Machine Learning Meets the Lean Startup

We just finished our Lean LaunchPad class at UC Berkeley’s engineering school where many of the teams embedded machine learning technology into their products.

It struck me as I watched the teams try to find how their technology would solve real customer problems, is that machine learning is following a similar pattern of previous technical infrastructure innovations. Early entrants get sold to corporate acquirers at inflated prices for their teams, their technology, and their tools. Later entrants who miss that wave have to build real products that people want to buy.


I’ve lived through several technology infrastructure waves; the Unix business, the first AI and VR waves in the 1980’s, the workstation wave, multimedia wave, the first internet wave. Each of those had a set of common characteristics that the Gartner Group characterizes as the Hype Cycle .


The five stages of the hype cycle are:

Stage 1: The Technology Trigger: A potential technology breakthrough kicks things off. Early proof-of-concept stories and media interest trigger significant publicity. Often no usable products exist and commercial viability is unproven.

Stage 2: Peak of Inflated Expectations: Early publicity produces a number of success stories—often accompanied by scores of failures. Some companies take action; most don’t.

Stage 3: Trough of Disillusionment: Interest wanes as experiments and implementations fail to deliver. Producers of the technology shake out or fail. Investments continue only if the surviving providers improve their products to the satisfaction of early adopters.

Stage 4: Slope of Enlightenment: More instances of how the technology can benefit the enterprise start to crystallize and become more widely understood. Second- and third-generation products appear from technology providers. More enterprises fund pilots; conservative companies remain cautious.

Stage 5: Plateau of Productivity: Mainstream adoption starts to take off. Criteria for assessing provider viability are more clearly defined. The technology’s broad market applicability and relevance are clearly paying off.

Shiny Object meets First Mover Advantage
What’s become apparent in the last few technology hype cycles is that for startups and their investors there is a short multi-year window of opportunity (at the Peak of Inflated Expectations) to sell a startup at an inflated price. This occurs because large technology companies (Google, Facebook, IBM, Microsoft, Twitter, Apple, Salesforce, Intel, et al,) and increasingly other non-tech firms, are in an arms race to stay relevant. For example, according to CBInsights nearly 140 machine intelligence have been acquired since 2011, with over 40 being bought so far in 2016.


Most often the first acquisitions in a hype cycle are for the “shiny objects” – the technology, the team and the tools. The acquired technical teams usually start up or complement the company’s research group in a specific new technology area.hype-cycle

If you’re a startup (or their investors) getting acquired at this point in the hype cycle is exactly where you want to be – short time in business, large acquisition price, value based on a frenzy, perceived scarcity of expertise, and fear of a competitor getting the key talent.

History shows that the acquirers often overpay buying this expertise early. While these acquisitions have teams of great researchers, they rarely contribute actual revenue generating products (because most never reached that stage when they were acquired.)  The irony is that the acquisitions made later in the hype cycle – when companies have built real products that customers want, are the ones that generate revenue and profit for the acquirer.

I had all that in mind as we watched our teams present.

Machine Learning Meets Lean – Berkeley Lean LaunchPad Class

Each of our teams in this class followed the canonical Lean model:

  1. Articulate your hypotheses using the business model canvas
  2. Get of the building and test those hypotheses using customer development
  3. Validate learning by building minimal viable products and getting them in front of customers

Each week the teams got out of the classroom and talked to 10-15 customers, testing a new part of the business model canvas.  And after week two, they had to build and then update their minimal viable product weekly. And present what they learned each week in an 8-minute presentations.

The presentations below are their final Lessons Learned presentations, along with a 2-minute video summary.

Three Berkeley PhD computer science students and an MBA working on machine learning. How can you not hit out of the park on day one?

This team epitomized rapid learning. Once their initial assumptions ran into the wall of actual customer feedback they rapidly built multiple minimum viable products (MVPs) and kept pivoting until they found product/market fit (i.e. a customer segment that was grabbing the product out of their hands.)

If you can’t see the video click here

If you can’t see the presentation click here

Before this class this team had spent three months in an incubator building the product after talking to only one customer. After  week two of the class they realized they had wasted three months building something no one actually wanted. What they next learned was pretty amazing.

If you can’t see the video click here

If you can’t see the presentation click here

Homeslice had a great journey. They came together over a personal pain – the inability to afford a house in Silicon Valley. Their initial plan was to provide fractional ownership to solve that problem. But they found that first serving an adjacent market – slices of investment properties – could serve as a launchpad for their initial idea of fractional home ownership.

If you can’t see the video click here

If you can’t see the presentation click here

Exit Strategy
Exit Strategy was building the penultimate planning tool. This teams learning that this wasn’t a business was as important as finding one that is. Really impressive process.

If you can’t see the video click here

If you can’t see the presentation click here

This class this was a team effort. Professor Kurt Keutzer and Errol Arkilic (former program director for the National Science Foundation Innovation Corps (NSF I-Corps), now founder of M34 Capital) were the lead instructors. Steve Weinstein (CEO of MovieLabs) and I assisted. Thanks to our TA Kathryn Crimmins and all the team mentors: Lev Mass, Kanu Gulati, Ewald Detjens, James Cham, Kanu Gulati, Patrick Chung, Rick Lazansky, Ashmeet Sidhana, Mike Olson, Michael Borrus, Fabrizo De Pasquale, Amit Kumar, Rob Rodick, Mar Hershenson.

How The Marine Corps Builds an Innovation Culture

marine-corps-logoJennifer Edgin is the Chief Technology Officer of the Intelligence Division at the Headquarters of the Marine Corps. As the Senior Technical Advisor to the Director of Intelligence, she is responsible for building and infusing new technologies within the Marine Corps Intelligence, Surveillance, and Reconnaissance Enterprise (MCISRE). Jennifer is one the “innovation insurgents” inside the Department of Defense driving rapid innovation. Here’s her story of the Lean innovation accelerator she’s built for the Marines.

If you asked 100 people to describe a United States Marine, they would probably use words such as “Warrior,” “Fierce,” “Patriot,” “Honorable,” and “Tough.” Marine Corps culture transcends generations and is rooted in the values of courage, honor, and commitment.  Marines are known for adapting to change and overcoming obstacles and adversity to meet new mission requirements continuously.  Three years ago, Marine Corps Intelligence outlined a mission to harness the disruptions occurring in the new frontier of warfare, the Electronic battlefield. To achieve this mission, we established a framework that leveraged Marine Corps tenacity, agility, and adaptability to create a persistent culture of innovation.

One of our primary goals in establishing this framework was to keep the user front and center, and to quickly deliver solutions to their challenges. To achieve this, we stood up the Marine Corps Intelligence, Surveillance, and Reconnaissance Enterprise (MCISRE) Accelerator. Like a tech startup accelerator, the MCISRE Accelerator assembles a cohort of active duty Marines of all ranks, experiences, and disciplines and pairs them with developers, designers, and mentors through a 12-week “Design—Develop—Deploy” cycle.  Marines are taught tools and methodologies from the Lean Startup, Design Thinking, and Service Design practices, which are then used to zero in on a problem; identify the target customer segment; validate the problem and solution by “getting out of the building” and submitting their problem and concept designs to peers for feedback, designing wireframes and prototypes, developing a minimum viable product (MVP); and finally pitch the MVP to the Director of Intelligence (DIRINT) and other leaders and stakeholders for a go/no-go decision for release.

Over the course of the last year, we have carefully measured and monitored our framework so we could quickly identify what was working, what was not working, and tune accordingly so that the end result created value for both the Marines in the cohort and the larger community of Marine Corps Intelligence. Below are the top 5 factors we found are necessary to successfully innovate.

1. Understand Your Customer and Teach Them to Solve Real Problems
Innovation begins and ends with understanding the customer and the specific problems they are facing. Too often in government, problems are talked about in generalizations, users are not part of the design and development of solutions, and anecdotal information gets passed around without data to validate it until at some point it becomes “truth” and is accepted without verification.

distributed-common-ground-surface-systemOne of the most difficult exercises for our cohorts is distilling “world-hunger”-level challenges into discrete, focused problems we can solve in 12 weeks. We learned that if you cannot define your problem in one sentence that a 7 year-old can understand, you don’t understand the problem.  If you want to create innovative solutions, you must start by defining real problems. Real problems—when defined properly—have metrics that quantify the scope, magnitude, and impact.

Before we launched the MCISRE Accelerator, we conducted site visits, and spoke with Marines from around the world to hear from them what wasn’t working and what was.  After our site visits, we identified common issues across each of the sites, disciplines, and ranks, and launched data surveys to explore and quantify problems.  The data showed us how users were currently performing a mission, where deficits existed in enabling technology and processes, and which anecdotal problems were actual problems and which were not. We then compared the results of the surveys with the site survey interviews and prepared a list of the top issues and challenges facing Marine Corps Intelligence. When we launched the MCISRE Accelerator, we used this information to quickly move the focus of the cohort from the world-hunger view to zeroing in one or two key issues that caused major disruptions in their tasking and productivity.

One of the biggest benefits to this process was that it helped shift the focus of the Marines from nebulous systems-centric thinking—“The network architecture sucks”—to identifying specific pain points impeding their productivity on the job. Because the tools and methodologies we use are simple but highly effective for analysis and problem solving, many of our cohort Marines take them back to their units so that they can reframe problems within their communities of practice.

2. Always Be Shipping
If vision without execution is hallucination, frameworks that don’t produce tangible products breed insanity.  Within the MCISRE we have two frameworks that engage Marines. Our yearly technical design meeting (TDM) brings together Marines to identify and address challenges and issues across the MCISRE. The outputs of the technical design meeting are used as primers for defining the problem themes that each MCISRE Accelerator cohort will work on.  The MCISRE Accelerator pairs Marines with developers and designers who work collaboratively both in person and virtually to design, prototype, and then develop and build a minimum viable product (MVP) in 12 weeks.  These two frameworks allow us to continuously innovate from within, creating a pipeline of challenges to be solved over the long term and implementing against them iteratively and quickly.  This rapid implementation creates real metrics that allow us to create quick, measurable value and kill bad ideas before too much time, money, and human capital has been spent on them. Rapid implementation also allows leadership to make quicker decisions on where, when, and how to apply resources.marine-map-photo

3.  Always Be Measuring
Big idea fairies live everywhere, sometimes for a very long time. This can be particularly true in the government where the acquisition lifecycle imposes “shipbuilding timelines” on information technology systems. Successful innovation requires one simple act: always be measuring. Using the Lean Startup methodology to build, learn, and measure in a 12-week cycle provides quantifiable data and user feedback that allows us to validate problem/solution fit quickly. By the time our cohort is pitching to the DIRINT and other leaders and stakeholders, the minimum viable product (MVP) has metrics that validate its value to users as well as prevent redundancy, loss of, or misalignment of capability, funding, and other key resources.

4.  To think differently…be different
Most meetings within the government involve PowerPoint, a conference table, and a bunch of subject matter experts espousing the relative merits or demerits of a point. These meetings can last hours to weeks, and at the end, there might be nothing tangible to show. When we set out to create this culture of innovation, we knew that to get people to do things differently, we had to get them to think differently.

Our first mission value was to create an experience for our Marines, stakeholders, and mentors that was counter to the typical meetings they were used to and focused on establishing a co-creative environment where everyone’s input had value regardless of rank, experience, and skill.  Our technical design meetings and Accelerators use little technology; are not set up as lectures; encourage jeans and your favorite T-shirt; and require constant, active participation. When you walk into our rooms, you will see Marines on their feet, Post-It notes and markers in hand, diagramming, sketching, and plotting furiously on white boards, flip charts, and any other available surface.  They paper the room with problem statements, lean canvases, journey maps, value proposition canvases, process flows, wireframes, and Pixar-worthy storyboards. By the end of the week, you can walk the walls and see the progression of problem to solution in their words, through their eyes, from their point of view. This process takes Marines outside of the normal rank structure that they are accustomed. It is admittedly uncomfortable for them at first. But within hours of the kickoff, these simple tactics result in ideation, collaboration, and production that is evident to them and fundamentally changes how they approach problem solving and conduct meetings when they return to their units.marines-hit-the-beach

5.  Do It Again… and Again… and Again
With any new skill, repetition is important.  Marines don’t learn close order drill with a one-time explanation, they spend countless hours on the drill field until they have mastered it.  We apply the same repetition mindset for our innovation methodology because it creates an environment of continuous learning.  With every repetition, we learn more about our problems and how we can solve them.  We learn which solutions are working and which ones are not.  We learn what techniques for engaging Marines are working and which ones are not. We learn how the operating environment is evolving.  Continuous learning is the objective of our innovation activities, and it is more powerful than the solution itself.  Learning means successes, learning means failures, learning means growth.

November 10, 2016 marks the 241th anniversary of the formation of the Marine Corps; 241 years of adapting to changes and 241 years of innovation. Innovation does not mean that it has to come from external entities.  Sometimes you just need to put Marines in jeans, challenge them to think differently, and give them another opportunity to adapt and overcome.

Hacking for Diplomacy at the State Department – Breakthroughs, breakdowns and relentlessly direct critiques

Time flies. We are already past the midway mark in our new Hacking for Diplomacy course at Stanford, and for both students and instructors, it’s an intellectually and emotionally charged period.


Having logged more than 550 interviews of potential beneficiaries, the teams have delved deep into understanding the problems sourced from the State Department. The challenges are tough – such as how to improve data on refugees who go missing or perish on their journeys, and how to better evaluate the effectiveness of peacekeeping forces. Each team of three to five students is being asked not only to become an expert on these complex topics in 10 short weeks, but also to learn and apply Lean LaunchPad methodology to put forth a solution.

Some teams have produced surprising and delightful minimum viable products (MVPs)  — one is a simple T-shirt, another is an app. These are generating actual customer feedback, which is accelerating the teams’ learning. Other teams are confronting the reality that they are off track and need to zag hard, or step on the gas and push past their comfort zones with their MVPs, before the quarter runs out.

In Weeks 5 and 6, we’ve encouraged students to make leaps and put forth concrete MVPs and get reactions – a scary process. We’ve challenged them to consider how they will acquire, keep and grow their “customer base” – whether those customers are Syrian refugees or U.S. government bureaucrats. We’ve asked them to think about how to get buy-in from beneficiaries, sponsors and other influencers – and to identify potential saboteurs.

As educators, it’s a sensitive inflection point. Besides myself, the teaching team includes Jeremy Weinstein, former deputy to the U.S. ambassador to the United Nations and a Stanford professor of political science; Zvika Krieger, the State Department’s representative to Silicon Valley and senior advisor for technology and innovation; retired U.S. Army Col. Joe Felter, who co-created last spring’s pioneering Hacking for Defense class and is a senior research scholar at the Center for International Security and Cooperation at Stanford; and Steve Weinstein, chief executive of MovieLabs who teaches entrepreneurship at Stanford and UC Berkeley.

With just four weeks remaining, we can’t afford to let teams drift – so we have to deliver what we call “relentlessly direct” feedback. The class is a combination of theory and intensive practice. First and foremost, it is experiential and hands-on. The teams live and die by the Lean Startup credo: “There are no facts inside the building so get the hell outside.” That’s why, just halfway through the class, they’ve already talked to 550 beneficiaries (users, program managers, stakeholders, etc.).

The Lean Methodology requires teams to abandon their preconceived notions of how one builds startups and solve problems – the class is designed to break students out of that all-too-common mindset that they understand customers’ problems, can design a solution and want to get right to work on building it – all without contact with the stakeholders, users, decision makers, etc.

After decades of teaching, I have found that getting students to really change these beliefs cannot be done with reading, case studies or in-class simulations – at least not in the short time we have them in the class. If we really want them to understand how to efficiently and rapidly understand and solve customer problems, we need to immerse them with customers on day one.

And if we want them to understand what life outside the classroom in an early stage venture will look like, then they need to experience chaos, conflicting data, uncertainty and good-enough decision-making for 10 confusing weeks.

We start by pushing the teams incredibly hard to set the pace (and wash out any of those who can’t work at this pace). Teams hit the class running. Before the first class, each team has already spoken to 10 customers, and they are challenged to present their Mission Model Canvases within 20 minutes of walking through the classroom door. Five minutes into a teams first presentation, they get hit with “relentlessly direct” critiques.

If you can’t see the presentation click here

This is a shock to students, many of whom who never have heard a direct criticism of their work in their lives. At the same time, we don’t want to demoralize the students, who are demonstrating incredible commitment. Right now, some teams are feeling very beaten up. But I’m encouraged — every team is working hard and learning a massive amount: how to become domain experts in a very short period; how to work under pressure; what qualities to value in team members. These are lessons that will pay off long after we leave the classroom.

By week 7 (next week), the teams have either embraced the Lean process or we’re not going to get through to them. So at this point in the class, we’ll dial down the tone and tenor of the comments, and become their cheerleaders rather than their taskmasters.

In Week 9 we’ll stop and use the class for “reflection”. We’ve found that getting the teams off the customer discovery treadmill at this point helps them to look back and reflect on what they’ve really learned, not just about their product/customers but more importantly about the Lean processes, themselves, and team work.

L.A. Times China bureau chief Julie Makinen, who is on a JSK journalism fellowship at Stanford, and is part of our mentorship team, has been taking notes on the rough and tumble of Weeks 5 and 6. She shares her observations below.

Six weeks in, many of the students who fought their way into Stanford’s new Hacking for Diplomacy course are feeling like first-time marathoners at Mile 20: They’re spent, they can’t see the finish line yet, and they are questioning their sanity for even signing up for this experience. Ask them how it’s going and they’ll tell you:

“I’m freaking out.”

“I feel like we’ve hit a wall.”

The first four weeks were a frantic but exciting sprint as teams dove into their challenges with their State Department sponsors. Tasked with conducting at least 10 customer discovery interviews per week, the students hoovered up reams of information about topics that many of them came to cold, such as tracking space debris and eliminating forced labor in manufacturing supply chains.

Even as they were trying to become experts on these topics, students were getting crash courses on State Department bureaucracy and Lean LaunchPad methodology. That rapid data uptake right out of the starting gate fueled an early sense of accomplishment, a sort of runner’s high, among many participants.

“The class is incredibly motivating,” student Leonard Bronner said after Week 4. “I’ve worked on a lot of project classes at Stanford. Oftentimes, you are tasked with finding the problem yourself and that alone can take three weeks. Here … you hit the ground running. You feel like you can actually go somewhere, which is empowering.”

But by Week 5, students came under the gun to synthesize everything they had absorbed and make some decisions: Which customers or beneficiaries were they going to target? What problem could they solve for them, and with what product? What pains could they take away for these customers, or what gains could they offer? Would their product concept prompt potential customers to snatch it out of their hands and ask: Can I really have this, and how soon? Or would their target market just shrug?

Some teams came to class in Week 6 acknowledging that they had met dead ends. Team Space Evaders, which is working on preventing collisions in space, admitted they were having trouble homing in on a customer to serve. Only one person was really excited about their MVP. Steve Blank encouraged the students to take this as a “big learning point” and to go back over the data and interviews they had already collected to see if there was an opportunity that they missed.

If you can’t see the presentation click here

“Be frustrated, not embarrassed,” he counseled them. “Life will be like that – in a startup, you’d either be talking about shutting down or pivoting like mad.”

Other teams came in for some even sharper feedback from the instructors. Team Aggregate DB, which is working on how the State Department can better gather and leverage information on informal leaders in foreign countries, was called on the carpet for failing to call on some high-level contacts provided by the professors that could potentially change their mission model. (Not following up on a teaching team lead is cardinal sin for a team supposedly driving on customer discovery.)

If you can’t see the presentation click here

Another team that is working on how to bring together technology, government, and communities to combat violent extremist messaging was told in no uncertain terms to go back to the drawing board with their MVP because it was too far afield from the initial problem posed by the State Department. “We just fired their idea,” Blank said as the team cut short its slide show presentation and went back to their seats.

A palpable tension, even apprehension, started to settle over the room. After a bit, I started wishing that 1970s game show host Chuck Barris would wander in and bang a gong, just to clear the air and say, “hey kids, none of this is personal.”

Soon though, Blank bounded to the front of the class and offered the students a pep talk. All of this, he insisted, was part of the normal, if messy and sometimes uncomfortable, process of trying to get stuff done in the real world.

“The teaching team is tearing you up and trashing your slides,” he said. “Don’t take it personally. We are asking you to accomplish unreasonable things in an impossibly limited amount of time. The journey is hard, but when the class is over you’ll look back and be amazed about what you accomplished. The chaos, uncertainty and pain lasts a short time, but the skills you learn here will be with you forever.”

“These students are really taking these problems to heart,” instructor Steve Weinstein said. “They feel bad if they’re having trouble solving the problems. And it’s kind of cool. They see the complexity their sponsors [in State] are facing and they internalize that complexity.”

T-Shirts and Sharpie Markers
Several teams did find their stride in Weeks 5 and 6. The four students working on the problem of how to improve data on refugees who go missing or perish on their journeys hit upon an elegantly simple “minimum viable product,” or MVP: What if we could just convince migrants to write the phone number of a friend or relative on their clothing with a Sharpie permanent marker? Not their own name, or any other identifying information. That way, if tragedy were to strike the migrants en route to their destinations and their bodies were found, those authorities handling the corpses could use this contact information to inform the deceased’s loved ones.  Last year alone, more than 3,700 people died at sea in the Mediterranean and only about a third of the bodies were identified.

If you can’t see the presentation click here

Initial feedback from refugees themselves was positive – a 22-year-old from Eritrea who crossed the Mediterranean to Europe from Libya told the team he would have put his phone number on his shirt – had he thought ahead of time to do so. If he had died, he said, at least he could feel like his family would have closure.

Yet there might be challenges to get migrants to “buy into” this simple act, the team learned. Could refugees find permanent markers in their poor and war-torn countries? What about Muslim women who wear all-black abayas, how would you write on their garments? Would smugglers object?

The team brainstormed who might be effective influencers to inform potential migrants about this tactic before they set out on their journeys. They identified refugee support groups on Facebook and WhatsApp as good starting points, as well as NGOs.

But even if refugees could be encouraged to start marking their clothing with phone numbers, would the people who find the bodies be incentivized to use that information? The team recognized that buy-in would also be needed on the side of forensic examiners and local law enforcement authorities.

The team’s interviewing suggested that human rights-minded entities like the European Parliament as well as NGOs, the media and each country’s Interior Ministry might be enlisted to encourage first responders to take advantage of these phone numbers. These groups might also offer resources (such as interpreters) to help local authorities make the calls.

As much as the teaching team appreciated their low-cost, low-risk solution, they encouraged the students to aim higher and design a prototype for next week that might entail more risk for refugees but more data that would better identify refugees if they perished along their journey.

An App to Streamline Field Reports on Peacekeepers
Another team that notched some wins in Weeks 5 and 6 is tackling the issue of how to help State more effectively assess the peacekeeping forces that the U.S. is funding. Currently, evaluators write up narrative reports on their fact-finding missions, sometimes months after the trip is completed. There’s no standardization of the reports, and no easy way to make comparisons across different peacekeeping units to measure their effectiveness to measure their effectiveness.

If you can’t see the presentation click here

For its MVP, the team prototyped a simple mobile app that features drop-down menus, check boxes and small text boxes that evaluators could use even while they are in the field to take notes. This data could be transmitted back to headquarters quickly for immediate feedback, and serve as a reference for evaluators to generate more extensive reports once they’re back in the office.

By equipping all evaluators with such a tool while they’re in the midst of the process, State may be able to standardize metrics and readily compare and analyze the performance of different peacekeeping units. More extensive written reports might even be rendered obsolete.

Shown this prototype, the team’s State Department sponsor was highly enthused, telling the team, “I absolutely love this.” Further proof of the sponsor’s buy-in came when she said she wanted to send the prototype around the office and get feedback.

The team followed Blank’s suggestion to show their potential customers a prototype they would “grab out of your hand” because they wanted it so badly, even if they hadn’t figured out how to make that product. While this team may have gotten that golden reaction from their sponsors, it remains to be seen whether they can deliver on what they proposed.

Takeaways & Week 7’s Special Guest
Shira McKinlay, a 41-year-old former lawyer from Orlando who is working on the peacekeeping evaluation team, said while she’s gratified by the sponsor’s reaction, she is concerned that their app may never be deployed to the field. Hacking for Diplomacy has made clear to her how challenging it can be to get things done in an institution as complex and overtaxed as the State Department.

“Everyone’s so overwhelmed and busy,” she said. “There are many different interests, and you can’t build ‘customers’ like you would in a business. When money is not a motivation, it changes the dynamic.”

But higher-ups in State are paying attention to the course — Deputy Secretary of State Tony Blinken will be on hand in Week 7 to listen to student presentations.

In the next few days, student teams will be focusing on how to deploy their product and the special dynamics that come with doing so within the State Department context – such as security concerns, cultural sensitivities, interagency dynamics and State’s relatively limited budget.

McKinlay, who’s studying at the Diplomatic Academy of Vienna and is currently on an exchange program at Stanford, admitted Hacking for Diplomacy has been more work than she ever anticipated. But the relentless interviewing and customer discovery process is teaching her, she said, to be comfortable being a bit more aggressive and “not take no for an answer.”

“I’m trying to do a thesis on climate change and international law, and there are all these people that before I probably wouldn’t have called,” she said. “After this, I feel like if I see someone’s name in an article and I want to talk to them, I’m going to contact them.”

Kate Boudreau, a 20-year-old who’s studying computer science, said she’s supercharged her interviewing skills. “I’ve really learned a lot about how to connect with people and how to ask probing questions,” she said.

Boudreau, who is on the Space Evaders team, added that she had been considering trying to get a job as a consultant after graduation and Hacking for Diplomacy has reaffirmed that direction.

“It’s so fun to jump into an organization I know nothing about and then try to figure things out,” she said. “And it’s also made me realize how important innovation is – I want to be somewhere where that’s a priority.”

The State Department Meets the Lean Startup – Hacking For Diplomacy

h4dip-screen-shotThe academic year is in full swing at Stanford and already we’re deep into our new Hacking for Diplomacy course. Building off last spring’s pioneering Hacking for Defense class, which sought to connect Silicon Valley’s innovation culture and mindset to the Pentagon and the intelligence community, we’ve now expanded our horizons to the Department of State.

The cross-disciplinary class brings students from widely divergent backgrounds together in teams of three to five, each aiming to tackle a gnarly international problem vexing Foggy Bottom in just 10 weeks by applying Lean LaunchPad methodology.

Guiding, drilling and grilling these teams are Jeremy Weinstein, former deputy to the U.S. ambassador to the United Nations and a Stanford professor of political science; Zvika Krieger, the State Department’s representative to Silicon Valley and senior advisor for technology and innovation; retired U.S. Army Col. Joe Felter, who co-created Hacking for Defense and is a senior research scholar at the Center for International Security and Cooperation at Stanford; Steve Weinstein, chief executive of MovieLabs who teaches entrepreneurship at Stanford and UC Berkeley; and yours truly, Steve Blank.

In addition, we’ve recruited a host of mentors, including folks from Google and the cloud computing firm SalesForce; Stanford Law School; and veteran State Department employees now engaged in studies, research or retirement in the Bay Area.julie-makinen

L.A. Times China bureau chief Julie Makinen, who is on a JSK journalism fellowship at Stanford, joined in at the last minute and is helping students with customer discovery techniques, particularly how to find and interview people. I’ve invited Julie to share some observations here on the class to date.
She writes:

When I stumbled into the introductory session for Hacking for Diplomacy a few weeks ago, there was a palpable, kinetic charge in the room at Stanford’s Tresidder Memorial Union. My reporter’s Spidey Sense began to tingle. It feels like something big is going on here, I thought.

The energy was also fed by the crowd of prospective students — a motley, enthusiastic and clearly wicked smart group including engineering PhDs and computer science whiz kids, U.S. Army veterans and mid-career MBAs hailing not just from the United States but from countries including Saudi Arabia, India, France, Israel and Austria.

And ultimately, it was the get-your-hands-messy conceit of the whole shebang:  Make the cut for this class and we are going to throw you headlong at some major real-world problems put forth by the U.S. State Department and see what products you can come up with to solve them. Along the way, you will actively learn the Lean LaunchPad methodology, the framework first developed for business start-ups.

This is no class for slackers, the students were admonished: You will work at warp speed in teams, trying to get your arms around challenges that experts have failed to wrestle to the ground, like tracking refugees missing at sea, countering violent extremism online, and evaluating the effectiveness of peacekeeping forces.

You will get schooled on the labyrinthine bureaucracy of the State Department, its alphabet soup of acronyms, its secret language of abbreviations. You will sniff out and pin down stakeholders and beneficiaries – from Washington functionaries to Syrian migrants – by conducting at least 10 interviews each week. You will meet and interact with VIPs.

You will learn a new rubric for asking questions, posing hypotheses and verifying those ideas. You will prototype solutions, developing and refining them over multiple rounds of iteration and feedback.  Your team will pen detailed updates of your progress and post them online each week for fellow students to learn from and for teachers to evaluate.

At every class, your team must stand up, explain – and defend – what you did in the last seven days and how you have moved the ball forward. In the back row, perched like a panel of fair yet tough judges, the teaching team will listen intently but also interrupt you without mercy, forcing you to drill deeper or acknowledge weaknesses. Simultaneously, your classmates will write critiques of your presentation on a sprawling shared Google spreadsheet in real time.

This is not for the thin-skinned or the weak-kneed. It’s “Shark Tank” meets “American Idol,” with heavy doses of foreign service school, business school and Journalism 101 thrown in. Here’s how it’s been going:

Weeks 1 and 2 – Mission Model, Customer Discovery & Figuring Out Who’s Who at State
For the students, the first week was a mad scramble through a gantlet of tasks: assemble teams, decide which challenge proposed by the State Department to work on, submit a written application and pass an interview with the teaching team. In the end, teams were selected to work on seven different challenges:

The teaching team began with an introduction to the Mission Model Canvas, a slightly modified version of the Business Model Canvas. This one-page schematic with 9 boxes, which students must revise and update each week, acts as the teams’ anchor throughout the class – guiding their research and discovery process and keeping them focused on who their customers or beneficiaries are, what value the team can bring to them with a new product, how to define success, and key tasks and relationships to identify and leverage.mission-model-canvas-sm

Krieger and Jeremy Weinstein took the class on a blitzkrieg tour through the State Department’s byzantine organizational chart, introducing different bureaus and their functions and explaining how their missions often intersect, overlap and sometimes collide.  Students also got a crash course on customer discovery in the foreign policy universe.  “Get out of the building!” Blank encouraged the class. “Talk to people in person or face-to-face on Skype! You need to see their pupils dilate!”

state-dept-org-chart“You have to be comfortable with the amount of uncertainty at the start of class,” said Leonard Bronner, a master’s student in statistics from Austria who is on the team working on the question of missing refugees. “You have to really actively work to consolidate everything that’s coming at you.”

In theory, teams were to start by liaising with their designated sponsor in the State Department, who is supposed to make time each week to check in with the students, offer information and guidance and suggest other people to talk to. In practice, students learned that civil servants are busy and sometimes hard to reach. The time difference between Washington and California often doesn’t help. For many, the first week was a jumble of what felt like disjointed interviews and a mad scramble for even basic information – who are the relevant people and organizations working on or affected by this problem? What do they do? How do they interact?

“In week one, everyone is totally disoriented,” said Blank. “That’s how it works.”

When students did make contact with their State Department sponsors and started asking questions, for some it quickly became apparent that sponsors themselves had trouble articulating exactly what problem they really needed to solve. Other teams learned that there was disagreement within and between different State Department offices about what they hoped the Stanford teams would be doing — or whether they should be working on the problem at all. Some teams got a quick crash course in turf wars and Government Bureaucracy 101.

“In some cases, the sponsors confuse the symptoms of the problem with the root cause,” says Felter. “Last spring [in Hacking for Defense], some of our sponsors didn’t have a good understanding of their problems. At a minimum, coming from the outside and using these tools we give them, students can help their sponsors understand their problems better.”

Because customer discovery is such a key part of the Lean LaunchPad methodology, the student teams have had to rapidly ramp up their interviewing skills. Many realized that it can be a challenge to find sources to speak with, or extract useful information once you do locate them. How do you go about talking to a Syrian refugee who crossed the Mediterranean on a rickety boat? How do you get a supply chain manager for a clothing manufacturer to take your call?

“We are asking students to do a lot in 10 weeks,” admitted Krieger. “They have to tackle a challenge that is new to them. In many cases, they are learning an entirely new field. Like the space team – no one on that team knew about satellites a few weeks ago.”

“Next they have to learn in minute detail about the State Department bureau,” he added. “And then they have to learn about other agencies, because almost all of these problems have an interagency dimension, whether that’s with the Department of Defense, USAID, NASA or beyond.”

And that’s just all prelude to the ultimate work.

“Finally, they have to find an opportunity – a ‘pain point’ as we say, come up with a solution, and prototype it in like six weeks,” said Krieger. “I’m impressed with what they’ve been able to do already. They probably know more about their corners of the State Department than most people at the State Department.”

Weeks 3 and 4 – Fast-tracked by Secretary Kerry, Diving Into the Value Proposition and Defining Mission Achievement
After being put through the ringer of the first two weeks, Hacking for Diplomacy teams got a boost in Week 3 when Secretary of State John Kerry made a swing through Silicon Valley and met with the students.

Asked by business school student Kaya Tilev whether the solutions the students were working on had any actual chance of being implemented, Kerry offered words of encouragement.

kerry-and-students“I have absolute confidence if you come up with a viable solution it is going to be implemented, adopted, and institutionalized,” Kerry said.  “You have a fast track for making that happen because you’re in the program and you know Zvika. So you’re on the right track. Just come up with the deal, okay?”

Later, Kerry would tweet: “Brilliant minds are applying technology to world’s toughest problems. Their perspective will inform our diplomatic engagement going forward.”

Krieger said he’s been overwhelmed by the level of interest among State Department officials in the class. Although the department has long had ways for university students to work on diplomatic issues, the end products have tended to be policy papers or think-tank like reports penned by political science or international relations majors. So the idea that students with engineering or other technical backgrounds might create actual tangible products for diplomatic ends feels fresh, novel and necessary.

“A lot of people are really excited about this, and I vacillate between being excited myself and trying to calibrate expectations” within State, said Krieger.

Unlike the Defense Department, which Krieger describes as “comfortable with technology” and flush with funds, State is a relative tech backwater with a much more limited budget. “Thinking in terms of technology solutions is a paradigm shift for the State Department,” he said.

In Week 3, teams began to really focus in on potential customers/beneficiaries. They rigorously tried to identify what “pains” could be alleviated for these people – and what “gains” might be offered to them that would be irresistible? What value proposition could the teams come up with to make each of their prospective beneficiaries’ lives better?

Many teams made diagrams to help drill down to beneficiaries.

Consider Team Space Evaders, which is working on the satellite collision issue. The team made a detailed flow chart of how data about satellite positioning and potential collisions is shared by entities ranging from the FAA to the Department of Defense and commercial operators. Then they started looking at which customers they might serve – is it a bureaucrat in the State Department? Or is it satellite operators, or satellite insurers? Who has what problems, and which ones might be ripe for solving?

space-evaders-diagramTeam Exodus, working on improving coordination among groups trying to assist Syrian refugees, made an elegant diagram of dozens of organizations involved the “customer workflow” of refugees, from U.N. agencies to the Red Cross and Doctors Without Borders.

Team 621, working on the missing refugees challenge, literally drew a map of the Mediterranean, showing departure countries, target arrival countries, and location of where boats sink.

team-621-diagramIn Week 4, a number of teams were sketching out minimal viable products, or MVPs. These super basic “products” are stabs at something that might address the pains and gains of their customers. team-trace-mvpTeams must take these products out to the real world and ask potential users for feedback. Would anyone want to buy or use these things – assuming the teams could even make them? Would the product be considered essential, a must-have? Or just “nice to have?” Would the product help each beneficiary achieve their core mission?

exodus-week-5-mvpThe rigorous, continual customer discovery process demands fortitude, students say. “Every week we learn something that completely invalidates something we thought we knew,” said Anusha Balakrishnan, a 23-year-old master’s student in computer science who is working on countering violent extremism. “So we have to keep iterating.”

Balakrishnan said the class has prompted her to think a little differently about how she might use her skills after graduation. “I didn’t realize before this how many companies are interested in applying machine learning to problems like countering violent extremism,” she said. “Even if I’m working at a tech company in the future, maybe I can still do stuff like this. So it’s opened my eyes to that.”

Even in the first few weeks, the response to the students’ work so far has been gratifying, she added.

“It’s great that people are taking us so seriously when we contact them and tell them what we are doing – it feels that we could actually add value to NGOs or the State Department,” she said.  “I didn’t think that was possible before this class.”

Why Tim Cook is Steve Ballmer and Why He Still Has His Job at Apple

What happens to a company when a visionary CEO is gone? Most often innovation dies and the company coasts for years on momentum and its brand. apple-equals-microsoftRarely does it regain its former glory.

Here’s why.

Microsoft entered the 21st century as the dominant software provider for anyone who interacted with a computing device. 16 years later it’s just another software company.

After running Microsoft for 25 years, Bill Gates handed the reins of CEO to Steve Ballmer in January 2000. Ballmer went on to run Microsoft for the next 14 years. If you think the job of a CEO is to increase sales, then Ballmer did a spectacular job. He tripled Microsoft’s sales to $78 billion and profits more than doubled from $9 billion to $22 billion. The launch of the Xbox and Kinect, and the acquisitions of Skype and Yammer happened on his shift. If the Microsoft board was managing for quarter to quarter or even year to year revenue growth, Ballmer was as good as it gets as a CEO. But if the purpose of the company is long-term survival, then one could make a much better argument that he was a failure as a CEO as he optimized short-term gains by squandering long-term opportunities.

How to Miss the Boat – Five Times
Despite Microsoft’s remarkable financial performance, as Microsoft CEO Ballmer failed to understand and execute on the five most important technology trends of the 21st century: in search – losing to Google; in smartphones – losing to Apple; in mobile operating systems – losing to Google/Apple; in media – losing to Apple/Netflix; and in the cloud – losing to Amazon. Microsoft left the 20th century owning over 95% of the operating systems that ran on computers (almost all on desktops). Fifteen years and 2 billion smartphones shipped in the 21st century and Microsoft’s mobile OS share is 1%. These misses weren’t in some tangential markets – missing search, mobile and the cloud were directly where Microsoft users were heading.  Yet a very smart CEO missed all of these.  Why?

Execution and Organization of Core Businesses
It wasn’t that Microsoft didn’t have smart engineers working on search, media, mobile and cloud. They had lots of these projects. The problem was that Ballmer organized the company around execution of its current strengths – Windows and Office businesses. Projects not directly related to those activities never got serious management attention and/or resources.

For Microsoft to have tackled the areas they missed – cloud, music, mobile, apps – would have required an organizational transformation to a services company. Services (Cloud, ads, music) have a very different business model. They are hard to do in a company that excels at products.

Ballmer and Microsoft failed because the CEO was a world-class executor (a Harvard grad and world-class salesman) of an existing business model trying to manage in a world of increasing change and disruption. Microsoft executed its 20th-century business model extremely well, but it missed the new and more important ones. The result?  Great short-term gains but long-term prospects for Microsoft are far less compelling.

In 2014, Microsoft finally announced that Ballmer would retire, and in early 2014, Satya Nadella took charge. Nadella got Microsoft organized around mobile and the cloud (Azure), freed the Office and Azure teams from Windows, killed the phone business and got a major release of Windows out without the usual trauma. And is moving the company into augmented reality and conversational AI. While they’ll likely never regain the market dominance they had in the 20th century, (their business model continues to be extremely profitable) Nadella likely saved Microsoft from irrelevance.

What’s Missing?
Visionary CEOs are not “just” great at assuring world-class execution of a tested and successful business model, they are also world-class innovators. Visionary CEOs are product and business model centric and extremely customer focused.

The best are agile and know how to pivot – make a substantive change to the business model while or before their market has shifted. The very best of them shape markets – they know how to create new markets by seeing opportunities before anyone else. They remain entrepreneurs.

One of the best examples of a visionary CEO is Steve Jobs who transformed Apple from a niche computer company into the most profitable company in the world. Between 2001 to 2008, Jobs reinvented the company three times. Each transformation – from a new computer distribution channel – Apple Stores to disrupting the music business with iPod and iTunes in 2001; to the iPhone in 2007; and the App store in 2008 – drove revenues and profits to new heights

apple-2001-to-08-arcThese were not just product transitions, but radical business model transitions – new channels, new customers and new markets–and new emphasis on different parts of the organization (design became more important than the hardware itself and new executives became more important than the current ones).

Visionary CEOs don’t need someone else to demo the company’s key products for them. They deeply understand products, and they have their own coherent and consistent vision of where the industry/business models and customers are today, and where they need to take the company.  They know who their customers are because they spend time talking to them. They use strategy committees and the exec staff for advice, but none of these CEOs pivot by committee.

Why Tim Cook Is the New Steve Ballmer
And that brings us to Apple, Tim Cook and the Apple board.

One of the strengths of successful visionary and charismatic CEOs is that they build an executive staff of world-class operating executives (and they unconsciously force out any world-class innovators from their direct reports). The problem is in a company driven by a visionary CEO, there is only one visionary. This type of CEO surrounds himself with extremely competent executors, but not disruptive innovators. While Steve Jobs ran Apple, he drove the vision but put strong operating execs in each domain – hardware, software, product design, supply chain, manufacturing – who translated his vision and impatience into plans, process and procedures.

slide1When visionary founders depart (death, firing, etc.), the operating executives who reported to them believe it’s their turn to run the company (often with the blessing of the ex CEO).  At Microsoft, Bill Gates anointed Steve Ballmer, and at Apple Steve Jobs made it clear that Tim Cook was to be his successor.

Once in charge, one of the first things these operations/execution CEOs do is to get rid of the chaos and turbulence in the organization. Execution CEOs value stability, process and repeatable execution. On one hand that’s great for predictability, but it often starts a creative death spiral – creative people start to leave, and other executors (without the innovation talent of the old leader) are put into more senior roles – hiring more process people, which in turn forces out the remaining creative talent. This culture shift ripples down from the top and what once felt like a company on a mission to change the world now feels like another job.

As process oriented as the new CEOs are, you get the sense that one of the things they don’t love and aren’t driving are the products (go look at the Apple Watch announcements and see who demos the product).

Tim Cook has now run Apple for five years, long enough for this to be his company rather than Steve Jobs’. The parallel between Gates and Ballmer and Jobs and Cook is eerie. Apple under Cook has doubled its revenues to $200 billion while doubling profit and tripling the amount of cash it has in the bank (now a quarter of trillion dollars). The iPhone continues its annual upgrades of incremental improvements. Yet in five years the only new thing that managed to get out the door is the Apple Watch. With 115,000 employees Apple can barely get annual updates out for their laptops and desktop computers.

But the world is about to disrupt Apple in the same way that Microsoft under Ballmer faced disruption. Apple brilliantly mastered User Interface and product design to power the iPhone to dominance. But Google and Amazon are betting that the next of wave of computing products will be AI-directed services – machine intelligence driving apps and hardware. Think of Amazon Alexa, Google Home and Assistant directed by voice recognition that’s powered by smart, conversational Artificial Intelligence – and most of these will be a new class of devices scattered around your house, not just on your phone. It’s possible that betting on the phone as the platform for conversational AI may not be the winning hand.

It’s not that Apple doesn’t have exciting things in conversational AI going on in their labs. Heck, Siri was actually first. Apple also has autonomous car projects, AI-based speakers, augmented and virtual reality, etc in their labs. The problem is that a supply chain CEO who lacks a passion for products and has yet to articulate a personal vision of where to Apple will go is ill equipped to make the right organizational, business model and product bets to bring those to market.

Four Challenges for the Board of Directors
The dilemma facing the boards at Microsoft, Apple or any board of directors on the departure of an innovative CEO is strategic: Do we still want to be a innovative, risk taking company?  Or should we now focus on execution of our core business, reduce our risky bets and maximize shareholder return.

Tactically, that question results in asking: Do you search for another innovator from outside, promote one of the executors or go deeper down the organization to find an innovator?

Herein lies four challenges. Steve Jobs and Bill Gates (and 20th century’s other creative icon -Walt Disney) shared the same blind spot: They suggested execution executives as their successors. They confused world-class execution with the passion for product and customers, and market insight. From the perspective of Gates there was no difference between him and Ballmer and from Jobs to Cook. Yet history has shown us for long-term survival in markets that change rapidly that’s definitely not the case.

The second conundrum is that if the board decides that the company needs another innovator at the helm, you can almost guarantee that the best executor – the number 2 and/or 3 vice president in the company – will leave, feeling that they deserved the job. Now the board is faced with not only having lost its CEO, but potentially the best of the executive staff.

The third challenge is that many innovative/visionary CEOs have become part of the company’s brand. Steve Jobs, Jeff Bezos, Mark Zuckerberg, Jeff Immelt, Elon Musk, Mark Benioff, Larry Ellison. This isn’t a new phenomenon, think of 20th-century icons like Walt Disney, Edward Land at Polaroid, Henry Ford, Lee Iacocca at Chrysler, Jack Welch at GE and Alfred Sloan at GM. But they’re not only an external face to the company, they were often the touchstone for internal decision-making. Years after a visionary CEO is gone companies are still asking “What would Walt Disney/Steve Jobs/Henry Ford have done?” rather than figuring out what they should now be doing in the changing market.

Finally, the fourth conundrum is that as companies grow larger and management falls prey to the fallacy that it only exists to maximize shareholder short-term return on investment, companies become risk averse. Large companies and their boards live in fear of losing what they spent years gaining (customers, market share, revenue, profits.) This may work in stable markets and technologies. But today very few of those remain.

In the 21st Century an Execution CEO as a Successor Increasingly May be The Wrong Choice
In a startup the board of directors realizes that risk is the nature of new ventures and innovation is why they exist. On day one there are no customers to lose, no revenue and profits to decline. Instead there is everything to gain. In contrast, large companies are often risk-averse engines – they are executing a repeatable and scalable business model that spins out the short-term dividends, revenue and profits that the stock market rewards. And an increasing share price becomes the reason for existing. The irony is that in the 21st century, the tighter you hold on to your current product/markets, the likelier you will be disrupted. (As articulated in the classic Clayton Christensen book The Innovators Dilemma, in industries with rapid technology or market shifts, disruption cannot be ignored.)

Increasingly, a hands-on product/customer, and business model-centric CEO with an entrepreneurial vision of the future may be the difference between market dominance and Chapter 11. In these industries, disruption will create opportunities that force “bet the company” decisions about product direction, markets, pricing, supply chain, operations and the reorganization necessary to execute a new business model.  At the end of the day CEOs who survive embrace innovation, communicate a new vision and build management to execute the vision.

Lessons Learned

  • Innovation CEOs are almost always replaced by one of their execution VPs
  • If they have inherited a powerful business model this often results in gains in revenue and profits that can continue for years
  • However, as soon the market, business model, technology shifts, these execution CEOs are ill-equipped to deal with the change – the result is a company obsoleted by more agile innovators and left to live off momentum in its twilight years
A shorter version of this article previously appeared in the Harvard Business Review.

The 11 Bad Habits Killing Innovation in Your Company

osterwalder-photoAlexander Osterwalder invented the Business Model Canvas, co-founded and was the lead author of Business Model Generation which sold a million copies in 30 languages.

Alexander and I often collaborate on new ideas for corporate innovation.  Here’s his guest post on what bad habits to avoid inside of a company.

Big companies have great execution habits to manage and improve successful business models and value propositions. But the habits that foster execution can easily kill new growth initiatives inside your company.

Bad Habit #1: The current business model dominates the agenda
In most companies the future suffers at the expense of the present. Companies are great at improving their existing business model and value propositions, but fall short when it comes to inventing entirely new business models, value propositions, and growth engines. In fact, by the time a company realizes it needs to reinvent itself for future success, it’s often too late. This happens because managing the present often takes oxygen away from inventing the future. Rita McGrath, a Columbia Business School professor says, “there’s pleasing today’s customers and there’s developing tomorrow’s business.” You need to be excellent at both.

Remedy: Create a protected space in your org chart where you invent and test new business models and value propositions. Equip this “space” with power and prestige. Become an ambidextrous organization — one that is excellent at managing and improving your existing business, alongside inventing new ones.

Bad Habit #2: One-size-fits-all decision making hurts speed & inventiveness
Companies that grow in size and scale proven products and services can quickly fall into a trap of slowness, unthoughtful risk aversion, and failure to experiment. As Jeff Bezos puts it, one-size-fits-all decision making “hurts speed and inventiveness” inside large organizations. In fact, Bezos constantly adjusts Amazon’s culture to ensure that the company never slows down and loses its entrepreneurial and nimble approach to finding future business success.

Remedy: Amazon distinguishes between non-reversible decisions with substantial sunk costs (like e.g. investing in a new warehouse in Amazon’s case), and reversible decisions like experimenting with a new offer. The former requires slow and careful decision making. The latter requires speed and agility.

Bad Habit #3: Insisting on untested and detailed business plans
Most established companies require detailed business plans for new ideas. This results in carefully crafted and thought-through documents with detailed spreadsheets and a great focus on how an idea will be implemented. However, the first goal of an innovator should not be to think hard about an idea and describe its implementation. First and foremost, an innovator’s job should be to rapidly, cheaply, and continuously test and adapt ideas until there’s enough evidence from the field to prove they will work. Only the latter helps avoiding big flops because it systematically reduces the risk and uncertainty of new ideas. Business plans actually maximize the risk of failure because of the focus on executing an unproven idea rather than testing it.

Remedy: Use business plans only for execution of existing businesses. Don’t ask innovators for business plans. Instead, implement processes that force innovators to systematically prototype and test ideas, reduce risk and uncertainty, and ultimately provide the evidence that an idea will work and is worth doing.

Bad Habit #4: Opinions matter more than evidence
Senior leaders acquire a lot of knowledge and experience about their business over the course of their  career. Unfortunately, this knowledge may be irrelevant when it comes to new value propositions and new business models. For example, the knowledge that Kodak’s leaders acquired during their successful decades in analog film didn’t equip them for digital photography. Quite the contrary. The rules to compete in the digital age are completely different. That’s why it’s so important for companies to “get out of the building” and interact with customers. Steve Blank, father of the Lean Startup movement, stresses that you will never know enough about your customers if you are holed up inside a boardroom. A good idea might still be a bad idea because customers don’t care about it. Michael Schrage, a research fellow at MIT, emphasizes that “a testable idea is better than a good idea”.

Remedy: Educate leaders that judging ideas for new value propositions, business models, and growth engines requires evidence from the field rather than just “expert opinion” from leadership. Implement processes that judge ideas based not on how they look, but based on the evidence from the field that support them.

Bad Habit #5: Outsourcing customer discovery and testing
Large companies have a habit of hiring outside agencies to do market research and customer discovery. That’s dangerous when it comes to developing new value propositions, business models, and growth engines. You can’t hire outside professionals to test and learn from customer interactions and make decisions for you. New ideas require many rapid iterations between prototyping, immediately testing with customers, and then deciding how to adapt your idea based on the acquired insights.

Remedy: For radically new ideas you should defer hiring outside agencies until you’ve found product/market fit. Instead, roll up your sleeves and internalize the hard work of rapid prototyping, testing, learning, and deciding. Third parties can help you with the process, but they can’t do the work for you.

Bad Habit #6: Senior leadership too busy for hands-on innovation
Senior leaders are very busy and time pressed people. Typically, they see the “getting out of the building” to test ideas with customers as a task to be performed by subordinates. But leaders have to be more than just sponsors of new business ideas. Decision makers are the ones who can make things happen. They are the ones who need to feel the market and talk to (potential) customers to learn that some of their initial assumptions or strong opinions might be completely wrong. Equipped with these market insights they can help move things faster.

Remedy: Distinguish between senior leaders who manage the present like running factories, and senior leaders who are involved in creating the future and need to “get out of the building”.

Bad Habit #7: Obsessing about competitors rather than customers
Unfortunately, many companies are more obsessed by their competition than their customers. Your customers are far more important than your competitors. Your (potential new) customers can tell you how to beat your competitors. Customer have the evidence your organization needs to validate or invalidate new business ideas and potential growth engines. That doesn’t mean you should completely ignore the competition. After all, business models and value propositions aren’t designed in a vacuum. However, competitors should not be your primary focus. As Steve Blank says, “You can’t drive forward by looking in the rear view mirror.”

Remedy: Obsess over your customers first when developing and testing new value propositions, business models, and growth engines. Then, evaluate how these new ideas perform in the competitive landscape.

Bad Habit #8: Focus on technology risk at the expense of other risks
New business ideas face many different risks. The California design firm IDEO distinguishes between three types of risk when they assess prototypes: desirability, feasibility, viability. Desirability is about the risk of your customers not being attracted by your new value proposition. Feasibility is about technology and infrastructure risks. Viability is about financial risks. We added a fourth risk, adaptability. Adaptability is about the risk of a business model and value proposition not being fit for evolving external factors, like competition, technology change, or regulation (risk: external threats).

Remedy: Make sure you test all four types of risks: desirability, feasibility, viability, and adaptability.

Bad Habit #9: Innovation is career limiting
In many companies being an innovator is not an attractive career path. First, in most organizations any type of failure is seen as a negative for your career. Yet good innovation processes require rapid experimentation and failure to gain insights, adapt, and ultimately succeed. Second, corporate incentives are geared to rewarding execution, where failure is not an option. Third, in most companies, innovation is still seen as a department for pirates and “the crazy ones” who really add no value to revenue and profit. And finally, prestige in companies is measured by who commands the largest budget and staff. But great innovation programs always start small.

Create different incentive systems for the people focused on execution, and the people focused on innovation. Make innovation a prestigious job in your company. After all, the innovators are ensuring your organization’s survival in an age of constant change.

Bad Habit #10: Innovation is siloed from Execution
Companies struggle to get the “execution engine” and “innovation engine” to collaborate, rather than to compete. Rather than realizing that managing the present and inventing the future are equally important and should be equally resourced, they often fight for the same resources. Often the execution engine deprives the innovators from access to valuable resources like customers, brand, or skills. That means the innovators end up competing without any competitive advantage against the more nimble and agile startups.

Remedy: Create a culture where executors and innovators collaborate because they understand each other’s value to the organization. Create processes and incentives that grant innovators access to customers, brands, and skills so they can outcompete the more nimble and agile startup ventures.

Bad Habit #11: Integrating new ideas into the execution engine too quickly
New ideas are fragile and they need to be carefully nurtured and scaled before they are integrated into the execution engine with its rigid processes, key performance indicators rules, and procedures. If you integrate new ideas before they fully mature you might kill them. For example, Nespresso, the successful daughter company of Nestlé, only survived and thrived because they were physically distant from Nestlé’s headquarters.

Remedy: Protect new ideas until they fully mature; for example, like in Nestlé’s case, by locating the project outside of the company’s headquarters.

Join Alex Osterwalder at the Business Model Canvas Masterclass in San Francisco, Nov. 3 – register here



Entrepreneurs are Everywhere Show No. 45: Dan Miller and Brian Zuercher

I always told myself that I would stop pushing forward when there was an overwhelming force from the outside saying that this is not working. But even when I reached that point, I continued to try to brute force it into existence. I wound up losing a lot.

We followed every test and experimental process from the get-go but we didn’t tell our investors we were doing that. They still thought we were building what we had presented in a PowerPoint slide. When they found out, they questioned my decision-making and me as an entrepreneur.

The same passion that got your startup idea off the ground can blind you to signs that your company is failing.

And not keep investors informed about changes to your business model can have serious consequences.

How to recognize when it’s time to pull the plug on your startup idea, and why founders can’t operate afford to operate in a vacuum were the focus on today’s Entrepreneurs are Everywhere radio show.

The show follows the journeys of founders who share what it takes to build a startup – from restaurants to rocket scientists, to online gifts to online groceries and more. The program examines the DNA of entrepreneurs: what makes them tick, how they came up with their ideas; and explores the habits that make them successful, and the highs and lows that pushed them forward.


Dan Miller

Joining me in the Stanford University studio were


Brian Zuercher

Listen to my full interviews with Dan and Brian by downloading them from SoundCloud here and here

(And download any of the past shows here.)

Clips from their interviews are below.

Dan Miller is the co-founder and CEO of Level Therapy, which provides access to psychotherapists through video, voice, and text. Before Level, Dan was the founder and CEO of Freshsessions, the world’s first marketplace for musicians to find and book studio time anywhere.

Before setting out on his own, Dan held various product, research, and operations roles at Salesforce, SurveyMonkey, Forrester Research, and The Ladders. He was also on the team that wrote the business plan for BlackGirlsCode. In 2014, Business Insider listed Dan as one of the top 46 African Americans in Tech.

With Freshsessions, Dan thought he was on to a great idea. Other people thought so, too, but weren’t willing to pay for the service:

It was going pretty well at the very beginning. We built a landing page, and ran some ads, and started to drive targeted traffic to the ads to see if people would be interested in it.

There was a lot of interest. But it was very difficult to find engaged studio owners that wanted to change how they were operating their businesses to adopt a technology-based model, and find musicians that had enough money to book consistent sessions through the platform.

We found the need, but no payers.

If you can’t hear the clip, click here

He wanted Freshsessions to work so badly that he was tone-deaf to signs that it was time to shut the business down:

I always told myself that I would stop pushing forward when there was an overwhelming force from the outside saying that this is not working. Even when, I believe, in hindsight, that I reached that point, I continued to try to brute force it into existence.

We were trying out different cities, and I was flying around the country, and we were trying different campaigns and various things, and ultimately they were not working.

Through that process, I lost a lot including some relationships with individuals, and I started to develop symptoms of acute anxiety.


If you can’t hear the clip, click here

Brian Zuercher is the CEO and co-founder of Seen, a marketing software platform that is helping marketers tell the story of their brand and build relationships with their customers through consumer generated photos and videos. Seen was recognized as the “Innovation Game Changer” at the 2012 Ohio Interactive Awards.  

Actively involved in the local startup community, Brian is a Startup Weekend Columbus host and MC at the monthly morning pitch event for entrepreneurs, WakeUp StartUp.

Additionally, he has an extensive background in building and launching consumer products at Seen, Clearwish (founder), and Woods Industries.

Today, Seen has achieved some success, but it took Brian and his team several pivots to get where they are. They didn’t always keep their angel investors informed about the changes they were making and it nearly cost them:

We followed every test and experimental process from the get-go but we didn’t tell our investors we were doing that. They still thought we were building what we had presented in a PowerPoint slide as the product, but that didn’t work out in our case. 

We did three iterations of the product in less than 12 months, each one progressively going off of different consumer metrics that we found and then partner feedback.

Ultimately, it didn’t work and we decided we had enough time to maybe do one last iteration. We did the tough thing of letting everyone go, reducing the burn down to two of us from almost 10 at one point, and gave ourselves six weeks to turn into a new product.

Meanwhile, the investors thought we were dead. We’d told them, “Hey, we let everyone go. We’ve got some money left, but we don’t know if we’ve got much left.” 

They questioned my decision-making and me as an entrepreneur. Fortunately we had the confidence of a couple board members as well who were able to stand up for us.

If you can’t hear the clip, click here

Seeking help to cope with the anxiety he developed as Freshsessions was falling apart, Dan tried some web-based mental health practitioners. The experience wasn’t positive, and led him to come up with the idea for Level Therapy:

I tried one that was solely web-based, with an interesting subscription model, but they didn’t accept insurance. I walked away from that experience feeling cold, and more like a number than a person.  

Immediately, the entrepreneur in me started to kick in, and I started to think about why, objectively, was I having those thoughts? Where did this company miss? How could I build a solution that would address those points?

If you can’t hear the clip, click here

Dan suggests other founders seek out not just a single mentor but a network of fellow entrepreneurs that can act as a kind of advisory board:

I would aim to find other entrepreneurs that are perhaps six months ahead of where you currently are as a company, or six months to a year or two years ahead of where you are, so they are contemporaries that have experience, potentially the same types of challenges that you’re experiencing. They can give you timely advice.  

Also, individuals that are further out, so perhaps they have either sold companies or they’ve been operating companies for three plus years, five plus years, etc. They can give you more long-term strategic advice.

If you can’t hear the clip, click here

Based on his experience with Freshsessions, he also counsels founders to never work with friends:

It was difficult to overnight go from being someone whose relationship was based around just having fun to actually motivating them and inspiring them and pushing them.  

Maybe that was a function of me being young in my professional career as well, but that was my experience. I wouldn’t do it again.

If you can’t hear the clip, click here

In his first startup, Clearwish, Brian learned it isn’t enough to focus on trying to realize your vision. The founding team must be on the same page about possible future directions for the company, too.

We were approached to get funded and had enough success and promise at the time, but we couldn’t find alignment around the notion of building a lifestyle business first, a venture-based business, and how big it could really be.

 We had never discussed everyone’s expectations when we founded the company. Oh we had a happy, fun conversation over a couple of beers, but we did not sit down and say, ‘Hey wait, where’s everyone want to go with their life?”

Instead, we launched the product, got momentum very quickly, and were swept up in this process. We misfired on that and broke down that team. It was a learning experience for sure.


If you can’t hear the clip, click here

Having worked with startups in Columbus, Ohio, Brian says it’s not necessary to be in a startup ecosystem to make your startup work:

When I was complaining about raising money in Columbus many years ago, Bill Diffenderffer, who ran SkyBus and is now one of the founders of Silvercar, told me, “You keep saying here, but there is no here here, so just go get the money.”

That really stuck with me. The place isn’t necessarily going to make the business go.

If you can’t hear the clip, click here

His advice for other founders? Think big:

I think I try to ask all the questions that I’ve been asking myself, like why are you doing this? Why are you making the decision? Why are you looking at that? Is this what you believe is right for the business versus right for what some financier asked you to do? I push them to think really big.

If you can’t hear the clip, click here

Listen to my full interviews with Dan and Brian by downloading them from SoundCloud here and here (And download any of the past shows here.)

Coming up next on the blog: Emily Kennedy, founder and CEO of Marinus Analytics; and Chris Cabrera, founder of Xactly

Tune in Thursday, Oct. 13, at 1 pm PT, 4 pm ET on Sirius XM Channel 111 to hear these upcoming guests on Entrepreneurs are Everywhere: George Zimmer, founder of Men’s Wearhouse and now founder, chairman and CEO of Generation Tux; and Scott Adams, creator of Dilbert


%d bloggers like this: