A Dozen Lessons about Minimum Viable Products

“It’s only cheap to build 2-3 person companies with sweat equity. The minute you start paying engineers you will realize it is quite expensive.” Bill Gurley. Assume a startup has raised a seed round of ~$2 million. Also assume that what the startup has is a hypothesis that a big market composed of dogs will want to eat the dog food described by the hypothesis. The founders of the startup have no proof that their hypothesis is true, but some investors have voted with their money that there is significant hope that the startup’s hypothesis is correct. Every penny of the $2 million raised by the startup is precious. If the startup runs out of cash it is dead, since that is the only unforgivable sin in business. Now let’s look at the overall context in which this is happening. The odds that the startup will be financially successful are, simply put, not good. How many startups raise a seed round? There is no way to know for sure since many startups at seed stage live and die and don’t leave a trace that can be tracked. Reported seed stage startups typically number about 1,200 in a given a quarter (plus or minus a couple of hundred) depending upon the business climate. Assuming ~5,000 seed stage startups a year both reported and unreported, only 800 of them raised a Series A round in 2016 says Mattermark. That’s about an 84% fatality rate just at seed stage. Mattermark also calculates the odds of survival here at far less than 10%. This calculation is based simply on a startup not getting to the next phase.

grad

Other research, which uses different definitions, concludes:

About 75% of U.S. venture-backed start-ups fail, according to Harvard Business School senior lecturer Shikhar Ghosh. Ghosh’s research estimates 30% to 40% of high potential start-ups end up liquidating all assets–a failure by any definition. But if a start-up failure is defined as not delivering the projected return on investment, then 95% of VC companies are failures, Ghosh said.

Being a founder or early employee of a startup is not a rational act given the odds of success. Of course, as George Bernard Shaw wrote in Man and Superman: “all progress depends on the unreasonable [human being].” The reason why books like The Hard Thing about Hard Things by Ben Horowitz and Shoe Dog by Phil Knight resonate so strongly with people who have been involved in startups is that they accurately describe the terror, inevitable setbacks and daily struggle of life in a startup business, not just the seemingly glamorous parts. .

“If you create something and no one uses it, you’re dead. Nothing else you do is going to matter if people don’t like your product.” Jessica Livingston. The first rule of startups is that without making something that people want to buy, you’re dead. The second rule is that you should not forget the first rule. Particularly when the odds of survival are low in an activity, it pays to be very aware of methods that can increase the probability of survival. Michael Mauboussin’s advice should be front and center in every founder’s mind: “If you compete in a field where luck plays a role, you should focus more on the process of how you make decisions.” What should that process be for a startup? In thinking about the right process it is wise not to forget that the startup’s goal is to establish product/market fit before they run out of money. Unfortunately, at the very early stages of the startup’s existence it faces many challenges related to at least one untested hypothesis. “Hypothesis”, of course, is just a fancy word for “guess.” Steve Anderson the founder of the seed stage venture capital firm Baseline Ventures points out: “Generally speaking, most of my investments are pre-product launch – they’re just an idea. My goal as an investor is to make sure there’s enough financing to give companies time to do that, a year to 18 months. The worst scenario is to try to raise more money when you haven’t achieved that goal. If you don’t have it, eventually you’ll run out of cash, say the experiment is wrong, and fold up your tent. That’s why when I invest I want to leave enough room for pivoting or reexamining your goals.” Making matters even more challenging for the early stage startup is the point Ev Williams makes here: “You know that old saw about a plane flying from California to Hawaii being off course 99% of the time—but constantly correcting? The same is true of successful startups—except they may start out heading toward Alaska.”

“A full executive team with a salesforce and all that stuff before you have a killer product is a complete waste of time.” Marc Andreessen. A startup should defer spending time and energy proving and developing its growth hypothesis until it has established the value hypothesis. I have recently written a blog post on precisely this “first value THEN growth” point here. The key point in that post is made by Andy Rachleff: “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. Companies often go through many iterations before they find product/market fit, if they ever do.” When the startup is still searching for the elements of its value hypothesis, money and time spent on growing the business is a bonfire of cash generating zero value. The early days of the life of a startup are focused on “search” rather than “execution” advises Steve Blank, a serial entrepreneur, professor and author who is justifiably famous in the startup world.

“The minimum viable product (MVP) is that product which has just those features (and no more) that allows you to ship a product that resonates with early adopters; some of whom will pay you money or give you feedback.” “The lesson of the MVP is that any additional work beyond what was required to start learning is waste, no matter how important it might have seemed at the time.” Eric Ries. The goal of the MVP process is to validate the hypothesis in a speedy and cost efficient manner. The key word in this quote from Ries above is feedback since that is how anyone learns. The most effective processes are based on feedback loops which are in turn based on the scientific method: build, measure, learn. What the startup offers as its MVP should be compete in what it does to deliver and capture value, not a fully complete implementation of the vision. The MVP is an experiment that is intended to generate validated learning about what customers value enough to pay for. An MVP approach is not the only way to go forward with a startup. Eric Ries describes two extreme alternatives:

“One, which I call maximizing chance of success, says ‘Look, we only got one chance at this so let’s get it right.’ We’re going to ship it when it’s right and that actually is perfectly rational. If you only have one shot, you want to take the best shot you can and build the most perfect product you can. The issue is, of course, you know, you can spend, I don’t know, say five years of stealth R&D building a product you think customers want and then discover to your chagrin that they don’t. The other possible extreme approach is to say, ‘Well, let’s just do ‘release early, release often.’ This approach is: ‘Look, we’ll just throw whatever crap we have out there and then we’ll hear what customers say and we’ll do whatever they say.” But the issue there is if you show a product to three customers, you get 30 opinions, and now what do you do? So minimum viable product is kind of a synthesis of those two possible extremes.”

“As you consider building your own minimum viable product, let this simple rule suffice: remove any feature, process, or effort that does not contribute directly to the learning you seek.” “If you want to do minimum viable product, you have to be prepared to iterate. And so you have to have the courage to say, ‘Yeah, we’ll ship something, get negative feedback and respond.’” Eric Ries. A minimum feature set is not a goal but a tactic to create cost-effective and speedy validated learning about the hypothesis. The goal is to learn and steer based on feedback rather than try to predict and emerge with a killer fully formed product. Some people like Peter Thiel who is quoted just below, have a different view:

“Even in engineering-driven Silicon Valley, the buzzwords of the moment call for building a ‘lean startup’ that can ‘adapt’ and ‘evolve’ to an ever-changing environment. Would-be entrepreneurs are told that nothing can be known in advance: we’re supposed to listen to what customers say they want, make nothing more than a ‘minimum viable product,’ and iterate our way to success. But leanness is a methodology, not a goal. Making small changes to things that already exist might lead you to a local maximum, but it won’t help you find the global maximum. You could build the best version of an app that lets people order toilet paper from their iPhone. But iteration without a bold plan won’t take you from 0 to 1. A company is the strangest place of all for an indefinite optimist: why should you expect your own business to succeed without a plan to make it happen? Darwinism may be a fine theory in other contexts, but in startups, intelligent design works best.”

Thiel or an entrepreneur like Elon Musk are not as capital constrained as the typical seed stage startup. They can afford to adopt what Ries called a “maximizing the chance of success” approach. Thiel in particular makes many bets and is nicely hedged since he owns a portfolio of wagers. In contrast the founders and early employees of a startup typically have all their eggs on one basket. The founders and early employees are far from hedged. What is right for Thiel may not be right for founders or early employees for that reason.

“An MVP is a process that you repeat over and over again: Identify your riskiest assumption, find the smallest possible experiment to test that assumption, and use the results of the experiment to course correct.” Yevgeniy Brikman. The MVP process is depicted as a flywheel or loop for a reason. Most of the time actual testing of a hypothesis will reveal that customers do not value the product or even the vision the product represents. If the hypothesis is not validated by the experiment the business must iterate by replacing the hypothesis or shut down. I like this description of the process from an interview of Steve Blank by Chris Dixon:

“An MVP is really just a tool for discovering a scalable business model through customer development. An MVP should have the smallest possible feature set that creates gains for customers and reduces pain—but it can’t be so small that customers have nothing to evaluate. In other words, an MVP gives startup entrepreneurs something to demonstrate when they get out of the building and talk to current and potential customers about what they really need.”

“The worst fate of any shipping of any product is that nobody cares. You don’t get any feedback at all. That’s what most features or most products do. They’re just dead weight.” Eric Ries. What Ries says here is an unfortunate fact. Chamath Palihapitiya describes reality bluntly: “Core product value is really illusive and most products don’t have any.” Faced with the reality of shutting down many companies just push the button and start working on the growth hypothesis without having solved the value hypothesis, starting a process in which they will usually fly the business at high speed into the side of a mountain.

“The common phrase that most people use today is,”You should build a minimum viable product.” And I underlined viable because I think a lot of people skip that part and they go out with a feature and the whole user experience in the very beginning is flat. Minimal viable product pretty much means what is the smallest feature set that you should build to solve the problem that you are trying to solve. I think if you go through the whole story-boarding experience you can kind of figure that out very quickly. But again, you have to be talking to users, you have to be seeing what exists out there already, and what you should be building should solve their immediate needs.” Sam Altman. The graphics which best describe what Altman is talking about depict the MVP as being complete in terms of what it does but not as complete as it will eventually be in implementing the vision once the feedback is obtained from early adopter customers.

slice-1

slice-2

“A minimum viable product is not always a smaller/cheaper version of your final product.” “Launching a new enterprise—whether it’s a tech start-up, a small business, or an initiative within a large corporation—has always been a hit-or-miss proposition. According to the decades-old formula, you write a business plan, pitch it to investors, assemble a team, introduce a product, and start selling as hard as you can. All MBA tools are irrelevant on a startup’s day one. This wrong belief is based on that we can start absolutely any company just by spending a lot of time on writing complicated operating plan and financial model and then hire people to execute this plans. But now we know that no plan survives first contact with customers! First days of startup are completely unpredictable. Business plans and financial forecasts are just silly as it was in the Soviet Union.” Steve Blank. A classic example of a MVP is what was done by Zappos Founder Nick Swinmurn: “My Dad told me, you know I think the one you should focus on is the shoe thing. That’s a real business that makes sense. So I said okay, focused on the shoe thing, went to a couple of stores, took some pictures of the shoes, made a website, put them up and told the shoe store, if I sell anything, I’ll come here and pay full price. They said okay, knock yourself out. So I did that, made a couple of sales.” If you can validate your thesis without paying to create lots of code that approach is like found gold. As another example, the MVP for AngelList mostly took the form of making introductions by email. The Virgin Airlines MVP was just a single plane flying back and forth between two cities. The less money spent on proving the hypothesis, the more money that is left to pivot or execute on the idea.

“A MVP is not just a product with half of the features chopped out, or a way to get the product out the door a little earlier. And it’s not something you build only once, and then consider the job done.” Yevgeniy Brikman. The MVP should deliver value to the customer even though it is not as complete at is could be. Some people argue that an MVP can be as simple as a landing page, but I am skeptical. Eric Ries writes: “The idea of minimum viable product is useful because you can basically say: our vision is to build a product that solves this core problem for customers and we think that for the people who are early adopters for this kind of solution, they will be the most forgiving. And they will fill in their minds the features that aren’t quite there if we give them the core, tent-pole features that point the direction of where we’re trying to go.”

bike

“MVP is quite annoying, because it imposes extra overhead. We have to manage to learn something from our first product iteration. In a lot of cases, this requires a lot of energy invested in talking to customers or metrics and analytics. Second, the definition’s use of the words maximum and minimum means it is decidedly not formulaic. It requires judgment to figure out, for any given context, what MVP makes sense.” Eric Ries. It does not make much sense to build a MVP unless you do the work to collect data about the experiments and conduct an analysis using modern tools. This data collection and analysis is a lot of work and is not as glamorous to some people as product design, creating marketing plans and attending fancy conferences and parties.

“In the real world not every customer is going to get overly excited about your minimum feature set. Only a special subset of customers will and what gets them breathing heavy is the long-term vision for your product. The reality is that the minimum feature set is 1) a tactic to reduce wasted engineering hours (code left on the floor) and 2) to get the product in the hands of early visionary customers as soon as possible. You’re selling the vision and delivering the minimum feature set to visionaries not everyone.” Steve Blank. Every potential customer does not need to value the MVP for it to be a success. Eric Ries elaborates: “Early adopters can be very forgiving of missing features. They see the vision and you can be in dialogue with them going through that learning feedback loop.” Operating in this process is faith that the customers will help evolve the MVP into something fantastic that will support a very profitable business with a scalable and repeatable business model.

Notes:

Eric Ries: Minimum Viable Product: a guide http://www.startuplessonslearned.com/2009/08/minimum-viable-product-guide.html

Eric Ries: What is minimum viable product? http://www.startuplessonslearned.com/2009/03/minimum-viable-product.html

Steve Blank: Perfection by Subtraction. https://steveblank.com/2010/03/04/perfection-by-subtraction-the-minimum-feature-set/

75% of Venture-backed Start-ups Fail http://www.inc.com/john-mcdermott/report-3-out-of-4-venture-backed-start-ups-fail.html

Eric Ries: https://m.youtube.com/watch?v=1FoCbbbcYT8

Steve Blank https://m.youtube.com/watch?v=Fj0qsAyKPN8

Neil Patel: Developing an MVP: Your Key to Success https://medium.com/@NeilP666/developing-an-mvp-your-key-to-success-43333610ab12#.yq9f3lecb

LinkedIn Founder Reid Hoffman’s Advice for Entrepreneurs https://blog.kissmetrics.com/hoffmans-advice-for-entrepreneurs/

Wikipedia https://en.wikipedia.org/wiki/Minimum_viable_product

Minimum Viable Products in Biotech http://blogs.nature.com/tradesecrets/2011/12/20/minimum_viable_products_in_biotech

Chris Dixon interviews Eric Ries: https://techcrunch.com/2011/09/27/founder-stories-eric-ries-lean-google-plus/