The book “How Google Works” makes this disturbing statement:
most companies are slow by design
At Metro we try to avoid this but people are used to the “slow” ways. The slow ways may seem like common sense, but they’re not, and here’s why.
1. Effectiveness over efficiency
Efficiency is worthless if you’re efficiently doing the wrong thing. Many businesses fizzle out because they spend too much time and effort doing things that didn’t matter.
We’re guilty too sometimes. Last year we spent 9 months efficiently developing an Android app. The only problem was, we had no real way to drive installs – we spent 9 months efficiently building something that nobody used. We were efficient but not effective.
Our weapon against ineffectiveness is to focus on outcomes not deliverables…
Outcomes not deliverables
This is the essence of the Lean movement, popularized by Lean Startup.
Deliverables are things like wireframes or new widgets. Outcomes are things like “increased return visits” or “more page views”.
“Outcomes not deliverables” is how the Government Digital Services team won a design award for gov.uk (detail here). A government site winning a design award?? They must be doing something right.
For people coming from a Scrum background, think of Lean as an upgrade. (Lean Startup landed in 2011, ten years after Scrum. A lot changed in those ten years… social networks, smartphones and the Cloud to name a few).
2. Connecting the dots
In today’s digital economy, companies either innovate or die. Steve Jobs described innovation as “connecting the dots”.
Google echo this, saying “combining technical depth with business savvy and creative flair… they are the key to achieving success in the Internet Century”. (Detail here).
We’ve certainly found at Metro than the best ideas come from every corner of the company.
Lean UX describes it this way,
Lean UX demands a high level of collaboration between these disciplines…
The creation of these diverse teams collapses the gated-handoff process known as waterfall. Insight on each idea is brought in from all relevant disciplines earlier in the process… Conversation is encouraged across functional silos…
Now for the science bit
The book Social Physics describes the latest science behind effective teams, backed by big data. It finds that the biggest driver of team effectiveness is the “pattern of idea flow”, specifically:
- A large number of ideas
- Dense interactions (conversations, not monologues)
- Diversity of ideas
In other words, connect the dots. Surprisingly, they found that this is more important than individual intelligence, personality and skill! (Detail here).
3. Small steps
Earlier we mentioned “outcomes over deliverables”, but how do you get meaningful feedback about outcomes?
The most robust way is to do controlled experiments: launch one change at a time and measure the impact.
It’s summed up by Matt Damon in The Martian, “In the face of overwhelming odds, I’m faced with only one option. I’m going to have to science the sh*t out of this”.
Take small steps in a promising direction, and get meaningful feedback about each step. It’s better than running a mile in a promising direction, and then finding out it was actually the wrong direction.
Even the first version of the iPhone was relatively basic – it didn’t support third party apps!
I think of it as “big ideas meet small experiments”.
4. It’s an ongoing conversation
It’s not enough to just “connect the dots” in a planning meeting or once a week. There needs to be ongoing interaction and feedback.
Lean UX says, “Lean UX demands a high level of collaboration between [all] these disciplines. Their involvement must be continuous, from day one… until the end”.
We really value these conversations. The problem is, they can get out of hand when there is a lot of concurrent work going on. We prefer to reduce the amount of concurrent work, in order to deliver a more “joined up” product.
Limiting work in progress is a popular ingredient of lean development. In the words of Steve Jobs,
Innovation is saying “no” to 1,000 things
We introduced a roadmap at the start of 2015 and found that this helped us stay focused. It’s simply a list of the next half-dozen themes that we want to tackle. (Successful themes have included editorial feedback, engagement and SEO).
- Limiting work in progress combined with “small steps” means that we’re laser-focused
- “Outcomes over deliverables” means that we’re laser focused on the right thing
- “Connecting the dots” means that our efforts are more likely to deliver results
There’s just one major ingredient left…
5. Motivation 2.0
The biggest factors that motivate people (assuming they get paid a fair salary) are: Autonomy, Mastery and Purpose. Check out this awesome TED talk about it…
An example of purpose is when someone isn’t just given a task to do, but is given the context about why it’s important. This also makes it easier to give people autonomy.
An effective path to mastery is to learn from other people while collaborating with them (when two developers pair up on a task, for example). Our approach of “small steps” and “less concurrent work” makes it easier for this sort of collaboration to happen.
If you’re freaking out, it’s ok. Some of us freaked out too when we joined the team. Now we freak out at the thought of going back to the old ways.
It’s ok. Grab a developer, they can buy you a coffee and start to explain how all this works. It’s not as crazy as it sounds, or your money back.
Want some more first? Here’s how other companies are following a similar path:
- How we apply focus to grow faster by the VP of growth at HubSpot
- ThoughtWorks: Great post on Lean UX by their UX team and a great post on Continuous Design by their design team.
Still want more?
Check out this great TED talk about cultivating innovation.
Plus, these awesome slides from the head of Google about how they stay awesome.