course did this with razors, and Hewlett Packard does this with printer ink.
For this to work, the company would have to lower the barriers to ownership, potentially subsidizing the machines significantly while a critical mass of regular users would be created. Making people shell out a lot of money up-front for a machine whose utility they still have to be convinced of is a big ask. And it’s the wrong way to build a platform – as any expert will tell you, you want to lure customers to your platform first, and make the money from their presence there later on.
An alternative to Drinkworks, the Bartesian cocktail maker, debuted as a project on Kickstarter in 2015. Focused on the drinks, the company licensed the manufacturing of hardware to Hamilton Beach. It started shipping units in 2019. It announced a $20 million Series A round in April of 2021, and according to the funding announcement had already sold 5 million cocktails. The big difference between the Bartesian system and the Drinkworks one is that with Bartesian, you supply your own alcohol – the pods are for flavors. Another big difference is that Bartesian is a startup, scrappily bootstrapping its way into the market. The 2020 Consumer Electronics Show was a bit of a coming-out for the company, and it has been looked at with admiration by savvy observers. By all accounts it is growing rapidly.
So the Keurig for cocktails idea may be one that works, but with the right model and execution. As Alberto would say, building the wrong “it” will leave you exposed to the beast of failure.
De-risking innovations and creating customer insight
The techniques I’ve described here – creating real customer insight and de-risking projects through discovery driven planning – can now be learned in a series of short on-line courses I’ve developed. They are perfect for people facing important decisions about whether innovations make sense and how to develop them. You can find out more here.