Companies for craftsmen

(1)

Companies for

craftsmen

Carl Erickson

@carl_erickson

Software Craftsmanship North America, 2011

Sunday, November 20, 11

I co-founded Atomic Object in 2001 with my partner Bill Bereza.

Then as now we wanted to build great software and get better at doing so.

Since what I’m going to talk about this morning represents things I’ve learned from building and growing Atomic, I thought it might be nice to give you some context...

(2)

@carl_erickson

These are the Atoms of Atomic.

In the 10 years we’ve been in business we’ve grown to 33 people, mostly developers and designers. And there’s always been a dog.

We build web, mobile and embedded software applications for clients ranging from startups to the Fortune 500...

(3)

@carl_erickson

web

This is catalog choice, the don’t call list of paper catalogs. You might be familiar with them from their prolific sponsorship of National Public Radio. It has 1.2M users.

(4)

@carl_erickson

mobile

We build iOS, Android and BlackBerry native apps and web mobile apps.

(5)

@carl_erickson

embedded

We got into embedded development 6 years ago when a firmware engineer at one of our customers told us “you can’t test my stuff”.

Telling someone at Atomic that you can’t test something is like waving a red flag in front of a bull.

As a result of that challenge, we pioneered agile development practices in the embedded space.

You might notice I didn’t mention desktop.

A Java app for automotive factories was actually our very first project. We still do that work, but it’s been significantly displaced by web over the years.

(6)

@carl_erickson

We practice our craft from an old post office building in the core city of Grand Rapids, Michigan.

Having our own building is great, until the air conditioner or the roof needs replacing. About a year ago I started to feel a little restless.

Upon reflection I noticed that my adult life so far has fallen into pretty neat decade boundaries

-the 1980s being a student in higher education

-the 1990s teaching at a university

-the 2000s starting and building Atomic

When I thought about the next decade I realized that I had the luxury, from our size and success, to do something different.

I was still having a lot of fun, and being challenged on a daily basis, so I didn’t want to make a wholesale change.

(7)

@carl_erickson

GreatNotBig is a blog I launched in February of this year as an outlet to share what I’ve learned about building and running a company for software craftsmen.

It’s just a hobby, really. And it doesn’t get that much time from me.

It’s part giving back, part documenting and teaching at Atomic, and part therapy for me. So that’s that context from which I’m speaking to you this morning on companies for software craftsmen.

When I talk or write about this subject I feel good. I feel good because there is an alignment between the interests of the software craftsmen, their companies, and the companies’ clients. I see a virtuous cycle...

(8)

@carl_erickson

customers

value

demand

makers

Customers are what we all need...

They expect to have value delivered to them or they won’t be customers.

And in fact they need to be happy customers for this to be sustainable. Value covers a lot of ground

quality, predictability in time & budget, creativity, respect for their time

satisfying their emotional needs, responsiveness, alignment with their business goals By receiving value, the customers generate demand for our services...

That brings makers into the picture...

that’s a word I use for developers, designers, testers, etc.

And of course it’s the makers that close the loop by delivering the value which generates demand...

Since people generally do their best work when they’re happy, we can control the power of this virtuous cycle by concentrating on the happiness of our makers...

(9)

@carl_erickson

Happy makers and happy teams. Then how do we get it?

(I think this was from one of our monthly SpinDown parties, so these guys might look unusually happy.)

Do toys make you happy? nerf battles in the office, anyone? Is it your title that makes you happy?

How about pizza parties? Fancy ergonomic chairs? Can we buy happiness?...

(10)

@carl_erickson

How many of you would be happier on Monday morning if your boss gave you a $10,000 raise?...

The results of a very large study came out in the last year or so.

It showed that your day-to-day contentment and mood can indeed be improved, but only up to the point where you’re making $75k a year.

After that, more money doesn’t equate to more daily happiness.

Studies have also shown that we all adjust quickly to changes in compensation. We’re happy at first, but it quickly wears off.

So while money can be a powerful de-motivator (fairness issue), it’s not a very effective lever for creating happy makers.

(11)

@carl_erickson

Pink differentiated between extrinsic motivations such as money, and intrinsic motivations. His claim is that intrinsic motivations are much more powerful.

Pink identified three intrinsic motivations...

Mastery - mastery is the opportunity to get better, to master your craft...

Autonomy - people want control over the work they do and how they do it...

Purpose - belief in the work you do, or alignment with the mission of the company

These things are common to all humans, not just makers. I find them very useful touch points when we’re working on how Atomic does things.

But there are also two others that I’d add that I think are relevant to the goal of happy makers...

(12)

@carl_erickson

Love of Craft

Team Work

Love of craft - interesting work, variety, ability to control quality, seeing people use what you build, building elegant programs

Team work -

I believe that accomplishing something with other people is a great way to live a satisfying life,

team work is knowing you’ve got help when you need it,

it’s a sense of camaraderie that makes people connected and happy.

It also might be important if you want to see your work used, or if your purpose requires more than solo effort.

I think the team is the best unit of organization for a company of craftsmen.

-it’s project- and hence client-centric

-it focuses on the makers

-it’s an alternative to functional or hierarchical organizations

If working in a team is important, and we all need jobs, then the question arises as to what sort of company is best for craftsmen...

(13)

@carl_erickson

?

The two ends of the spectrum are freelance, solo work, and working for a big corporation. What I’m excited about is something in the middle...

(14)

@carl_erickson

Innovation Services

Innovation services firms sell their expertise in innovation as a service.

I’ve found examples of innovation services firms in software, product design, mechanical design, entertainment, special effects, and environment design.

Software development companies like Atomic Object, 8th Light, Obtiva, Relevance, and Edge Case fit my definition nicely.

It’s all too common in the technical world for people to conflate their passions with their preferences. That’s where the endless flame wars of Ruby vs Clojure or PHP vs Rails come from.

I suspect a common element of innovation services firms is that they show a passion for

higher-level concerns -- things like quality, fit, predictability, maintainability, productivity -- and a preference for lower-level concerns like languages, platforms, and tools.

(15)

@carl_erickson

•Sell services, not products

•Makers, not just advisors

•Work in teams

•Collaborate with client

The common traits are that they:

-sell their services and don’t have their own products

-make stuff. they aren’t just advisors or consultants

-work in teams, often coordinating multiple disciplines or skill sets

-collaborate with their clients

I believe innovation services firms are an under-appreciated yet critical component of the economy and of our communities.

-They are incubators of talent. They attract the best practitioners and train new people.

-They are expert at innovation. They focus on the process of innovation rather than a single product line.

These firms exist in a symbiotic relationship with their clients. They use their expertise to create products that increase revenue or define new markets, and they do it alongside their clients...

(16)

@carl_erickson

Innovation Services

Software Craftsmanship

+

Because they sell their services, these business have no leverage.

-It’s an hour out, a dollar in.

And they don’t typically have an exit the way a product company might.

After all, who’s ever heard of a software development services firm being purchased, right? Innovation services firms usually have two types of clients:

-well-funded startups (who don’t know how to build a good team)

-large companies (who have budgets, but other priorities)

One of the things that make an innovation services firm a challenging place to work is their competition:

-they need to be significantly better than an internal team, and their external competition

-as a consultant, they are only as good as their last project

On the other hand, following the path of software craftsmanship is not taking the easy route either.

-many more ways of building software wrong than right

-software systems are naturally unconstrained, unlike other other engineering disciplines

-our craft requires life-long learning, lots of which is done on your own time

I think in fact that innovation services firms are a natural match to software craftsmanship. So what do these kinds of companies for software craftsmen look like?...

(17)

@carl_erickson

There are three things I think you need to get right to have a sustainable company. First, you need to have a reasonably good idea. And you must be able to execute well. But idea and execution aren’t enough.

You also you need to create and nurture a powerful culture.

What I’m going to talk about next are aspects of culture. These are more abstract and fundamental than particular business practices.

Over the years as I’ve traded stories with owners of other innovation services firms I find that the cultural aspects are common, but the practices vary between companies.

(18)

@carl_erickson

transparency

Transparency has at least four important dimensions: facilities, business practices, clients, projects

I believe it is really important to have alignment and consistency across those dimensions. Facilities can be “open” while being very closed...

(19)

@carl_erickson

I happen to live in the region where the “open office” was invented. Sorry about that.

As I’m sure any of you who have worked in these environments knows, cubes combine the worst aspects of private offices and one big room.

They tend to be sterile, isolating, un-human, neat, clean, and orderly. In contrast to this picture from our office which is...

(20)

@carl_erickson

messy, crowded, sometimes noisy, collaborative, informal, and human contrast those pictures <- ->

This environment spreads knowledge, lets you know what others are doing, makes it impossible to hide, makes it easy to ask for help or offer help.

(21)

@carl_erickson

finances

sales

marketing

strategy

We’ve practiced something called open books management since I invented the idea in 2001. Any laughs on that? I found out a few years ago that a guy named John Case also invented the idea of open books management in the early nineties.

What this means is every employee understands the financial model of the company, has data on the state of the company, and knows how his or her work impacts the company.

I periodically teach an “Economics of AO” class for new employees and anyone needing a refresher.

At quarterly meetings we review

profit & loss statement

sales pipeline

We have radiators for our marketing efforts...

(22)

@carl_erickson

This shows a snapshot of traffic on our website.

It also shows an internal measurement of participation in the company blog. You can see:

blog posts per month

blogging status of individuals (green, yellow, red)

progress against the goal of a consistent posting rate

(23)

@carl_erickson

clients

projects

teams

They should know who’s working on their projects, how much time’s been spent, and on what features.

There are vital human connections here:

Craftsmen should know for whom they are working.

Clients should know the people creating their products for them.

(24)

@carl_erickson

0 83 167 250 333 417 500 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Points Iterations

Total Points in Project Total Points Completed Total points line

The weekly burn chart shows changes in project scope, historical velocity, projected completion

(25)

@carl_erickson

Teams should know where things stand with their own projects and other team’s projects. One way of codifying your culture is through a statement of values...

(26)

@carl_erickson

values

If you know why you’re in business, and you have shared, clearly articulated values, you’ll find more alignment, easier recruiting, and a common purpose

To describe your values it helps to know why you exist. Your ‘why’ might not be obvious...

(27)

@carl_erickson

This is Simon Sinek’s idea. Every organization has these elements:

what - you do (“we build software applications”)

how - you do it (“iteratively, test-driven, in pairs, etc”) and lastly,

why - why does your organization exist?

Articulating your why helps bring alignment in your company or team and might even be valuable for marketing and recruitment.

I believe innovation services firms have a hard time articulating their why because of what they do.

My first “why” statements were all about our what and how. Then I realized that the what and how are part of why we exist.

Atomic’s “why” is

To build great software, and find better ways of doing it

That was true in 2001, is still true in 2011, and I would expect it to remain for the life of the company.

(28)

Atomic Value Mantras

•

Give a shit

•

Share the pain

•

Teach and learn

•

Own it

Our value mantras are like pattern names. They communicate a whole lot to people in the know.

We use them on a near-daily basis. give a shit...

share the pain... teach and learn... own in...

We refer to these when we’re deciding company guidelines, how to handle tricky situations, employee expectations, recruiting, assignments

(29)

@carl_erickson

It helps maintain our culture. which is all about:

co-located

polyskilled

self-managing

teams of makers

it’s what reminds us that spending $60k a year on professional development makes great business sense...

(30)

@carl_erickson

This is Marissa and Brittany on a layover in Denver on their way to attending a usability workshop

(31)

@carl_erickson

trust

I see multiple layers of trust at these companies.

There’s the trust that the employees have in the owners to treat them fairly, share the gains of the collective work, and shepherd the company responsibly.

There’s the trust exhibited between colleagues on a team to pull their weight, share the pain, do good work, and not sully their reputation.

There’s the trust that the leaders have in their employees and teams to fulfill the expectations of clients.

There’s the trust the clients have that their teams work hard to deliver value and maintain quality.

There’s the trust that the team has in the clients to allow them to do their work professionally and to not blame them for things no one could anticipate.

One of the reasons that growing these kinds of companies is difficult is that it is difficult to scale trust.

Trust is earned slowly and can be destroyed quickly.

Transparency creates opportunities to build trust. It keeps you honest, so to speak, when as a leader you might be tempted for the sake of expediency to do something that could

undermine trust.

Social connections are another common element of the culture of companies built for craftsmen...

(32)

@carl_erickson

friendship

I think that social connections, or friendships, are going to arise whether you encourage them or not.

Investing yourself socially in your work has some risks associated with it. It’s like an

investment portfolio over-weighted in one particular asset class: you have a lot of eggs in one basket if your colleagues are also friends.

But it’s going to happen whether you like it or not. They happen because of

-shared values

-common goals

-time spent together

Since innovation services firms are intense places to work, we create intentional time to be soical, have fun, blow off steam together.

I mentioned our monthly SpinDown events earlier. Last summer we took the whole company up to a historic in on Lake Michigan to celebrate our 10 year birthday...

(33)

@carl_erickson

This was not only a lot of fun, it gave us a chance to tell stories, bring new Atoms into the culture, strengthen existing bonds, and build some new ones.

Those are the common elements I’ve found at companies built for craftsmen:

transparency

clearly articulated values

trust

social connections

There is a much wider variety of business practices at these companies. I’ve left the end of this talk open for questions and discussion.

Here are a few specific areas that I have ideas on how practices can make companies for craftsmen...

(34)

@carl_erickson

34

hiring & firing

tools

innovation

professional dev

standards

assignments

compensation

governance

ownership

workload

marketing

(35)

@carl_erickson

Close:

Action you should take:

find one of these companies to work for

start your own

-to make more great companies for craftsmen

-to help the economic competitiveness of your communities Advantages you have

as technical founders

by being able to bootstrap