Lean Startup Book Roundup

Earlier this year I got involved with using Lean Startup techniques to help with a new business inside of Adobe. As is my normal style, I read a large number of books to help get my head around the techniques and build up a base of knowledge I could use in the future.  

As an entrepreneur/intrapreneur you have to be able to lie to yourself a little. Otherwise you would just stay at home and not pursue the new idea that you have. But lying to yourself until you ship your product to the marketplace can be expensive and emotionally devastating.
For those of you new to the Lean Startup Methodologies, they are tools that help you stop lying to yourself and check in with reality at all phases of developing your business. My own path with these tools has been somewhat backward. I originally thought we were further along with our business when I picked them up. But as I applied a tool that I thought was appropriate for the phase of development (for instance a retention graph for our private beta), reality would come through and we need to go a step earlier in the chain to find the problem. We finally ended up all they way back at the beginning.
Here are the books I read and what I took away from them.

“The Lean Startup” by Eric Reis : http://theleanstartup.com/
This is a very inspirational book. It goes pretty fast and doesn’t get bogged down. One of the most influential portions for me was the discussion of Vanity Metrics. These are ones that are generally easy to measure and create, but almost always lie and don’t provide real actionable data.
I consider this book a great seed of ideas, but too thin on actual implementation and case studies. Read it, get inspired, get familiar with the terms and concepts, then move on to a different book for implementation.
“Running Lean” by Ash Maurya : http://runninglean.co/
I consider this the next book in the series. A key tenet to stop lying to yourself is to go out and interview potential customers in as neutral a way as possible (it it very easy to influence people so they tell you what you want to hear).
Ash does a great job step by step explaining how to use a Lean Canvas (a derivation of the Business Model Canvas) to help clarify your ideas, find target customers, and then go out an do qualitative interviews. He has a lot of examples of how he used these tools to clarify his thinking and provides good guidance when you use them yourself. 
“Lean Analytics” by Alistair Croll  (Author) , Benjamin Yoskovitz : http://leananalyticsbook.com/  
This book is good when you have an idea of where you are going. While they talk some about the early qualitative interviews you need to get out and do (in their Empathy stage), I found “Running Lean” much more detailed and useful for doing that. But “Lean Analytics” does point out many common problems in interviewing techniques that aren’t really addressed in “Running Lean” and therefore is a nice complement.
“Lean Analytics” has my favorite graphic in it where they list the major lean methodologies that are kicking around and compare and contrast them. They then go ahead and invent their own. That just underscores the fact that LSM is just like Agile Engineering. It’s much more about philosophy and culture then any specific dogma. Dogma is good when getting started as it can help prevent you from heading into the weeds, but as you get familiar with the concepts you can be flexible.
This book was my first introduction to many different revenue generation models (I read it before “Business Model Generation”) and is great for that. Not only does it cover the models, but also what metrics should be tracked, and has many case studies as to how these were applied in the real world.
The book is also awesome for coming up with “The One Metric That Matters”. It can be easy to lose focus with all the things going on, and this is a reminder to regularly pick out what you should be targeting, figuring out how to measure it, and then head for it.
One thing I was disappointed about is that they don’t really address anything about the statistical and/or business significance that you can attach to numbers. When starting out at the beginning of a product, you normally have very small sample sizes and it is very easy to be mislead by your data. When working with small sample sizes you need to be vigilant when to consider your quantitative data suspect and just throw it out lest it influence your thinking (see “Thinking, Fast and Slow” chapter 14). Stay tuned for a blog post on this problem area.
“The Startup Owner’s Manual” by Steve Blank and Bob Dorf : http://www.stevenblank.com/startup_index_qty.html
I really wanted to like this one. It’s intentionally not done in a narrative style and I found it hard to digest. While I occasionally would reference it, I found myself using the other books more often. The book does call this out at the beginning. I think the book would work much better in a classroom setting then how I used it. I found the explanations of _how_ just not fitting how my brain works. The why parts are great.
The book does have a great discussion of the Customer Development methodology (which helped spawn Eric Ries’ thinking in “The Lean Startup) and is good for that reason alone. The other parts are useful, and it has some good case studies. I just opened it to refresh my mind while writing this and I saw a few nuggets that I liked.
On a dorky note, I had a really hard time with the graphic design of the book. I didn’t like the main serif font chosen, and I found the mixture of serif, and sans serif fonts jarring. I also didn’t like the use of whitespace or even the general typographic layout (the leading felt off, particularly in bullet lists). But I’ve worked at Adobe for 15 years so I’m probably more sensitive to these things.
“Business Model Generation” by Alexander Osterwalder and Yves Pigneur : http://www.businessmodelgeneration.com/
Speaking of design, the graphic design of this book is luscious. Do NOT buy this on Kindle. The physical book has a wonderful layout along with useful and interesting imagery. As a business and tech person who has a foot in the visual arts, I found that the graphic design really helped pull me into the book and keep my attention. 
If you have a hard time sitting down and plowing through page after page of text, just go and experience this book. You’ll need the tools in the other books at some point, but this will get you started in thinking like an entrepreneur.
This book provides a common language (and visual language) for communicating and reasoning about all the key parts that go into a business, not just a product. It can be used for any kind of business, so certain parts didn’t make sense for us, but it was a good start. In the end we ended up using the Lean Canvas from Ash Maurya since it was adapted to fit our space better than the generic canvas in this book.
This book really helped my team come to agreement on terminology and allowed us to communicate more clearly with each other. A canvas is much quicker and more effective then a powerpoint with a long presentation.
As you can see, no one book does it all. That’s a good thing since this is a big thing to wrap your head around, and frankly there is no one true way.
I just picked up a copy of “Disciplined Entrepreneurship” (http://disciplinedentrepreneurship.com/) that has come out of MIT Sloan Business school. I’m thrilled to see the East Coast represent! I grew up in Newton, MA and started working in tech back there before relocating to San Francisco and then Seattle. I’ve only read the forward, but he starts by recommending all those books above and pointing out that there is no single book to cover all the steps.
Happy Reading!
   

Thoughts on flying the Cessna 162

Yesterday I had a chance to fly the Cessna 162/Skycatcher. This isn’t a review, just some thoughts on the experience. We just had almost 75 days straight of rain in Seattle. My drive to renton looked like: Joy; Despair; Joy; Despair; Joy as I drove through alternating spots of fog and sun. Luckily for me KRNT ended up in a spot of very clear skies centered around the Lake Young area. Pretty much everyone else looked socked in.

A little about my perspective: I’ve got about 70 hours, all in J-3 Cubs or Champs, only the first six were in a round gauge Evektor Sportstar. I got to the point in my training where I could take the sport pilot checkride back in August 2011, life intervened and I couldn’t. This was my first flight since then.

I’m also out to fly as many different flying contraptions as I can. Gyroplane, weight shift trike, powered parachute, glider, motor-glider, balloon, did I miss any?. Therefore my training is biased heavily towards fundamental stick and rudder skills and learning a lot about aeronautics and aerodynamics. I’m not very focused on getting from point A to point B in an easy and quick manner.

I have a love hate relationship with the classic taildraggers. They definitely made me a much better stick and rudder pilot, hands down. But I occasionally wish for some creature comforts, like cabin heat, and not being passed by a big rig when following I-5 with a small headwind. Or an electrical system. Had I not stopped to explore what the rudder exists for, I’m pretty sure I would have my license by now.

RFS has three 162’s on the line, and they do most of their primary flight training for private pilot in them. They are currently renting at $99/hr wet with no fuel surcharge.  I met with John Miller, one of their full time instructors who recently moved here from Arizona.

The 162 is well thought out for life in a flight training environment. It’s barren inside. In their desire to maximize useful load with the 1320 pound LSA gross weight limit, there is a lot of bare metal and unadorned plastic. I like the Mad Max look and a rattle can of flat black would get you there in a heartbeat :). Nice thing about barren? There isn’t any upholstery to get stained and ugly looking. The only fabric was the carbon fiber seats done in a black cloth.

Fuel management is very reliable. There are fuel gauges in the wing roots that are just floating plastic balls in a tube of fuel. They are only calibrated for in-flight level reading. But the filler necks in both tanks are marked with little holes at 1/2, 3/4, and Full. RFS keeps the planes at 3/4 tanks to improve the available load for students and instructors. That’s around 18 gallons, or 3 hours fuel. Fuel is just on or off and is always fed from both tanks via gravity.

I’m torn about the glass cockpit. So far in my training I’ve chosen to eschew use of a hand-held gps to try and hone my map reading skills. This has been a lifelong weakness for me, so I really wanted to learn how to do it better.

Surprising things for me about the glass:
– no need to slave the HSI to the ‘compass’
– the compass is driven by a magnetometer. That means no whiskey compass errors. Ever.
– You can set heading and altitude bugs and the computer will announce on your headset as you approach them and deviate
– It shows TAS too (derived from the OAT prove and pressure from the kollsman window).
– It has synthetic vision, supposedly useful for accidental VMC->IMC. I fear that it will breed more deliberate flying through a layer by non IFR rated pilots.

The plane was pretty easy to fly. Very little adverse yaw, and man, I’m a little jealous of tricycle gear pilots right now. Takeoff was easy, with none of the excitement when the tail comes off the ground. I think density altitude was approaching -2000 feet, so we launched pretty quickly.

Landing was also straightforward, though it was a very different experience for me as we did a power on approach and had flaps to use. It was a little confusing, with power off abeam the numbers and no flaps (my comfort zone), there aren’t too many variables to play with, pitch to set airspeed, add power if you are short, slip or go around if you are too high. We did full stall landings, so felt very familiar from that angle. It theoretically has a higher wing loading than a champ (11lbs vs ~8lbs), but aside from a very gentle crosswind I didn’t get a chance to test that. Another day.

One big problem I had with the landing? I stayed off the heel brakes like was drummed into me. That means I was applying brakes on touchdown with those handy toe brakes… Even with that abuse the plane refused to depart the runway and groundloop (did I mention I’m a little jealous of the physics of tricycle gear). I’m a little worried about retraining myself when I fly a taildragger with toe brakes.

In flight? Not much to say. Doesn’t buffet as hard as the champ before dropping a wing when stalling, but the horn screams for quite a long time. Departure stalls were more exciting as the pitch was held at 25 degrees until stall. Steep turns were easy, though I had a tendency to lose some altitude there. Site picture being 6’4″ tall was difficult with steep turns to the left.

I’m 6’4″ tall and weigh 170 lbs. Most of my height is in my torso. Comfort was about what I expect. Much more roomy than a 152. The stoke (stick/yoke) just barely hit my knees on full aileron deflection. Better than the back seat of a cub (but really, what isn’t), more cramped than the Champ, though getting in and out was easier. I’m finding that high wing tandem works best for me. While in all cases I stare into the wing root, the tandem gets me further away from it and gives me more sky to look at.

The 162 looks like a nice addition to the planes used for training. I know if I was just starting (and didn’t get bitten by the taildragger bug), going for the 162 with the glass would be a pretty straightforward decision for me.

Am I going to jump ship for the rest of my training? Unlikely. I’m a sentimentalist at heart. Not only does the slow drafty Champ have a long history, this particular one was owned by my friends father back in the 50’s before it was sold to Snohomish Flying Service. I expect I’ll get checked out in the 162 at some point (and the Evektor Sportstar) so those days when I’m looking for something a little more snappy I’ll have choices.

Now if only I could find a SportCub to rent… (http://www.cubcrafters.com/sportcubs2).

Making Machinable Wax

I recently converted my mini-mill to CNC and watching it run full speed with aluminum is a little terrifying. I thought having some machinable wax to play with while I learn would save me a lot of anxiety.
After much searching on the interwebs I decided on a roughly 4:1 mixture of paraffin to LDPE. I decided to add some stearic acid and colorants so my final ratio was more like 4.5:1.
My recipe:
  • 1564g paraffin
  • 340g LDPE 
  • 47g Stearic Acid

All ingredients ready to go
Molten wax and plastic slurry poured into molds.

Milling tests are going to take some time as it’s in disarray at the moment.

Sunrise: 7am; Sunset: 4:45pm

In preparation for short dark days, I decided to build a stand to mount my sunlamp in a better position.

And if you leave in Seattle without a sunlamp, ask your partner (or close friend) what your mood is like in December. I don’t notice my slide into being a crankypants, but my wife certainly does. 20 minutes a day under this puppy from the Indoor Sun Shoppe in Fremont and I’m good to go.

But I’m tired of scrunching under the lamp on my dining room table and instead want to use my more comfortable workstation. Some 1″ 065 steel tubing, tig welder, and drill and I have this nice stand.

0.5 hours Pilot in Command

How do you know someone is a pilot?

They tell you. Again, and again, and again…

Today I managed to convince a different instructor (Peter Swift) in a different plane (1946 Aeronca Champ) at a different airport (Harvey Field, S43) that I’m highly unlikely to kill myself in a plane. So unlikely, that he hopped out and let me fly it myself for a half hour!

I never learned how to smile for a camera. And the sun was in my eyes too.

I’m pretty damn excited.

BTW, I’m getting a Sport Pilot license which requires about 50% of the time and cost of the Private Pilot license. I can fly planes like the one above, and the new flying car that should make it to market in the next couple of years.

Abandoned: CNC Mill

I don’t know about the rest of you, but I start a lot of projects that don’t reach completion. I’ve become emotionally attached to the parts that I dragged around with me so I’m loathe to part with them, but I thought I would start documenting them.

Back when I lived in San Francisco and had full use of my arms I bought a Rong Fu RF-31 Mill/Drill from Enco:

I was having a blast with it and lusting after making it CNC. But then I got a repetitive strain injury and lost the ability to wrench the bolts that hold the head in place. The mill languished for a long time at my studio at the Cataclysmic Megashear Ranch. When I moved to Seattle I left the mill behind in the care of Jamie Nasiatka. This was back in 2004, I haven’t been a great friend and slowly lost touch with him. But a quick search just showed that he finished what I always intended: converting the mill to CNC.

I’ve since purchased a Mini-Mill that fits my shop and arm capabilities. Due to my CNC lust back in San Francisco, I purchased a number of the parts needed to convert the mill to CNC.

The parts are totally useless for my Mini-Mill.

Mini-Mill, small but easy on the arms.

Here are my parts that I’m still trying to come up with a creative use for:

36V 20A Linear Power Supply for the Motors, also from Dan Mauch

Old Servo motor drives from Gecko

Servo motors with encoders from Dan Mauch
Parallel Port breakout card (sorta useless since I don’t have a computer with a parallel port anymore).
Possible uses:
power drive for the lead screw on the 9×20 lathe. I doubt I’ll cnc it, but it would be an mighty accurate power feed if hooked up to an arduino.

data geekery and tuning

I finally finished my data acquisition for the bike. I used an Arduino with an SD Shield to capture the TTL serial data as output by the Cycle Analyst.

Arduino with SD Shield taped to back of bike.

I actually took the bike out to run some errands. One of the annoying things is that I’m using the current and speed control on the Cycle Analyst. It uses a PID loop to do this, and the defaults are tuned for a bike with less power than mine. That means it oscillates. If you look at the image below, this represents me running on flat ground with the throttle wide open. You’ll see the PID loop cranking the amps up and down and the speed oscillating about 3mph. I’m trying out Tableau so you should be able to play with the data.

Full run coming back from Bartells :

Sheet 1
Sheet 1

eBike construction details

I had a great time giving a talk at dorkbot-sea about my build. Lots of great questions.

A couple of people asked how the motor is attached to the controller. Here’s a close up that shows the torque plate machined out of 1/4″ mild steel along with the motor wires with a drip loop.

Torque plate on left side of axle.

Battery Management System showing fully charged (except for one cranky cell)