Lean Startup Conference Live Blogging
@ericriesThe Lean Startup
Sarah Milstein @sarahM
Lean Startup: not
Eric Ries introduced the conference with a challenge that there be "no vanity metrics" at the conference. He encouraged s all to "Do the work."
"Whether you succeed or fail, if you learn something, tell us, and we'll make yu famous."
He then talked about something that has been on his mind lately.... If you hate big companies so much, why are you trying to create a new one? The answer is that each of us believes our big company will be different.
We are good at buzzwords and slogans -- important to use the right words to describe things, but where are we as a movement? Move beyond initial successes to something much or difficult and complicated.
99% of what's in the book is not stuff that fits on a bumper sticker -- MVP, pivot, get out of the building...
What people don't talk about are innovation accounting and engines of growth.
now that your MVP is in the field, how do we figure out were' making progress after initial success? We are not done after getting initial prodcut fit. This process is difficult. It doesn't fit on bumer sticker, so no one talks about it -- that's what we're going ot talk about today.
The goal of the Lean Startup movement is not just to make random startups more succesfful and more money. We're not against that, but that's not our purpose. Our purpose is to create an ecosystem that makes really successful long-term companies -- to build the next toyota. We want you to master the discipline of entreprenesureip -- "entrepreneurial management" in the book. One day, you'll have to hire at least 1 other person (LOL). How do you institutionalize the management and process practices that will alow those people to become good entrepreneurs? when you build your company, I challenge you to build something new: a company built for continuous innovation.
We're gonna start far afield, in strange places like the US government and education, then wind back home to silicon valley. We'll cover the three big strands where lean practices are ... large companies, governments, and places like silicon valley. Use every minute to tell as many true stories as we can about these ideas
@Todd_Park United States Chief Technology Officer
Eric introduced the first session, with a secret amazing surprise guest! The very first Chief Technology Officer in the United States (2009-2012), Aneesh Chopra (@aneeshchopra), who will introduce our first speaker.
Eric and a bunch of entrepreneurs came to the white house and asked: why is it so difficult to be an immigrant and also founder of startup that will make American jobs and help the American economy? a startup visa is not yet law, but we were able to borrow some of Eric's own ideas and principles and advise a new lean startup of our own.
We created a new service created a year ago in December, which started with a call to action: dozens to hundreds of entrepreneurs. A group of 6 private and 6 government folks designed and built: a checklist that allows anyone to be an immigrant entrepreneur in this country, to have all the tools they need. We had a simple hypothesis: there are existing authorities not being leveraged for immigrants. Our thesis is we can solve our nation's most intractable problems.
Harnessing the power of lean startup mojo in order to better serve the american people.
Background: 24 years old, health, castflight health (wsj #1), in india: healthpoint services (water, drugs, diagnostics to rural poor). All of that is to privde context: the most amazing entreprentuial experience i've had is working with the american government for the last three years. In 2009, first CTO / successor, tech entreprneur in residence. What do they do? run an incubator inside the government, apply power of tech and innovation to do projects (public safety, jobs, education, energy). Each project has a different focus with one explicit common denomiantor: powered by small teams that run efficiently to provide maximum awesomeness for the american people. Get stuff done in an environment of significant uncertainty.
I'm a big fanboy of what the FDA has been doing with its new entrepreneur program - in six months, design and deploy a new "innovation pathyway 2.0" - a new regulartory approval pathway. It turned out to be a wonderful concept for a Lean Startup. rapidly deploy on small scale to test vs reality, characterized by hypercollaboration, shortened time to market, cutting time it takes to go through clinical trials by half (from 1 year to 6 months). 3 companies: one wearable artificial kidney.
5 projects "presidential innovation fellows program" - inspired by code for america (jen paulka out of SF) - over the summer, brings innovators from outiside govt plus inside govt to make
launched at tc disrupt (after a pitched 5 products. 600 applications within 24 hours. selected 18. moved to DC in summer to do unbelievable work. now: pitching 5 products.
1. open data initiatives program.
seeks to liberate data in order to spur entreprenurial innovation. (the "Tim O'Reilly plan", the godfather of it). started it decades ago with weather data. turns out the US Govt collects all weather data -- make it downloadable in machine readable form and . turned it into weather.com, weather channel, improved our lives, made jobs.
In 80s, Reagan (then Clinton) did the same with GPS. "Reagan is the godfather of foursquare." It turns out that giving entrepreneurs access added $100 billion in value to american economy. without legistlation, without regulation -- takes data we've already paid for (bought the military a fabulous gps system) and turned it into awesomeness on a massive scale. The president LOVES this play. GPS is just the tip of the iceberg. Raiders of lost ark - govt is infinite size. that is a good metaphor for the treasures in the us govt that you paid for, and dammit, we have to give back to you.
We take public data that's already out there but unusable to developers (pdfs, etc), then spread the word that the data exists. Data by itself is useless - only useful if you actually apply it. and we need you to actually produce value. if entrepreneurs don't realize data is there, can't add value to it.
We are targeting a number of sectors, but started with health. 2010. small data sets: patient satisfaction, etc. gathered small number of healthcare people in a room and they were skeptical - 9 days later, they showed up at the first datapalooza in madison , 20 apps they built using this data.
Inspired more entrepreneurs to use data AND inspired government to release more data. Spread the word about data through hackathons, meetups and more. Latest medications from FDA, latest publications from national institue of health... got feedbak about how to make data most useful.
2 years later: 1600 entrepreneurs - 200+ companies competed for the right to be one of the 100 comopanies to present at the datapalooza, most of these companies
cloning it across other sectors: energy, education, public safetey, finance, nonprofit, global development.
- catalyze ecosystems of innovation fueld by open data. iTriage has helped 8 million consumers understand systems. oPower - energy efficiency of appliaces (helped 1.4 terawatt hours of energy, 165 million on energy guard). billGuard - help find erronious charges. ondex capital - census to facilitiate underwriting of loans to small businesses. many more.
check out our data.gov site! and its subsidiary sites! take advantage of the free data that you paid for and by goodness, you have the right to get back and turn into awesomenss for the american people!!!!!
rfpez (@projectrfpez), small business administration, enables the govt and tech startups to do business with each other. It is hard to sell to the government - VCs say "do not talk to VA, too much red tape, not worth it" -- why is this a problem? we spend 80 billion dollars a year on IT. If our tech startups can't access that market, big problem. other problem - we want to be able to buy that stuff. So rpez has built, starting in august, an online marketplace and platform that evatoraptes. It's out on github - five american cities have rfpez in "that leads to a double helix of civic awesomeness"
Please consider selling your awesomness to us. try bidding on our projects, and give us feedback about whether we have to remove more suckage from the process.
Right now we have 24,000 websites - our web presence is organied how our gvt is organized - not useful for citizens. mygov is creating alpha platform to organize how gvt interacts with citizens with a focus on the citizens' needs.
4. better than cash
in developing world, making for recipients of our American dollars. We're creating more transparency, more efficiency by paying on mobile phone. This actually had the effect of increasing salaries. What was happening: the money was traveling through sticky fingers - with mobile payments, individuals got what they should have been getting all along.
Another impact: when being paid much lighter salaries, were offered more by aliban. the electronification of money means fewer aliban, more policefganistan (mobile ) salaried payments over mobile for police, teachers, 100,000 usersby ecember
indocesia - large employer [plus literacy programs for 250,000 workers and contractors
5. blue button for america
eter evin veteran to download their own copy of all their own medical records securely. "maybe 25 will be interested if breakout hit?" ctually today 1.5 million veterans have downloaded a copy of their own records. e are getting questions form the private sector saying "are you allowed to give patients the ability to download their own medical information under hipaa?" the answer is yes... in fact it's the right of patients to get their using open technical template that any organization can adopt - to let patients download info from you. 88 million americans have access to their blue button information from at least one source. t's spreading like crazy thanks to this product.
hese are the five running in our incuboator, powered by ean tartup techniques, need your help to to do this. ric, anice rasier, uxor doing workshops with teams. o to http://www.whitehouse.gov/innovationfellows to learn more about the projects, follow on twitter and github (all code opn on github) - help products thsemvesl, but even more ipmortantly: by heling them be succesful, inspire other governmetn
email@example.com if i don't answer in 72 hours, then hit reply and put:
you idiot: subject
hack this to put "you idiot" and save 3 days!
help save lives of american soldiers in afgnantistan
help government save money
these are missions that do not suck. big difference on behalf of the public.
i suspect if we were playing word association, public schools do not come to mind when we say Lean Startup. Public schools are designed to hold the status quo: guaranteed revenue and customers, even when we are losing customers.
Charter schools are public schools with the express intent of innovating.
Charter schools oeprate under extreme uncertainty.
10 years ago, I launched our first school. at the time, following reall well-established facts:
only 1 quarter of them are able to attend or apply to college. #1 reason: they don't complete the required coursework. This can be directly atributed to the quality of teacher.
Our goal was to create a school where every student enrolled completed rigourous set of courses that made them eligible for college. Every student had a high-performing teacher every period, every day.
Mission: create a public high school that gets 100% of students accepted to a four year college (not 24%, our national average).
4 years later: 98% accepted. 8 times more likely to
named by us news, 7 applicants for every seat, raised a lot of money.
4 years after that, less than half of those students actually graduated from college.
nothing bad actually happened: even with only half graduating, still doubling the national average. still among the highest.
However, it still took us 8 years that our product only delivered half the value we expected, unacceptable. also, 8 years too long -- this is these kids lives. if we could teach them a little more a little better, we could ge tthere.
So, we designed a new math program. Kahn academy, technology, after 10 months it's working: they were learning more math than their peers. The NYT wrote an article, we raised more money -- but we were NOT SURE which element led to kids learning more math.
- kahn academy?
- was it that we assessed them 4 times per year?
- simply that they were doing more math?
Then we discovered lean startup! We realized that we didn't hae a discipilined and principled approach to innovation that let us build, measure, learn effectively.
We started with a relatively small batch of students, for 1/3 of school day, for 1 subject.
Hypothesis: if students can learn to drive their own learning (determine own path and pace to goals they set themselves), they will be succseful.
We had just 3 months: pushing ship date back is not an option.
One rule: remove any feature, process or effor that does not contrubte directly to learning.
We started removing walls between classrooms. In order to test that "when kids that direct their own learning, kids learn more," then we have to change schools fundamentally. In old way: teachers direct everything.
To test the hypothesis, we needed to have the kids do everything, teachers are there to provide resources.
We built an online math guide showing everything a student needs to know from kindergarten up, using red, green, and yellow. We created online playlists to go with concepts on guide: Kahn academy, in person, online, on-demnad testing system: show exactly what they learned and get immediate feedback. If they can show they learned a concept, it turned green.
We created a unique space with big tables, open workspaces, looks more like a high tech startup than a typical school. We gave every single student a computer, showed them tools, said go. We commited to empirically figuring out what works.
1. did kids learn math? could they show they learning
2. to what degree did they drive their own learning, make decisions and engage in behaviors that led to learning?
These seem obvious and seem like what every teacher does on a daily basis (measuring what kids are learning)? Sadly, everyone is still using vanity metrics. They are not measuring where each student begins and ends in a short learning cycle. They're not looking hard enough. Is it specifically because of what the teacher did, or because of other reasons (parent, self, other)?
We create in-depth focus groups. We looked for observable behavior data. Teenagers embrace this, she gets emails with the subject "user feedback" from high school students. A cross functional team meets weekly. We have made more progress in 14 weekly cycels than we have in 10 year long cycles. Focused on learning something the kid didn't know when the day started.
They started with 20-30 minute introduction by teacher on the subject in breakout rooms.
We found that students who attended the lectures were no more or less likely to learn.
Teachers predicted: if we knew which students would coming, if students were more prepared and brought questions. This went on for several cycles without aything changing.
but student reported "lectures aren't timely. they're not customized: not exactly what I need to learn. They are not peronsalized -- I don't get help with what I need to learn." Attendance lowered, and when the groups got to be 2-3 students effectiveness increased. They noticed that teachers were throwing out their prepared lecture -- teachers tutored instead.
The tutoring bar was born. like the genius bar at the Apple Store.
We continue to experiment to see what the biggest impact on learning.
We stopped giving lectures. If not for the rigourous path, we'd still be spending an extraordinary amount of teacher time on lectures and wasting our most valuable resource.
This decision cuts directly to the core of what the teacher things their job is and where they'd add value. It would have been impossible to change without clear and indisputiable evidence. This is how we got them to move to different behaviors quickly and without resistance.
Twitter and facebook or visit, we're here in the bay area.
Ten Ways to Get Out of the Building
Last night's ignite session was very engaging. the diversity of backgrounds and sectors in which lean principles are being applied.
Tendai Charasika (@XcellentTea) EnterpriseCorp is from rural Kentucky. He works for a chamber of commerce state organization: the Kentucky Innovation Netwrk. He's a business coach in community, working with all ages and states of compnay to pull and push their companies forward. He talked about 10 ways to "get out of the building" with examples. We need to "get uncomfortable." He wants all of us to tweet our own ideas with the hashtag #getoutofthebuilding and add to the list.
e are the front line of entrepreneuship of rural kentucky. in the trenches, working with 100s of companies. what i've found is: lean principles help us manage entreprenurs to hit milestones in their development.
n working with so many companies, get asked: what's the one commonality that you observe that these early stage companies go through? tihout a doubt, what gets these early stage entrep: lack of passion and lack of desire to get out of the building. talk to to get early feedback about validating their assumptions one on one. aybe you're getting unconformatble as well about taking that idea you have and sharing it with people outside your friends and family.
few things: first one is a fear of rejection. no one wants to be told your baby, which you work on so much and feel so passionately about, is ugly.
n a tactical level: some entreprenerus literally don't know how to get out of the building.
hat we risk by not getting out of the building: the opportunity cost of pursuing your next best idea. need to know up front if no customers want it. e also waste a lot of time -- our time and other p time. ut importantly: opportunity to learn what customers want and will pay for. miss out on oportunity to iterate on products and make them better, someting worh pursuing.
10 pragmatic ways for folks to get out of building.
everage a combination of these. have found that the various ways to do this are as unique as the entrepres in this room, so please add to this list! be creative!
1. don't ask your uncle
this is about what not to do. when you sit at a table of entreprenurs and ask "have you talked to anyone?" they say "sure!" and you say "who?" and they say "my uncle"
2. set up a booth, do a public demo
3. interview potential customers
2 and 3 are scary. if you have a demo and you observe their reaction to your idea and
4. put your office where your customers are
5. throw a party
4 and 5 are about being creative, and you as a unique opportunity. 5 is to receive testers for their first mvp.
6. talk to experts in the field
leverage what others have found for you, if someone else has done the legwork already.
7. find the decision maker
that'll be the one that buys the product and selling them on the values and benefits of what your'e building is critical.
8. listen to wht customers are demanding
9. pre-order, landing pages, analytics
8 and 9 show demand for product. if you have the ability to follow up with them (email or contact information), there's no subsitution forthat.
10. ask for the introduction
to get in front of first customer, or getting a mentor, at the end of the day, it's all about building relationships.
Now, some examples.
uilt a version of an app that teaches basics of programming to kids 5-7 year old, fun and engaging.came with idea, we helped connect to in the community.
nterviewed 25 people face to face - parents and children - brought them in like office hours and observed children playing with app, and interviewed parents. found 2 critical things
1) children in younger demographic didn't have fine motor skills to play the game. so they changed
2) after talking to parents, saw that what was most important was seeing their kids engaged in learning.
ohn and retchen wouldn't have been able to do this without getting out of the building.
help them by downloading app and help them figure out if app sucks and stop wasting time, or actually improve product.
ooibee ed ea
reat eamples of being where the customers are putting product in the mouth of customers who will help them get it out on shelves. usic festivals, store openings, farmers market - to get feedback from experts in field; people who love teafrom 2010 to 2012, now have national distribution.
aaron marshall and the app ver - picturs on text you love the most, simple. experimented his way to over. aron doesn't put his office in his basement or apartment doesn't sit in office building sits in coffee shop in cool part of town, called "quills" -- and talks to customers 5k downloads in first 3 days
ric ies goal of a startup right to build customer want and will pay for as quickly as possible.
What Is Innovation Accounting?
TEREZA NEMESSANYI@terezanHigh Ridge Group / Honestly Now
This Lean Startup Conference is, at its core, a Lean Startup revival meeting. We've heard these amazing storis of people taking flight through these principles. To have a kickass revival meeting, you have to start in the lows.
Accounting is the foundation of how we hold people accountable. These methods are not just keeping track of where the money went - but which managers are doing a good job. Make forecasts from the long and stable history that the company has.
Making accurate forecasts. Didn't even know this was possible for startups.
We need to create a new accounting system that helps us report progress to people who hold us accountable that are not based in vanity metrics, and this is hard.
Ground rules: these are things I would do differently next time.
Hard to talk honestly about these issues. Where'd they get the vanity graphics? From the pitch deck that WE made when we were trying to get money. Ridiculous hockey stick graphs. We say that we'll have X revenue in year Y. When VCs come back and say "your'e failing" because we're not meeting those numbers, that's our own fault...
Start with the minimum viable product phase, establishing the baseline.
MVP: that first prototype out of the gate so you can have something to interact with and have a baseline for metrics.
I talked to my husband, put some money down, worked with some people to put something out that was sort of ugly but also kind of cool... serving multiple masters, difficult to focus our development energy on who we're tryign to please. creative challenge from the outset.
Start from point of time when we took more substantial funding.
I would really like it if you could rate my mistakes in terms of cringe points (1 to 10 where 10 is "eric is crying")
Teresa: I got a term sheet. we were negotiating. i didn't have alternative term sheets to negotiate -- either taking this, or game over. as the CEO, I own this decision. However, this metric was thrown in at last minute about traffic, not related to core model (which was subscription based). So, I accepted a vanity metric in my term sheet.
Eric: this is common. I want to takthe VC's perspective: they want something to hld you accounable for. if all they know is eyeballs, then that's what they'll use.
I once had a term sheet that had requirements about how many press releases we'd put out, so I've made even worse mistakes.
Teresa: Phase two of innovation accounting: instrumeting the machine. I was managing two companies. we had a setup where we had montly board meetings (a lot) - busy pushing business forward (team believed it was supposed to be) - wound up spending a few weeks on what i thought the business was, then a week prepping for board meetings reporting on thse other metrics - and then hard to weave the two together. i have 20 years of experience in innovation, very good strategitcally, but even so it was close to impossible.
Eric: one of speakers last night likened getting stuck in vanity metrics to meditation. We need such absolute focus, but most of them we're distracted. There's a board meeting coming up: starting night before, "better run some metrics, but we've done 4 weeks of great work so the metrics will be great" ... metrics are bad, then we thought "OMG what if we ran the report a week ahead of time so we can do the fire drill not the night before?" then we realized: well if we're going to be driven by these metrics, maybe we should look at them all the time...
if we looked at them once a week, then we notice they haven't changed even though we've done a week of work. The existential angst of what we're doing as entrepreneurs is that most of the things we're doing have no impact on customers whatsoever. and no one wants ot know this.
Teresa: Phase three: pivot or perservere.
When you go back to people that you sold a vision to and say "we need to change," that's very hard. if the data sayswe're doing the wrong thing, i can be mercenary of throwing that out the window -- but your'e the steward of the capital that came in and you have to social that situation.
Eric: investors have an option other than pivot and perservere, which is focus on one of the other companies that's doing better. It's really pivot, perserve, or portfolio. Our investors have the option to focus on one of their other portfolio companies.
How do we show progress in meaningful ways even when the vainty metrics aren't awesome? during the long flat part of the hocky stick, in successful companies, somethin is happening. In order to demonstrate that to the board, we have to do math. if we master these techniqutes, and demonstrate that the work we're doing is actually making a difference long before it shows up in the vaniy metrics (like the summit school exaple)
we are fortunate in what we do: go out every day and try to create a future. even though we fail and fail a lot, this is honorable work. we should share the mistakes we made with each other. if i'm not the one who succeeds, someone will stand on my shoulders and make
for the good of the ecosystem, and the good of the world, fail. fail fabulously, and share the good news.
A Conversation with Beth Comstock, SVP at GE, about Bringing Lean Startup to Life at One of the World's Biggest Companies
ERIC RIES@ericriesThe Lean Startup
Intro: GE is #6 by revenue. As far as we know, this is the biggest company using lean startup practices. 300,000 customers. Started by Thomas Edison. Its been through a few iterations.
Eric: What is a chief marketing officer?
Beth: Connecting the great science and things that we do to a customer need, to a market. it starts with innovation, it starts with a customer.
Eric: There are startups inside GE. Waht does tha tmean?
Beth: There are entreprentus everywhere. they exist in the company -- inside and outside. Internal and external opportunities. Within ourselves: be humble, we don't have all the answers, partner externally without "it's our way" -- it's a shared vision.
Eric: tell us about anexample of successful startup in GE
Beth: a decade ago, our chaiman: guaranteed pipeline of incubated
has to start at CEO level. We need to fund and protect them. If it's a bad idea we'll kill it, but won't kill it for crazy day to day. We have a "protected class of ideas" called "imagination breakthroughs" There a lot of exciting examples. One is an energy storage (battery) business. A hybrid locomotive has a sodium based technology. The team said, great for trains, how else can we inonvate? Power cell phone towers in africa. from year 1 to year 5, 100 million dollar business in five years.Prescott Logan, who runs this business, had to be different and try things different than running normal course of traditional P&L Buseinses.
Eric: when i meet the actual entrepreneurs that make these things happen, we speak the same langauge. and yet, have ability to live in corporate environment. causing problems cuz constantly surrounded by chaos. how help protect
beth: keep them away from the mainstream. integrate selectively. they have to integrate to get right funding, technology and poeple, but also have to leave them alone.
self-selectiong. there are people who want to go this path and others who dont. filtering the poeple who are risk takers and have the ability to deal with ambiguity. you don't now if your tech works, or if you have your first customer -- need to find these poeple and protect them and give them space
Eric: why would someone with these skills wnat to work with ge?
Beth: access, technology. cutting edge. this is good stuff.
other thing: technology that matters. telecom in africa that is not only powering an ecosystem and an economy. there's a lot of good in that, as well as building something from scratch.
Beth: Hardware people say - "okay software guy, it's easy for you to iterate, but i'm building a jet engine. what can you do for me?" concepts like MVP help tether and translate. Engineers want quality and perfection and that's gerat - they need to push the limits of science - but it also means they have to be perfect and appeal to every use case, so before you know it, you have something that's overengineered and have something that's hard to get out. So what resonates in any industry setting is something you can get out there and test.
Eric: You have a 5 year product development cycle -- this stuff takes a long time to build. But how much of the 5 years is mitigating the risk of the engine itself, vs waterfall? What would it take to get a low-quality thing to market? Engineers get excited.
Beth: We get ideas like "wow, i could put an engine in the market and have a service person standing by 24 hours a day." It stimulates engineers' imaginations that we're embracing Lean Startup in company: what would an entrepreneur do? you can fall in love with tehcnolgoy, and how user uses technolgoy; we're using lean startup methods to marry those two.
Eric: At your level, in a company like GE, if it's a good idea, someone will say "Let's put a hundred million dollars into that ...Excuse me? If everything that eventaully became big has to start small, but you need to make big investments, how do you encourage your team to celebrate small wins?
Beth: one toll gate at a time, one small step at a time. small scucesses, microshare. It works for this customer, now how do we scale? it can't work for every customer out of the bag.
You break it down -- you can be intimidated, a hundred million dollar commitment over 5 years, but you have to break it down... get to proof of concept, first customer: you have to be very patient.
Eric: It requires executive leadeship from the top down. Jeff Immelt and yourself: how do you create this executive leadrship?
Beth: top down - can never be too connected to the market, so time to market is critical.
Without a customer, there is no business. It starts and stops there. So you have to have a leader that breathes customer. Then you have ot have the right fudning, the right tools, and teamwork. It's a marriage of technology and marketing. Mark Little, our CTO (and heads up glboal research) are I teamed in this effort. We also brought in a Chief Learning Officer so cultural pieces are there as well. It's very integrated and seamless.
Symbolism: "it may not make sense to you but we're going for it, we're funding it" sometimes you have to do that.
Eric: I saw a giant jet engine (as a prop) -- why software?
Beth: we're investing very heavily in software for a couple reasons.
1) better analytics for our jet engine puts off a terabyte of data every day. single engine. how do you take intelligent machine and ocnnect to internet to help airlines and hospitals?
jeff came here: we need the kind of consumer social media analytics skills that are here to go into industry.
150 million dollar service backlog. help us build this.
partnered more in open challenges. data quests.
2) flight - alaska airlines - take most comprehensive data that exists publicaly and compile . pilot sees weather different from air traffic control. people don't know this.
same with hospitals - how do you encourage developers and designers to
connect our access to big data with epople in this community who have access to build algorithms and write prgorams.
how and why?
gequest.com. 600 thousand dollars of prizes. special pool of prizes just fo rlean startup people (just this crowd can vie for). take all the frameworks you know from another industry and apply it on the industrial side.
if a train could move one mile faster, a railroad could save $200 million. this i sreal money, all built on sensors and analytics.
people want to have the permission, tools, and freedom to do what they want to do, so i'ts been great to connect with you!
Lean Startup Case Study: Getaround
JESSICA SCORPIO Getaround
Jessica Scorpio (@jessicascorpio) one of founders of Getaround spoke about her experience "Prototyping to Validate a Big Idea".
The big idea of getaround: solving a problem called "over car population" cars that sit idle connect to people who need a car.
5 billion daily un-utilized cour hours. Cars are really expensive to own. it costs 19% of their household income to own and operate one car. People who don't have a car would love to get access to these unused cars.
We heard "great idea, never gonna work." That's the thing about big ideas: lots of road blocks. so you prototype (legaslative, tech roadblocks)
1. used a university class to test it. moffit field. 150 students/faculty/staff to test. started very low-tech. getting owners to hand off keys to strangers -- was enough to prove taht they'd share with strangers.
2. yahoo iphoneDevCamp.
unlock quick win, found our co-founder through that, won "Best Money Making" app.
3. needed a world stage to show we can get traction. needed to show that thousands or tens of thousands could get this. in 2011
16,000 drivers signed up (some percent of zipcar over 10 years)
10,000 cars (more than zipcar in 10 years)
partnered with hathaway? for insurance
worked on legislation. this is a new industry. got new laws passed in ca, oregon, washington. government supporting this idea.
a few years ago, one car one driver. now, one private car shared amongts different peopel. (even a lotus elise!)
some trouble with slides - minimally viable AV problems
Testing the Market for a New Kind of Car
Danny Kim (@LitMotors) founder of LitMotors.
Actual prototype in audience! It looks cool.
We take the romance and efficiency of a motorcycle and put that on a car.
Building a car is hard. I'm the only founder with an actual track record of building vehicles (video!) If you want to start a car company, you have to be a car builder. (If you want to start a web company, you better be a web developer.)
This project led me to ask, how to cut the car in half? Most people drive alone. We're building the ford model T of the 21st century. Creating a sustainable mode of transportation that's affordable (like model T for masses).
Lots of risks. but if you do it right, huge returns for 90-100 years.
What we've done for $800k: major risk reduction.
Remember Segway? What WE did is - not focus on engineering first. 80% of our financing showed that there was a market. reduced our market liability. proved that there's a business case to devleop technology.
We built our own showroom in our lab. We started to de-risk the market. We brought in 54 people: baby boomer motorcyclists and younger professionals.
15.7% of everyone who came into this simulated dealership experience put money down on the spot. That's 30 times what we actually need to have a viable company. Most investors will take the number you feed them and divide by 10. Every time I spend a buck, I want 10 extra return. We killed six birds with one stone. (engineer, design, media...)
We started taking preorders. First prototype: built it by hand. Saved 2.8 million dollars because we built a unibody chassis by hand. It takes six iterations to get to a beta.
After 3 months, we had a driving full scale prototype. This led to killing another six birds with one stone: we had an accurate bill of materials, and weight (needed for manufacturing). Also wecreated huge amount of media attention and increased preorders. We are half sold out of first production run, we proved to investors that we can build it.
Two wheeler that doesn't fall over (even when pulled by another vehicle)
Iku who rides a motorcycle says: "it can back up!!!!!" (Sarah, drives a car, didn't even notice!) They totally nailed target audience with their video :)
"We derisked all of our hardware. we have a prototype. now we just have to work on the code. We all know how expensive that can be."
David Kim, @litmotors
I (Heart) Ugly
LANE HALLEY @thinknow Carbon Five
Sketching is the most valuable tool.
Everyone can sketch. Combine circles, triangles, and squares to make visual interfaces.
Team sketching is a form of making, which is better than talking. As you sketch as a group, you become more failiar with techniqutes. you put on a wall: you can't tell what was drawn by engineer, designer, or product owner. everyone draws.
cross pairing is important. designers and developers working together is important. increase code and design literacy
iterative design -
quick visual collaborative continuous
Innovation Games - a project where I elarned to get over myself and embrace the ugly.
We had a very audacious objective: new game platform that woudl be web, mobile, lots of ideas. Start by making a consumer pass and play game on iPad into app store in 1 month. audatiocus cuz noen of us had made an ipad app before. our first screen designs didn't work because we didn't have a shared language.
instead, used workflow sketch and externalize what we were thinking.
vizio wireframes - designer bottleneck from great ideas - what's the lightest weight, quickest thing to do? make a paper prototype. made the game in a layer of paper with sticky notes moving around. played the game by passing it around the table. our faces lit up and we were having fun and we realized we understood the game. we could break apart and front end, back end could work independently.
lighterweight, clickable balsamiq prototype.
we got it into the store in a month. you can learn a lot more by selling a small game to the store and get it out there in the real world than months and months of thinking.
consumer games, and exploring
Tips for Rapid Mobile App Development
Matt Breznia (@brezina) an early Lean Startup adopter, as a founder of XOBNI. Sincerely helps the world be more thoughtful. It allows people to send physical gifts throught the mail.
Postagram (picture from phone as postcard). something card (holidy cards). setsme (themed gift sets from phone)
We started from simplest gift you can send in mail: a postcard.
In the first year: we created a five apps.
With a team of six people, in under one year. We had also launched 4 other apps using libraries and such. 9 products in 12 months on mobile.
we wanted to learn a bunch of things. printing, physical delivery, mobile commerce, etc.
santa postage - "postcards from santa" - parent send postcard from santa to kids in mail. choose one of few pictures, get credit card, send in mail.
solved billing, printing
Then added one thing and made 99 cent: solved library photos + image upload
THEN made Sincerely
2. Off-brand testing
(or test in canada) -- off branc
best thing: developers don't have name on it. don't care what it looks like. don't care if breaks. all we care is learn: does this product work.
Sincerely was on brand.
3. De-emphasize visual design
- don't de-emphasize interaction design
- engineers can actually do design; dont' care what button LOOKS liek
4. Reuse common components
they're different types of products, all use same address book.
5. Buy cheap, disposable users
- no cross promotion
0 no pr
- buy users from sumbleupon, admob, iads, tapjoy, google
disposable: teach us what we need to learn, kill product that they're on.
6. Be willing to kill apps
- we're killing an app this month (dotty... but it's not important to our business)
- you will have upset customers
- must do this to support what is strategically important to the company
7. Use android for quick testing
- no app approval
- go live in hours
- apple updates go live in 7 days
- closet sthing we have to web in the mobile app world
"google we love you, thanks for being out there"
8. APIs are everythig
Payment processing, fulfillment, user accounts, address books, emails -- all internal api.
no sincerely product engineers touch our database.
Then made it external.
Now we have over 60 mobile apps that use our external api. Including Facebook.
9. Team structure
Pair up developer and designer to build an app.
If your'e buildin an offbrand app, dont' even need a designer -- just have a dev do it.
10. Minimize interdependencies
One of worst things you can see is "i couldn't get this done cuz design mock wasn't ready" or "i didn't know how ot send emails" -- instead, goal as CEO is: each develoepr have 4-6 hours of uninterrupted magic each day.
On the Way to Lean Startup: Curvy and Working it Out
Ron J. Williams (@ronjdub) founder of Knodes
Some of us aren't quite lean et,
Non-technicall co-founder in a technical organization
"To thine own self be true" -- Shake
We need to be honest with ourselves. Build - Measure - Learn for everything.
When I talk to a customer, I ask: what am I going to learn from them?
Telling your team to BE lean is like a crash diet. It breaks your body. Don't do this to your team "Hey guys, happy Monday! We're going to be a Lean Startup now. We're going to start doing 15 things differently" I hadn't gotten buy-in. I hadn't created a sustainable plan.
Changing bad habbits is really challenging. Working out is easy if it is convenient. Know who is on my team, what are their strengths? This is not rocket science, but you are not impervious to bad decisions just because you are drinking the Lean Startup koolaid.
Meetings were running long. I noticed that I was throwing things at him in the meeting. Quick fix: prepared an email in advance.
Sometimes there is no quick fix. You need to get help! Like when you are getting into exercising, a personal trainer will look over your shoulder and help you create good practices. What are your goals? Make resaonable ones.
BEING Lean is not our goal. Our goal is to have fun creating a product our customers love.
"just like science, which also takes human judgement" -- Eric Ries, 2011
Failure is Great and Other Myths About Adopting Lean Startup
Adres Glusman (@glusman) Meetup
Share 3 stories on the transformation to customer development
Myth #1: Poeple give a damn about Lean
The four steps to the epiphany (successful stragetis for products that win) - Steve Blank
Made a long hypotheses, test problem hypotehsis, test product hypothesis, verify. 18 steps! go through this iterative loop! isn't this awesome?! OMG crazy.
I was dejected, drank a few beers that night -- turns out i was selling her a process. no one wants to buy a process (like a root canal). dont' need people to buy into a process - just do it.
pual miller showed up in our office and i give you early access to something in beta while
1. the product we were thinkign about launching was aboslute crap (killed it few weeks later)
2. something extermely powerful in this approach we're taking. we brought paul into the building. weekly, then twice a week... now we do 4 sessions every day. 600 usability sessions this year. will cost 30k, less than 1 outsourced usability project. this has been open sourced, we can look up how they do this.
Malkovich bias: the tendency to believe that everyone uses technology like you do. if your'e an iphone user you don't understand that android exists, etc.
shifts the conversation from
reality 1: People give a damn about their work (not about lean)
seeing where their asuptions were so flawed - becomes part of their workflow.
Myth #2: people was to test things
Reality: People like buliding things. People want to build things.
Large risks. Notion of testing felt like we were tweaking the edges. "i'm not really psyched about spending my time doing split tests" but perservered and did it anyways. first tested easy things: sales funnel <-- one got a 10% lift (one out of 10 tests)
subject lines, buttons in email, onboarding flow, algorithm changes... each win had a dozen failures behind it but we got some momentum behind it. then the more you test the more people got into it. walked by peoples' desks and staretd to hear, let's test it!
also shifted, fundamentally, our way o fthinking about failure. what would you attempt to do if you knew you could not fail? instead, think: what would you do if you knew you were going to fail? you need to embrace failure.
you can have an attitude of moving gingerly towards it slowly and back out and look away to avoid failure, BUT reality is: when i try to delay the failure i'm nervous about, the more i find that i just fail anyways and i'm miserable because i'm stressed out waiting for the bad thing to happen.
So what we've learned: embrace the failure. go after the things that will cause us to fail as fast and as often as we can. opourtity to get the trials. in.
Not just that people want to build things: they want to build things that peole use.
Reality: People want to build things that people use
Of course you want to test and make small mistakes and make bets on big things that aren't that risky. (wat?)
Myth #3: Can test your way to a great experience.
"let's test it!" the problem with this is testing your way to an experience leds your way to the homermobile. everything that works and mash it together. when you design purely through iterating through tests, you end up wtih something not coherent.
prioritize this now: design + lean. how do we do design + lean?
Sum up all three myths in one big bang.
The way to create a new meetup group.
- user groups
went back to drawing board and tid dthis many times and iterated. julia child thing. series of steps, fill it out, click next: final step of
rearrange the entire experience, more dynamic, people get excited about it when we tested it.
in old days, would have locked selves in room for 3 months, rebuilt whole thing, then launched it.
1st core idea: people will get excited about interaction. first thing we tested: flow becomes more systematic and stepped. 2, theme. then merged 3rd thing in. meetup.com/create to see where we're at in the process (as opposed to where we want to go)
vision guides experiments. after experiments, come back and validate visino.
is the thing we're excited about in the vision manifesting itself in the experiemnts? if not, we need to invest in something that is meaningful.
How Engineers Embrace Lean Startup
director of engineering
How to move beyond engineers resisting culture change.
Central to the idea of all of these startups is that they're technology. Need the engineers building the technology. How to get these engineers on board & engage them to be full participants in the process
1. Role of prototyping and thinking about the MVP
Melissa: hybrid of UX and product (by trade, designer). validate as many assumptions ahead of time using paper prototypes and clickthrughs. anything we can put up on wall, anything we can put in front of customers - smallest thing we can do befoer we build code. i have been wrong more time than i can count - i was sure my customers . our mantra: our ceo says this all the time: "my opinion doesn't matter." it's hard to do and say as a designer, but it's working for us.
sam: when i arrived at change.org a year ago, we had a culture of testing, but took a long time to run our experiments. what we found: our engineers, who are all well versed in agile and tdd, were still building experiments with very heavy process for production code. we had to try things to run a little quicker, try things with smaller sets of users, . real struggle for engineers to adapt to. took a long time to relearn that. approach: divide into two separate types of work.
1. unvalidated hypotehseis where goal is to build something as quickly as possible. validated learning - to prove or disprove hypothesis as quickly as possible
2. things that HAVE been validated that we'll build more robustly.
as long as we were clear at the outset which type of work it was, gave engineers a goal to head towards and cleared up confusion.
evan: other thing you hear a lot: what metrics, data are you collecting? need to have, from the start, integrating mixpanel or kissmetrics from day 1. do you have experience with getting the developers to colelct the metrics?
for us, matter of doing it quickly and cheaply. struggling with "build your own" syndrome. several attempts with building our own systems. took us a while to realize that this was a big waste of time and money and there are wonderful tools like mixpanel out there.
We Went to West Africa and Learned Our Key Assumptions Were Wrong
Provide toilets in peoples homes
Put together 6 prottypes of products
initial assumption is that people wanted "water flush"
People prefered a really simple toilet model
customer were willing to pay for the toilet service
Would you prefer to dump your own watse?
or would you be willing to pay for a service where someone would come, clean and remove the waste
After a few days they went back and asked them if they would still eb willing to transport their waste. Everyone said NO they didn't want to transport there waste and that they would be willing to pay.
11 out of ___ said they would like to put the clean team sticker on their toilet
they had 4 stickers they asked people if they wanted to put on their toilets
150 families with toilets to 10,000 families
Moving Fast While Caring About Design at Hipmunk
published a book at 16 with O'Reilly
They encourage people to use the live chat feature of their website
Everytime they launched their first product they had a button on the bottom of the site that said click her to talk with a human. The most popular questions was "Are you really a human?"
See which flights had wifi - put a logo on the site next to the results
They thought most people would ignore this feature
What actually happened was that users loved the feature
The decided to make this the number one feature on mobile and it drove 10s of thousands of new downloads
rolled out a new version of their hotel prodcut recently
Where they showed people a subset of thweir hotel data and they had a plethera of users come on to their live chat and tell them how much they hate the new feature then they rolled it back.
The way they collect feedback from users: the live chat feature on their website
Testing MVPs with Crowdfunding
JUSTIN WILCOX @Justin_Wilcox Customer Development Labs
your startup isn't a business it's a hobby
both easy to get obsessed over and they both cost a lot of money
a business will make more money then you will give up
first thing we do is to come up with an idea
does this solve a real problem?
our own problem - we've run out of tools
how much will they pay???
how much would you pay for this?
critically important to our success
kickstarter - our savior
pebble went from hobby to business without a product
Bounce: be on time
kickstarter - will they pay at a specific price point
what we need is: crowd testing
self starter an opensource corwd funding platform - where we integrate a/b testing in their campaign
$5 x 1.4% = .07/visitor
$10 x 1.7% = .17/visitor
crowd testing a business or hobby before you build it
Making Decisions By Ignoring Sales Metrics
ALEJANDRO VELEZ@BTTRVenturesBack to the Roots
NIKHIL ARORA@bttrventuresBack to the Roots
Two full tim ebankers became muschroom farmers
two and a half years ago went to UC Berkely
a professor said you can grow mushrooms on coffee grounds
@ farmers market let's use a timer - test how long someone would stay
A lesson on engagement
Our vision - making food personal again
Went from a product where they were making fresh mushrooms to an educational kit on growing mushrooms
Lean Content: Testing Marketing Copy (Instead of Spinning Your Wheels)
STEPHANIE HAY @steph_hay FastCustomer
using words to aquire users
people choose what they understand
when you're driving through iowa and you see a sign that says we have the best buger
but we know not everyone can have the best burger
lean content speaks to the first time user (not to you)
describing your product in a way that a user understands
marketing goal: be understood
3 things you can do to get your content understood
a) watch for a-ha bodty lang
b) write down their questions
2.) mom test - if you feel like a tool saying it to your mom you probably are a tool
a) why did you sign up?
b) why did you come back?
once you nail that messaging now you need ot get found
test yuour messaging in adwords
embrace the unsexy words that people use in organic searches
look at entry points and top content in google analyts
drop the crap be proud of whats real
learning and using the real person words that make it easier for people to find you
10,000 Startups - Teaching the Entrepreneurial APIAsh
It's about practice, it's about doing
no insites on the business model canvas
We used to belive that entrepenuership can't be taught
It wasn't can it be taught or not but the real question is who can it be taught to
ENtrepeneurship can be taught to those who colunteer
What we used to believe = years in school and tons of money
what we used to believe - All i need is the 5 year forcast
What we know - VC's and the soviet union are the only ones that require 5 year plans
we now know how to make startups fail less
1) we need to know what a business model is
Business model canvas
2) Customer development
3) Some type of iterative and incremental building process
Alexenader --- came on stage
What if we had a roadmap to collect data, get insights, and improve our business model?
What are the jobs customers are trying to get done?
We hire products to do things for us - CLayton Christensen, Harvard Professor
"People don't to buy a quarter-inch drill, they want a quarter-inch hole."
"The focus should be on what jobs customers are trying to get done"
What are the biggest pains?
What are the biggest gains? What should customers expect the outcomes of these jobs should be?
You map it out and then you go and test
We are going to map out three things to get to the product market fit
How do your products and services kill pains?
Fail/change until we get it
Steve is now back on stage
How do you create 10,000 startups?
Why? It's a nice number
we believe this can be taught and be practiced every where in the world
first step was to put the class online: udacity.com EP245
3-weeks hands on course
Running a Lean Startup Sales Process
GEORGE BILBREY@gbilbreyReturn Path, Inc.
Sales in a lean startup
new solutions new positioning
So what's the problem?
Initially it didn't seem like we had a lot of problems
built a high fidelity prototype
Horizon 1 0-12 months
current businesses - generate todays cash flow
Horizon 2 12 - 36 months
today's revenue growth + tomorrows cash flow
horizon 3 36-72 months
Horizion three selling is different
early evangelists are different
CEOs and GMs and PMs have superpowers
Horizon two problems are different
Focus on the key items your going to need to teach
Where are the leads going to come from?
How will you qualify leads
What are the key questions you will ask?
What's your story?
Get as much of that sales learning in horizon three as possible
get a sales person into the process as early as possible
The sales person must like experimentation
They are willing to do voice mail message one to one group of people and voice mail message two to another group of people
Compensation is different
Prepare for pivoting in Horizon two
start small and organize for experimentation
Not every suspect is a prospect
training will require pivots and iterations
Bonfire of the Vanity Metrics: Numbers You're Still Using and Shouldn't
IVORY MADISON@ivorymadisonRed Room
founder and ceo
"There is no saftey in numbers or in anthing else." - James Thurber
Vain, vainglorious Vanity: Not good for Business
- Vanity blinds you to a lock of product/market fit.
- It commits you to your original vision without agility.
- It gets you invested in apperances, not reality.
"Like dreams statistics are a form of wish fulfillment"
Vanity metrics: The most common form os success theater
- living beyond your means: unsustainable
- spending energy merely looking successful: stupid
- exaggerating your accomplishments: dishonest
Mark twain - "facts are stubborn statistics more pliable."
Notorious vanity metrics: what numbers not to use
"This is the way the world ends, not with a bang but a whimper" - TS Eliot
Characteristics of actionable metrics
- measure success at your core business
- show or directly relates to revenue
- track real, indicidual customers
- Illustrate cause and effect.
- Lead you to what to do next
I'm not prepared for a zombie apocalypse. I need more bottled water, a shotgun and stronger abs. - jenna fischer
The most important metrics
- measure customer retention
Find the big picture in the small numbers
real lean startups don't flinch at real numbers
Innovation Accounting in Practice
ASH MAURYA@ashmaurya Spark59
We want to be like Tony Stark, but instead it's like "WTF?"
2. Measure everything as a cohort. Not enough to know what ot measure, but also how.
You cannot step in the same river twice. More interesting events like revenue tend to be longer life-cycle events. Cohorts let you group your customers and keep your events together to more accurately derive cause and effect.
3. focus on a single key metric.
This is what our dashboard looked like earlier this year. First, takle the low-hanging fruit: the activation rate. First, write out activation funnel.
4. Build a continuous feedback loop with customers for rapid hypothesis generation.
Go to the people behind the numbers and simply ask them. Modfy dashboard to show u not only users that were activating, but also those getitn stuck. now, identifying reaching out over email every day was tedious, so
generated 20-30 email resopnses every day, provedL we were able to identify
"1. too busy, 2. need more info, 3. just kicking tires" -- decided to tackle second item by building a 7 day lean canvas course.
we thought this owuld have a major dent, but meansured no significant impact. made no sense whatsoever. normally based on this data, right hitng would have been to kill off this experement.... except we built a way to leave comments and we got a LOT of comments on this video and they were overwhelmingly positive
So our modified experiments dashboard looks like this and it tells a bigger story.
We invested in extending it futher, turned it into a longer, multiweek sequence; measured steady bumps in retention and revenue.
5. metrics alone are not enough.
only part of soultion to tell you waht's wrong -- but if you want to know really why, need to go to people behind them.
LEAH BUSQUE@labusque TaskRabbit
Not just changing the way we work, but: one of the companies with the most disruptive potential in the decades to come!
Early on, very little money, very few marketing dollars. Lean marketing: if you had $1 to spend, where should you put it?
Early on, when starting a company, lots of shiny things you want to go after -- but it could be easy to spend money on lots of channels without realizing what sort of impact you could have on your business. Learned that trying to do everything at once leaves oyu in a place where you've done everything poorly and your'e not sure what works or what doesn't.
Back in Bostom, tried many things: street team events, newsletter, SEO campaigns, daily deals sites (groupon)... but doing all of them at the same time giving mixed results about what was working. Wasn't until I started reading Lead Startup and lsinging to Eric's talks --> applying methodology to marketing campaign.
Key thingL understand where you want to hone it. build measure learn can be applied to marketing campains just as to product infrstrutrues.
Customer development. Do you know where these people eat, breathe, live? Go down list of customer acquisition techniques; run experiements; measure queickly.
Example: a few years ago, twitter was coming out with promoted tweets. Hey Taskrabbit, let's have you experiment with promoted tweets and experiment with geotargeted tweets. Developed some campaigns, started to measure: measuring the really key metrics for the business (not just #followers or -- instead, actually tracking how many people went to the site, put in their credit card info. Be disciplined about what's moving the needle for your business,and apply those metrics to your campaigns.
Learning // scaling the channels taht are working, removing the ones that don't.
3-4 years ago, when doing street team events on ground -- weren't seeing offline to online conversion that we wanted to. 2 years alter we decided to revisit the effort (evne though we thought we crossed it off the list) -- easy to say "it's a few years later! messaging is different!" but if you feel really good about how you conducted and measured this experiment, you should be able to put these things to bed and move on.
Test everything, not just channels - not just about channels but holistic approach to marketing/messaging.
brand, design, geo, messaging, value prop, AND channel testing.
What we realieD: peopel wanted a serene experience. feel a sense of release that we weren't offering them RAn a/B test on two sites plus qualatiative. HEard from users that serenity was what they wanted. saw that in our conversion numbers as well (5% for corkboard vs 10% for serene blue clouds)
Testing PR on local scale vs national scale. Went to local radio (wpix) vs today show.
We thought today show would
But actually measuring people posting tasks, putting int heir credit card and actually converting: national were better for conversion, not just for purely unique visitors and email signup standpoint. Showed that our band was more trusted (and trust barriers for taskrabbit is important).
(bar for people we do know is very high)
How To Run a 5 Whys (With Humans, Not Robots)
DAN MILSTEIN@danmil Wingu
director of product development. wingu in boston.
Post mortem where we actually talk about
Problem is: w'ere humans, not robots. shame is a very intense emotion. will lead people to have the instinct to do everything opposite of what people are talking about in this room. "we shoudl slow down. we should make fewer mistakes."
You're sitting in your seat thinking, "we should fail fast, we should make mistakes..." you're wrong. You will totally experience shame. (I still do.) F.A.E. <-- "Fundamental Attribution Error" (psychological term). humans fundamentally underestimate the power of the situtaion until your'e in it. Shame is very powerful and will warp your thinking and your team's thinking.
Big idea: adopt economic, not moral mindset.
The point of this: get your team to move away from the moral
Parable: a tale of two factories. They both make widgets. They're missing their goals for different reasons.
Factory 1: broken machine. The way that humans respond to a broken machine is the economic mindset. "how much is it costing us?" "how much to repair? kludge a partial fix? what are risks if we delay a fix?
COST, PARTIAL, RISK are key words.
Factory 2: one of the employees is an axe murderer. so they keep missing productino goals. now, if you ask the same questions, they don't seem quite right. "we can't stop him from killing people, but if we train a bunch of temps, we can bring them in as soon peopel get killed!" ... lol. Key thing: the questions are wrong. the questions are actually obscene. they're not okay. the moral mindset, instead, produces the following things: "search for villains, elevation of accusers, and mobiliations of authority to mete out punishment" (pinker, the blank slate)
most companies, unless they specificaly try to avoid it, will treat them from
When you're running a 5 whys, your challenge is: you want to get your team out of this moral mindset, but it's so hard. Why is it so hard? No matter what you
Everyone is waiting to see who's going ot get punished; everyone's looking for someone else to accuse. If you say "don't worry, I'm ot looking for someone to blame," they won't change their thinking. It's very hard.
Human-robot distinction. Robots don't have a sense of humor. Humor ==
Fundamental tool: make 'em laugh. After that, they're more likely to
Use humor to break them out of fear mindset, then break them in to economic way.
Humor is not a nice to have: it's necessary.
Here are a couple of specific techniques to address this.
Tip 1: always share worse "bad things" (the failboat has arrived). on the continuum of bad things, this is pretty bad. have we lost customers or data? no. (i have done that. it was bad.) did we send 100s of emails to everybody on our customers mailing lists? i have done that. so you gotta let people know about other sucky things so you can see them on a continuum of bad things.
tip 2: mock hindsight bias to its face.
"oh the problem you have with question marks... you have to deploy the config files, and bla bla bla" -- usually you'll have a conscientious developer who says "i totally knew that, it won't happen again." that is unacceptable: it's hindsight. thy can't say that because it's like saying "i was stupid once. i'll never be stupid again and no one at this company will ever be stupid again."
tip 3: relish absurdities of your system
you're gonna discover that your system is this weird mess. your team may feel ashamed of that.
two kinds of startups:
1. achieve some kind of traction on a code base of which they're kind of ashamed
2. the ones that go into business.
that's it. (round of applause)
tip 4: "boradest fixes" vs" root causes"
shared folder. outlook popping up little windows for 3 hours in whole office.
one of our developers had commited a change in the database. inserting rows in database every 10 minutes. we didn't find out for 3 hours cuz our system was emailing us errors 100s of times a minute. our provider (google) shut it down and gradually delivered it to us. then we got an email that was like "here's what's broken" and we could have fixed it if we got
2. our monitoring system failed under stress.
what's the root cause? think about that.
the answer is: we don't care. there's never a single cause. there's always contingencies.
"who really screwed up/ what was the real reason?" this is moral thinking.
if we made a slightly better system to iprove database errors, vs a slightly better system to improve monistoring, which is a bigger win? answer: monitoring.
so we invested a little in database, then a lot in monitoring. much better conversation to have.
remember: there is no axe murderer (probably) so push your teams to go for more economic thinking.
The Challenge of Sustaining Disruptive Innovation When You Meet Success
What we do: work with brands, first
Once ou build your first disruptive innovation,
balancing success and disruptive change is hard, especially in a small startup.
step 1: create the right environment. identify the right resources based on their ability to aid your learning (vs scale). this is after we've done the real prioritiation (we're not working on this, this this). Deep analysis of the types of employees that you have. who is leaning towards bulding new product and new iterations of learning... vs building out
separating them out and giving them these projects. this is actually really empowering.
we actually physically moved this team of engineers into anotehr space. often times skunkwork types do this thing. cultural things (day to day) make a different in the minds of the team. isolating this team out went out with freedom to do what they wanted. gave them power to innovate.
as we grow, there's more risk. want to give them time (but cant say "they'll change the whole company omg" which sets them up for failsure), butn ot too much time (can't just say "do whatever you want"). this leads us to step 2.
step 2: set goals and boundaries. use internal customer development to develop initial hypotheses.
aim to build an mvp that solves a simple business problem.
build a usable product, not just a hack - don't skimp on design. if you get "this looks hacky, i wont' use it" your'e getting feedback on design, not your product, when oyu wanted to get feedback on your product.
separated out feedback and way things were flowing back into company. speed up communiction/learning time
step 3: milestones & rpgoress checks.
keep the progress updates separate from the existing product workflow.
re-asses high-level assumptions nad hypotheses every 3 months.
Be ready for disappointment and some pivoting.
We wanted to
Also wanted you to hear from someone whose credentials were unmatched. This is osmetone who can really
Scott is a legend, I thought I was in trouble! As I account in the book, it was the start of a relationship. It's not enough to be a great entrep yourself -- if you want to have a great company, you have to create entreps out of maybe 1000 people who work for you. Fortune 500 company. Since we've done the onstage interviews before, I asked him to bring some people to see
Creating a Culture of Experimentation: Ideas and Best Practices
SCOTT COOK@intuit Intuit
DR. BHARATH KADABA@Intuit
Journey: convert intuit to become a network of lean startups.
To do that, pioneer real change. Create new businesses and reinvent old businesses.
To fix that, you've got to change how and where decisions are made. Enable decisions to be made by the best idea you can validate in the market. Move decisions from bosses moving with their opinions to customers voting with their feet.
In order to do that, have to change the roles of leaders. Principles of lean startup apply to leaders too. Brad and I have changed the questions that we ask. We used to ask things like "what's your answera dn what's the analysis behidn it?" now we ask "what's the fastest way to have an experimetnig to test that idea?"
In this age of innovation, the role of a leader has to change. The leader has four roles.
1. The leader has to set the grand challenge.
2. The leader has to install the systems and culture to enable people, even junior people, to run fast experiemnts to achieve that grand challenge.
3. The leader has to model: pulling insights from those experiements, no matter if they succeeded or failed.
4. The leader has to live by the same rules and disciplines themselves.
Walk through exaples of each.
Barath Kadaba - VP of global businesses division? He brought lean innovation to Intuit in india.
My story begins in august 2008: Alex, my boss, set a grad vision: create new businesses that improve the financial lives of indians.
I was given enough funding to hire three people. I thought this was insane. How can we impact the lives of 1.2 billion people with 3 people? Undaunted, we quickly put 3 people in place in India and gave them the challenge of figuring this out.
We narrowed the problem a bit and chose to focus on Indian farmers. Typically own less than three acres of land who live on the edge of poverty.
150 million of them.
Spent 3 weeks living with the farmers. After three weeks, choose a narrow problem to solve: who should i sell the produce to in order to realize the "best" price? price of produce varies every day. supply and demand changes at local markets every day os it's hard to decide where to take your stuff to. solution: personalized market price information and connection to buyer via daily text messages. they all have cell phones. number of cell phones in india is approaching a billion right now.
late september figured out what to do.
early octover: we ere "leveraging the cloud" to manually send dialy market prices via text mesages. test in market within days!
overcame many "death threats" from management by staying lean. hard to shut them down cuz theyw ere using so little money.
always passing ipossible toll gates
customer benefit: 20% increase farm income.
1.2 million active farmers, adding lots each week.
- set grand vision
- team very small
- many rapid experiments
- don't be afraid
And you will survive.
So we train our leaders on how to set the grand challenge. We do that by asking leaders to ask questions like these: "what's the biggest pain point that we can solve? And how does the customer measure the gain, and how can we move the needle the most?
Second role of leaders in innovation age: install the systems and the culture to enable anyone, even brand new employees, to run experiments on real customers. this means removing barriers
3 principles/keys that help
mindset -> behavior
becomes how yo work
new tolls so teams can run many expriencts without talking ot legal at all.
take ourselves out of the equation to let
empowers teams to go off and experiment.
have some examples. "if oyur experimetn looks like this you're good to go. if it's an edge case, let's sit down and figure out how to get to yes."
the guidelines themselves were an eperimetn. we applied lean startup principles by working with teams to discover what the answers were.
learned while working with business teams: people want to do the right hting. anything that creates doubt in their mind or make them feel scared: they'll not want to do that, and that's not what we want.
quesiton is: how
document capture app for turbotax online.
we did discvoer that people in general do want to file taxes on their phone. we heard from customers with more complex returns from other states - they wanted to file their returns on their phones too.
today, the app has expanded. the app is availabel in 50 states, for all kinds of customers.
goal: change peoples lives so profoundly that they can't imagine going back.
3:45 PM - 4:15 PM
4:15 PM - 5:30 PM
Late Afternoon Talks
One Year Later: Dropbox Answers Your Questions
DREW HOUSTON@drewhouston Dropbox
What are the top three metrics you focus on at all times?
Maximize the number of happy dropbox users around the world.
It's not specific metrics I look at: you gather them, and monitor in the short term for unexpected changes. You want to be able to identify issues quickly (a lot of money could be at stake, or deeper problem). Cohorts over a long period of time - how does our conversion rate look. Platform mix - what kind of devices do people use dropbox on? There's seasonality in our business. We don't have to become depressed over the summer and manic when kids come back to school.
Are you seeing mobile use go up, or do you want mobile use to go up? Is that strategic for you?
How much share of the overall computing experience. Peple are spending a lot more time on their phones and ipads than on their PCs. Our product needs to evolve to accomodate that. You probably owulnd't start where we started and build a windows syncing client. Given all the things that have chaged in the last 5 years, how would we end up at the same place if we were to start all over?
Ask people (like my sister's friends at college)
Paul Graham had this great essay recently on startup ideas. More about paying attention than inspiration. Quote: You want to live in the future and build what's missing. Try to be a little bit in front. We use things like Office all the time and use Dropbox shared folders for that. And flood of email we have to sweep away every morning. this stuff lives in a hundred different places; be creative to be like, how can this be made better? Problems hidden in plain sight that we can deal with every day. that's what i like to thin about.
process of adding employees from you to employee 25ish? was it engineers? when did marketing people get into the mix?
In the early days, it was all from our network. Met my cofounder through another ycomb friend. Some know them for
I knew arash for an hour before I was like, "yes let's get married" ... fortunately that worked out really well for us. Usually the odds are not in yoru favor for that sort of approach.
The first ten people were engineers. All people we knew from MIT, where we went to school; friends and friends fo friends. Community manager who was arash's friend from high school
One thing that's hard for an engineer: evaluate what a great business person looks like. Imagine startnig a band. You're a really good singer but you don't play guitar. So you need someone that plays guitar. You see someone playing guitar, "that's a beatles song, that's amazing" you ask them to join your band -- that's kind of what hiring businesspeople is like. Then one day you run into jimi hendrix and oyu're like "oh." It goes the other way around too. Anyone who is not technical, they look over and they're like "this kid is typing these incantations and out came a website!" but then we all know there's a huge
What is busiess development at a company like dropbox? If you hire someone that's not good, they hire a team that's not that good -- risk.
do you ever feel like you're doing the wrong thing and you'll succeed...?
Everything starts small. Codebase that's slightly embarrassing but somehow works.
At first, the goal was to make me happy. Best case scenario: something oyu're excited about. While getting code to compile, in back of mind thinking: dropbox is usseful for anyone with a computer and phone. if you have a pulse and an internet connection, you can probably use dropbox. but they weren't like, "let's get a hundred million users... let's just get 10." i bothered 10 random people and asked them what they thought.
whiteboard in boston: took annual salary as engineer in day job. how many people must i get paying me $5 a month so I can do this? few thousand. and i could bother a bunch of people on hacker news or digg and scrounge up that level of interest - and i was like, okay i can quit my job.
Try to prove the idea as quickly as possible. Every time you have a goal, 10x that.
Making the Call on a Platform Pivot
CHARLES HUDSON @chudson SoftTech VC
If you walk away with one thing: even if your'e a VC, you can still make mistakes o_O
We started life as a company on Android.
Defining A Strategy
Why should we exist? Why should anyone care? By the time we started in 2010, the world was awash in mobile games companies. My cofounder (acquired by zynga) and I said: we want to do something different. A lot of really smart people we knew on iOS - what's somethign fun, different, impactful if it worked out? Looked at Nexus One, said "no good games for google play" ... at that time, Android was the ugly stepchild for mobile; weren't that many devices out there; this can't persist! eventully the checkout experience and devices will get better; will be hudnres of millios of people on this platform. we decided to take a close loo kat it.
Focusing on Android
Thought that all three thing would converge, but interesting side approach: we were able to identify ourselves as The Android Guys. recruiting + getting people excited about startup: when we tell people, they'd either recoil in horror, lagh, or give us quizzical looks. helped us separate out people who were just looking for a job or doing something interesting.
highlight the fact that in retrospect, it sounds like a trivial decision but: platform choices matter. different tools in tech
consumer experience + expectations are different
what your company will turn into; who you'll hire
why we pivoted
1. try as hard as we could, could not get monetization on android to hit that celing. looked at the kind of company we were trying to build,not gonna get there alone.
2. funny thing happened. social feature: #1 requested feature was "when will my friends on iOS be able to play this game with me?"
1. do i have the resources to pull this off?
2. do i have the right team to switch gears?
3. how long will it take me to learn this platform, monetization etc? (we thought 6 mo, took a year)
4. what will happen if i continue my current stragetyg? it's one of the most disruptive things you can do to your company
Instead of trashing everything we've done on android, we tried to take the knoweldge we had on the android platform and leverage that into an emerging picture on ios and kindle fire. i'll probably try to talk you out of it but thanks
(We're lucky he's still alive! Welcome dave with an update)
One Year Later: Lessons from Votizen
DAVE BINETTI@dbinetti Votizen
Necesarily implies one year ago. At this conference: using lean startup techniques into tool to.
Steve Blank's customer development + dave maclure's startup metrics for pirates
more than just linearization of build measure learn.
Final thought: importance and centrality of vision.
All of this is dependent on your vision as entreprenuer. decisions you make when the path forward is not certain.
Often, this key point is overlooked while we're in the tactics, techniques, tools, how we do our day to day life in startupland.
After last year, I walked o
your vision is a constant. it is the only constant on which you can rely.
in the vast plain of boundless uncertainty and hope, it never changes.
you need your vision to guide you, inform your decisions, help you accomplish what you want to accomplish.
don't change your vision. share it with others so you can say "that's the mountaintop!"
so no matter what happens, and even if it takes a form you didn't expect when you were first taking off -- you may wander, but you will never be lost.
(change in strategy without a change in vision - is hard!)
thank you, sponsors!
next speaker is winner of lean startup machine workshop.
weekend hackathon, friday night to sunday night - how well they had built a cool widget, but also how they'd experiment
mark abramson @mark__a
We generated $4,500 in the last couple of days using lean
six bi-monthly events at SF's top restaurants.
winner of lean startup machine - two day bootcamp.
validation board: tool we used to run 10 experiemetns and 5 pivots in just two days.
pitch: resto tax. turns out restuarants with
got out of buliding on friday night talked to 7 general managers at resaturants. 0 or 7 experience pain themsevesl. send to account. PIVOT. (4 hours) this experiemtn took us
next up: reached 3 of 40 accountants; 0 of 3 experience pain on this. PIVOT (3 hours)
Number one problem SF restaurants face: getting more people in the door! SALES! So let's drive business to them using
0 of 6 want more Happy Hour patrons. No pain. PIVOT (2 hours). Notice: we were twice as productive (lol).
Let's go back to something we learned during our first experiment.
Restaurants need a better esimate of how many people are going to come in the door on any given night.
Whether hotels or conventions are around nearby: PatronCalc! 0 of 8. NOPE. PIVOT!!!! (20 minutes!!!!!!!!!!!) (audience laugther)
FoodRightNow. We are going to drive business in the door using a location-aware mobile application to send deal alerts driven by restaurant managers in real time! Groupon and $MM VC's doing it - 30 minutes PIVOT!!!
Now it's Satruday night. "We are lean. We are fast. Our ideas suck." Morale is low. We are awesome at invalidating our crappy ideas.
Then Fiona says: hey, what about a wine club for resaturants?
So we iterate on this and come up with chef's table. Six bi-monthly restaurant experiences in 2013. SF's top restuarants featuring SF's top chefs. Seasonal menu not available rest of year. Personal time with chef. Signed menu and prep notes. $1500 for 2 people for an annual subscription.
Riskiest assumption: Can we get $1500 from hardcore foodies?
THE ANSWER IS YES. We closed $4500 in 90 minutes on sunday afternoon.
1. You can do this. we went from selling vaporware to vaporfood.
2. you have to practice to get lean. (i've been to two of these conferences before, and now i'm just getting okay.)
3. as dan martell just said, once you have collected a dollar on the internet, you have gone pro!
+ leanstartupmachine + mentors + wife
looking for investors!!!!
livestream hosts. went up to thousands through different groups!
thank you @michkimble !
What I Learned: A Report from the Weekend Winner
A Conversation with Marc Andreessen
MARC ANDREESSEN @a16z Andreessen Horowitz
ERIC RIES @ericries The Lean Startup
We pivoted twice, which is like 8 fewer than we should have needed to.
The bad news is this was when pivots were still called fuckups. LOL!
1. Interactive television. Time warner had rolled out 60 homes in Florida (40 thousand dollars), gave up on that idea. gave up on that.
2. xboxlive or psnetwork for... n64! We were slightly premature. lol! luckily we punted on that one.
And that was the beginning of netscape.
University == lack of security, so I kept myself on lall the mailing list. Customer suport emails bombarding in which is a good sign. Had inquires for commercial inquiries. Free for academic and nonprofit use and pay for commercial use; were fishing (no price) -- at certain point with 1400 requests in inbox, thought -- this might be a business.
Hit product market fit as a company 1.5 year later.
"product market fit" -- learned from your blog -- inspired me to write everytihng about startups while you were doing it. People throw this term around all the time and ask for help, and if you're clalin gme to ask, you don't have it.
mark: Kind of like calling you and saying "did i just have sex? i'm not sure." looooooooool. "do you have a whiteboard?"
It's actually a feeling. The key to being successful in life is to find a parade and jump in front of it.
startup everyday: pound your head against a wall, is this gonna work? oh my god.
once you hit product market fit, you're like "omg this is running away from me." oh my god, i need to figure out how to get my company to grab this opportunity that we are not yet prepared to grab. oh my god, we have to go get the market. oh my god, if we don't do it, someone else is going to (if you're paranoid like me). lots of stories about people who got product market fit and then someone came out of left field and stole the market out from them. it is a very distinct feeling. from talking to people who've experienced it, you KNOW when you hit it.
eric: people think after you hit it, you're done
marc: people are social, so when your friends are doing something, very strong inclination to jump on; feel is like demand is all coming at same time. you have to build the company to do that. that's when hiring explodes, or international becomes a really big deal. you decide: do i go international or do i wait and let others do it (and then buy them later)
skills / marketing skill set becomes more important; build the engine for customer support and do these things. conscious decisions: we're not lean anymore. we have to become fat because if we don't, someone else is going to come do this
eric: in the months leading up to that moment, were there significant predictors that you were gonna get it?
marc: a couple. real early adopter phenomonen: really smart people that
small group of enthusiastic adopters that really love it, but everyone is laguhing. 3d printing is going through that right now. there's a group of people who are actually manufacturing guns. (wired magazing, last week, funniest headline of all time: 3d printed gun only fires six shots before breaking. should have been: 3d printed gun fires six shots OH MY GOD. hahah
twitter went through this too.
microversion of that: when employees themseves are enthusiastically using the product.
not even friends and family cuz they can be coerced. but when employees are using it, you can feel it.
3 gradations of this thing
1. founders who build the product for themveslve. huge advantage: founder market fit.
2. hp theory (in the old days) - the "next bench theory" - each bench would make the product that the next . voltmeters --> oscilloscope --> down the line; adjacent markets.
mixpanel (spin off of slide, analytics platform). they really wrapped their head around it cuz they've had that problem and are building for adjacent.
3. hardest, and lucrative becuase it's harder: building for something oyur'e not.
lots of enterprise companies are this: workday (hr and financial software). professional product management really matters. super serious, can't let it slide like
guage them on the basis of how strong their product management is.
eric: you and i both lived through the dot com crash. people don't understand what that is
marc: it informs a lot about the psychology of
lean startup is a direct reaction against "go big or go home" from then.
a lot of psychological damage done to people who went through it -- both entrs and investors -- i think we're halfway through it.
i actually think all of the dot com ideas were correct ideas: they were just early.
it's hard to name an idea, even the crazy ones (like webvan) -- those business all work today. amazon grocery stations; diapers.com was succesfull -- industry magazine (800 pages every 3 days of ads of companies selling to each other... good business model while it lasted. legal pyramid scheme lol) -- it's literally a catalog of startup ideas that are working today. a lot of these were - market wasn't ready yet. in early 90s, only 50 millino people online and mostly on AOL. now market is ready for these things.
the dot com bubble itself, i think, was only 18 months long.
we think it's 5 or 6 years
only from the 4th quarter of 98 to the first quarter of 2000. this is really important
95, 96, 97: people were gettin ga little happy (one or two martinis in, not three or four)
98 was actually a financial crisis. asian , russia defaulted, mini version of sept 2008. that rattled investors hard. ebay went public that fall, barely - cuz investors were completely freaked. after asian flue didn't infect the rest of the world, peopel went nuts, came to a screeching halt in 2000.
dotcom bubble was bad, crash was bad, only 18 months out of 20 years - put a stake in it and bury it. get a little worried it's the boogeyman and it's worse cuz it scares us off form attempting big things.
the good news: the 22 year old founders don't have this fear.
one of the challenges of the 35 or 40 year old entr is to function without the trauma that was afflicted then.
eric: i hear of complaints form all my friends who are VCs: entres are coming into their office with the same garbage pitch but leanwashed. would oyu please tell them to stop doing that?
marc: yeah for a while "disruptive" was the thing.
eric: yeah so sorry about that. most common misconceptions taht people bring to you. you et the front row seat of what they look like in the wild, not the rarefied
and i wish we could license only certain people to use our ideas but that's not going to work
marc: there are three.
1st: not evry startup can be done lean. there are some, especially the ones that have the really audacious goals. in some cases, they start audacious and just have to be audacious
old days: macintosh. nobody ever asked for mac, the product had to exist on people
intel 8086 major engineering
silicon graphics, major
workday was a fat startup, 400 employees from
gonna rip all the poeoplesoft customers
they had to build that mass
losirra? vosirra? sold to vmware 1.6 billion. got this price tag because it was a major league engineering effort, going against networking as done by cisco
elan musk - neither tesla nor spacex could have been a lean startup. you gotta get the rocket into space. there's al ot of effort.
eric: in commong: significant techincal risk in addition to market risk.
marc: yeah, had to build up the company before you coudl sell your first computer (in 70s)
cautionary note: shoudln't let the lean startup idea preclude us from
challenge: how to extend the lean startup theory to e
number two: thing that drives us most crazy: lean startup used as an excuse to not take sales/marketing seriously. see this all the time.
"all that matters is product! if we build it, sales/marketing will happen by magic! if you buidl it they will come!" that's the salesmarketing versio of the rpoblem that lean sovles on the prouct develop side.
theory must be
steve blank and mark leslie have made headway with pushes in these areas
these entrs are fooling thsemseves. ways: 1) extrapolating argument in delusional way from precedence. salesforce.com is self service, so my saas app is . well, salesforce once speng 80% on
entres also say google is a service. well 15,000 service
facebook's sales force
twitter's in the process of building up a huge sales force (you need to go get The Money) - ad sales
we see them thinking that thy can take their consumer methodalogy (besides they haven't realized google adnfacebook have big sales forces) - businesses and applications do not sell thesmelves. you have to get the money.
number three: viral. one in a hundred or one in a thousand.
safe assumption is: your product ain't viral. the number of things that are actually viral is very small. i love the magic business models but it's really hard to find an exape of someoe who goes to get th emoney without taking
the third is the most profound or most psychologically interesting. taking the stigma out of the "pivot" is very exciting. one of the huge advantages of the startup culture of the united states and the mentality we have w/ lean startup theory: failures are acceptable as steps on the path to success.
if a faulre is a step through the search, you're running the search to find the right idea, right fit - you fail and you learn ad you ultimately figure it out yoru sixth time. much better than failing
on the other hand, we see entres who give up too easily.
permission to give up too fast. by taking the stigma out, are they really goinna do the havy lifting over long time? whe you talk about the realy succesful entres,
sean parker likess saying being a founder is like chewing glass 0 eventually you enjoy the taste of your own blood O_O
had they given up in year 1 or 2 or even 3 or 4, they would never have made it.
worst case is the failure fetish: it's a badge of honor! i've failed four times! i'm experienced! this is wonderful! there are serial entrs where we say "wow thye've relaly let repeated failrues go to their heads." a little too gleeful about number of failures.
you want to preserve the good about experimenting and pivoting but you don't want to give eople
sarah palin. at least when sarah palin quit after 2 years (governer of alaska), then she decided it was too hard and quit, at least they called her "QUITTER" -- maybe it's time to add a little more stigma. i really admire the ones who succeed, and how have persisted. it's a really big part of success in this industry.
eric: one of our ambitions w/ lean startup is to give people during the flat part of the hockey stick: some kind of early indication of progress. if your'e not at product market fit yet, indicators your'e getting close. you believe andreessen horowitz is the long term vision. long term is what i admire most. not just celebrate pivots but long term vision thinking.
marc: we and warren buffet are the last long term investors in the stock market
we're 10+years on new stuff, they're 10+ years on old stuff, and that's it. everybody else is high frequency trading.
there is very little investable capital - ask any CEO and find out how short term the capital is.
VC is an asset class. there are some VCs that are taking a long view - magic -
entres almost uniquely have a problem with timing, in my experience. one thing successful ones have in common is: they're in complete panic that they're doing it too late. on my god, this idea is so obvious, someone else is going to take my market, work week is m/t/w/th/friday, friday, friday... there's a too-late feeling. in reality, entres live in the future, your'e anticipating a world that doesn't exist yet.
2012 is the year of saas. it's in this year, finishing up, where saas has tipped. even the largest coopanies in the world are goig to move to saas. People have got through issues with security reliability. pipelines of saas companies who are succeeding are incredible.
1999 to 2012 - 13 years to get to that point. why? it just was. stuff finally just got to that point. business model got that point. mass psychology.
1976 microsoft. didn't emerge until 81, 82 when ibm came knocking
finally in 2012, fifth company with this idea
for VCs, that's fine - you just keep reupping. for a founder, though, you only get one swing. the psychological damage from one failure will prevent you from doing it a second time. raise more, rather than less, to give yourself more time, you honestly have no idea how long it's going to take.
eric: one of the biggest misconceptions is lean == cheap.
but both steve blank and i - we raised significant amounts of capital from VCs!
marc: goes back to big reason we stated our firm and i'm on this side of the table now.
you see so many people with such great ideas and so much potential, and so many companies sell out too quickly, give up too quickly, don't raise out of money, the one and only reason for going
corollary: cash is more important than your mother. (old saying, but still applies)
i call it the puzzle of the missing campuses. you drive down 101 and your'e like, oh there's oracle, appel, google - miles where there are no campuses. why aren't there 10 times the number of really imoprtant successful companies? why aren't tehre 100 times? we have the talent and the ideas, but they get taken off the table too quickly.
get taken off the table if they fail, or if they sell too quickly. you get product market fit and the offer comes in and it's lifechanging money so you take it and *snap*
there are 10 big
eric: do you feel that way about netscape sometimes?
marc: a little bit but that's a much longer conversation.
eric: you have two bad options of being acquired by a public company which sucks, or going public which sucks -- really relentless short term pressure. advocating in book: long term stock exchange. different trading rules to faster the kinds of missing companies you're talking about. if we can slow the trading down, impose higher fees -- forget "is that possible," do you think that would nurture the kinds of companies you want to build?
take capital gains tax
the world we live in right now, in the US, is a reaction to the 2000 crash. we live in a world, where on the public side, it's incredibly inhospitable. all in view of protecting investors from themselves. side effect: the number of public
as a long term private investor who can do growth investments and hold them for 10, 15 years -
on the other hand, our companies don't go public for much longer - not great for public. retirement money (schoolteachers to employees of companies) - if you suck all the growth money out, can only go into companies that aren't growing anymore
do not go public until you have built a fortress.
fortress: build all the defenses to all the attacks you're gonna get once you're public.
commanding market positing
executive positions filled
brand/marketing up and running
strong IP position, these days (tragedy). public policy on this has to change, there is no momentum for this from washington.
1999 w/ companies
i took both of my companies public within 18 months of being founded. would not recommend that today. like an extreme sport, not a good idea.
heroes are great: google, mark zuckerb, pushing their IPO out as far as they can, building up their companies. have to keep in mind how incredibly strong your company has to be
eric: "i'm sure a lot of these peopla are saying i'm going to be the next mark andreessen"
marc: lean startup is like discovering the theory of relativity
eric: go ahead and tweet that if you want to
marc: a lot of stuff was not understood. then general relativity came along and it made a lot more sense. now we have a process and methodology for this. advances have been made by erc, and folks like steve and marek lesley --
very important if a rocket ship can go between stars, make sure it can still go inside atmosphere.
it's great if it's good at lean startup, and customer development, ALSO great at old fashioned blocking and tackling sales and support, all the things needed in the real world as it eists.
when you say entrepreneurs, newtoninan physics, you get really excited. now: opportunity to define both old and new.
eric: thank you for
founder of veritas, one of the most successful companies. The sales force learning curve: theory around sales force.
ERIC RIES @ericries The Lean Startup
the actual work of entrepreneurship doesn't happen on stage. it happens in the salt mines! my hope is that all of you learned something of value and want to
"let us make you famous. that's our job." if all that you get from this is some entertainment, then we've failed, but let me know.
to the dedicated fans: we honor your deication. trending topic of day (vanity metric youer' supposed to boo). this is jsut the beginning.