There’s a funny scene in the Matrix when the oracle first meets Neo. Having spent her whole life waiting for his arrival she’s pleasantly surprised by his good looks, but disappointed when he opens his mouth.
I thought about that as I was trying to teach my dog some tricks the other day. If you’re new to the world of animal training, as I am, you might be surprised to learn that when it comes to intra-species communication the less you talk the clearer your message will come across.
The idea is to minimise the signal to noise-ratio and if you think about it, voice commands are really horribly imprecise. By the time you’ve said “Goood booy” millions of neurons have fired in your dogs brain. He did something right but can’t figure out exactly what or when.
So most people serious about communicating with animals have learned to talk less, and instead use a little device called a clicker. It’s deceivingly simple, just a chunk of plastic wrapped around a sheet of tin that makes a click as a button is pressed.
The sound can’t be mistaken for any other random noise generated by human beings and the fact that it exists in time only a few milliseconds means it can be closely tied to certain events. It turns out these exact qualities are key when it comes to having animals decode our intentions. In this case at least, less really is more.
(Side-note: For a fascinating backstory I recommend reading Karen Pryor’s memoirs Reaching the Animal Mind, where she tells the story of how she experimented with no-punishment approaches to shaping animal behaviours and eventually came up with the clicker method. For a more hands-on instruction manual there’s also her best-selling book Don’t Shoot the Dog. As a funny side-note within the side-note, what got me buying *that* book was the fact that a psychologist friend — with no interest in dog training — revealed to me that she regularly recommended her clients to read it, since to her mind it says just as much about how human behaviour works. I guess we’re all animals at the end of the day.)
The clicker technique is now being used far beyond animal training. Olympic coaches apply it to fine-tune their gymnasts. Speech therapists use it to help clients get rid of verbal ticks. It’s said to work wonders for golf swings.
And I’m starting to think this school of thought can be applied to how we work, too.
I recently experienced a beautiful application of its core principles. A team I had been following for a while had grown to the point where it became unwieldy. The daily standups used to be snappy but had started feeling like they just wasted everyone’s precious time. In spite of the best of intentions, people were busier keeping each other up to date than with making actual progress. Things were getting out of hand.
So the founder introduced a new routine that literally got people to shut up: Instead of sticking to the standard format of morning standup meetings where everyone goes through what thet did the day before; what what they’re planning for the day and what help they need need to move forward, each team member had to pick a color. You’re either green, yellow or red. And here’s the thing: only those who are in the red gets air-time.
Red means you’re having serious trouble and won’t be able to deliver on what you’ve committed to. There could be any number of reasons for that, but whatever they are, the rest of the team have to find a way to adapt, and so that’s what everyone is focusing on. The standup meetings are back on track.
Such a simple little hack, and it works because of what’s omitted.
Jake Knapp, Google insider, writer and self professed “process geek” makes a similar point in his book Sprint, where he presents a methodology that allows teams to Solve Big Problems and Test New Ideas in Just Five Days.
It is one of those books I keep coming back to, because I’ve seen its tenets work well in the wild. And here is why: it provides a number of tools that allows a team to make progress together while avoiding group thinking. The trick, largely, is to be disciplined with parallelising processes and thereby resisting the natural tendency for the group to “get to the bottom” and “talk things through”.
The process seems a lot like concurrent engineering or deep collaboration, which were two terms Steve Jobs used interchangeably to explain Apple’s secret sauce; a company with zero committees and uniquely responsible individuals for every last detail.
But wait a minute, Jobs used to describe that process as super messy and chaotic, the opposite of a controlled sequential workflow where design departments place orders to development teams and gets disappointed with the watered down results that eventually gets delivered. The Apple way seems more like a mediterranean family dinner where everyone is talking over each other. So how does this fit in with the theme of minimising noise?
Well, let’s not be too literal. What the clicker method boils down to really is that communication needs to be timely and precise in order to have maximum effect. That doesn’t necessarily mean that a quite team is more productive than a noisy one, but it does mean that the same number of words can carry different amounts of information, just as a gallon of rocket fuel has higher power density than the equivalent volume of diesel.
Speaking of Apple; the talk-less principle also seems to apply to the art of pitching. Carmine Gallo spends a chapter of his book The Presentation Secrets of Steve Jobs to contrast the exquisite simplicity of the icon of Apple to the cumbersome jargon Bill Gates typically used on stage.
Running transcripts of both speakers through algorithms for quantitative linguistic analysis results in clear objective metrics: the Apple message always did come across with a lot more clarity.
And it’s not just what’s said. While the average powerpoint-slide contains 44 words, a Steve Jobs slide would typically have one short sentence, or often simply one word.
So how can the talk-less principle be codified in a way that makes sense for human-to-human communication?
I think it can be inspiring to look at proven design patterns for how to construct highly scalable complex systems. Take something like Service Oriented Architecture. It’s really an umbrella term for a number of concepts, one of which is the principle that every individual service hides its complexity to the rest of the system. The functionality is exposed via a simple interface that conceals everything that goes on under the hood.
Millions of moving parts make up a system where the highest abstraction layer is blissfully unaware of its own complexity. (Incidentally this pattern is strongly inspired by how complex organic systems work.)
One of the particulars with Service Oriented Architectures is the fact that subsystems are often written in different special purpose languages. (In fact, this can be seen as its main distinguishing aspect compared to good old object oriented programming).
The magic that allows for this, the cost of concealing complexity if you like, is the universal adherence to a generic communications layer, a protocol, that allows just enough expressiveness for services to be able to send each other the necessary high octane signals that make the whole more than the sum of its parts.
And here’s where we’re back to what makes great teams tick: It’s the fact that they’ve arrived at a common protocol that efficiently cancel out noise, thus shielding each other from the messy complexity that needs to go on behind every curtain. In a word, they’ve learned to be brief.
***
PS.
(What concurrent engineering feels like for someone in the trenches is rendered beautifully in Ken Kociencda’s book Creative Selection : Inside Apple’s Design Process During the Golden Age of Steve Jobs.)