Five Books — ACBDE of Communicating as a Tech Leader

Always Listen — it is two-thirds of effective communication.
Be Precise — human-to-human data bandwidth is 10 bits-per-second.
Choose Phrases Well — there are hundreds of time-tested rules.
Debate using numbers directly- whenever possible, make numbers relatable. Extract data into right chart — pay attention to color, axes and scale.

1. You’re Not Listening

Why

When we do not have time to listen, we say “I don’t have time to talk now”. That probably says everything about how much we value listening. Hearing is one of the first senses we gain — a fetus can distinguish between human voice and other noise from 16th week, and the last but one sense we lose — right before we lose feeling someone’s touch — and yet, the “2048-page International Encyclopedia of Interpersonal Communication” has only 1 entry for listening. It is a skill and one can never master it enough.

I particularly am a bad listener. While trying to improve it, I luckily came across this gem of a book. It is full of actionable insights I believe will be useful to anyone, especially leaders whose key task is to figure out stuff from listening well.

• Most “business consultants” basically advise to echo and signal appropriately, e.g., nod, “you’re right”, “hmmm” as a proxy of “active listening”. It’s mostly useless.
• Active listening is an exercise in passivity. It is a form of “emotional attuning” — opposite of algorithmic approaches.
• Good listening starts with a belief — “what I know is different from what you know”, and accept that “negative capability” — even if they differ a lot — if not embrace it.
• What’s best in music is not found in the notes. Silence within conversations between two people is culture dependent. In Japan and Finland, the average pauses are up to 400% longer than in the US.
• “Tis the good reader that makes a good book”. Continual good listening feedback loop makes a successful conversation.
• More than half the emotional meaning is non-verbal. Why we often bring up emotionally draining topics during cooking, or driving. Also why confession boxes have a separator.
• “We can readily accept that we can be wrong, but we are never wrong about what we like or dislike.”

One thing from the book

Shift Response vs. Support Response — When someone says, “I just went to Hawaii last week”, we often shift the response to ourselves. I, especially, am very guilty of this! We often say, “Oh! I was just there last year. Which island?” This is bad listening. This is, as the author rightly points out, a form of “conversational narcissism”.

Support response is “So you finally got a good vacation. Tell me what was the best part?”

The book was completely worth it for this lesson alone!

Similar
A quote — “Wisdom is the reward you get for a lifetime of listening when you’d have preferred to talk.”

2. Make it Clear

Why

Good communication may not always result in success but bad communication most definitely leads to leadership failure. The biggest ones! Added to that risk, communicating “tech” requires an extra degree of simplification, especially if we have the “curse of knowledge”. Fewer than half engineers, leave alone others, will understand this line from a recent CloudFlare RCA — “While deploying a change to our prefix advertisement policies, a re-ordering of terms caused us to withdraw a critical subset of prefixes.”

This book, written by a late MIT professor who also very successfully taught this course for 40 years, offers a good framework for analytical people. It argues all good communicators are remembered much after it is over. People remember them because they generally offer (one or more of) 5S-s — slogan, symbol, saliency, surprise, and story.

Frameworks are like colors in a paintshop — none is wrong but you should choose the one that you like. Not all are listed in the book, obviously. For example, I find “What-So What-Now What” a great pattern when trying to influence very senior leadership.

One thing from the book

Shannon Channel Overflow — Every data channel has a fixed capacity. Error free transmission can only happen if the data rate is the same or lower than that capacity. Human-to-human bandwidth — through voice or writing — is 10 bits per second. Transmitting any higher than that creates data loss. This is why we must communicate using simple words, pass along known abstractions (e.g., metaphors to send a concept than the whole structure) and make it easy to recall (e.g., rhetorical tactic).

Similar
MIT has a superb 1 hour video that summarizes the book.

3. Elements of Eloquence

Why

Few years ago we wanted to launch a “Digital Wallet”. The idea came during fall. We then decided to “ship” it on Feb 14 for a GA. We named the target “Wallentine’s Day”! It was excruciating, but the team did launch it — I suspect because no one wanted to spoil a good slogan.

Good phrases are forever. A simple choice between two similar words could be the difference between passionate followers and a stifled yawn. If you doubt, just try replacing “I have a dream” with “I have a multi-year vision”. It is said Churchill mobilized the English language and sent it to the battles. He won! Good choice of phrases could mean your ideas wearing a tuxedo or LBD. This short, eminently readable and often hilarious book teaches how. It gives a wide set of examples from Shakespeare to JFK to hip-hop.

Rhetoric is a thousands of years old discipline. It also changed very little from the time Greeks provided structure and naming around it. For example, Diacope is a verbal hot dog — James sandwiched between two Bonds. Or, Erotesis — a question that is meaningless. “Shall I compare thee to a summer’s day?”. Or, the most important trick of all — tricolon — saying it in 3s. Churchill apparently said “nothing to offer but blood, toil, tears and sweat.” But the tricolon is so strong that everybody remembers the line as “blood, sweat and tears.”

If you read just one book on communication, this is it! If anything can survive for the practical length of human civilization, as rhetoric did, embrace it as universal truth.

One thing from the book

Steve Ballmer practiced Epizeuxis in this infamous video — he repeated a word immediately in exactly the same sense.

Similar
“Thank You for Arguing” is also very readable and equally useful, though slightly less funny, and a bit bigger.

4. Making Numbers Count

Why

Leadership is, essentially, setting a vision, frame a strategy to go there, relentlessly execute toward it, and measure what matters to fine-tune the execution. While vision and strategy are most important, measuring progress and acting on it occupy a much larger share of our time. Therefore the ability to deal with numbers at “downstream” is a superpower.

Most leaders, especially in tech, are good in math. That makes them bad at communicating numbers — curse of knowledge. This short book teaches how to contextualize, humanize and make numbers relatable with 30 rules. Numbers are not real things. They are symbols. Telling someone ”the solar system’s largest volcano is on Mars and it is 14 miles tall” is like asking a developer to code something in machine language. Reframing it to “Mars has a volcano that, if a United passenger plane could fly there in regular cruising altitude of the earth — it would crash in the middle of it” is like writing a human-readable high level programming language.

Few takeaways from the book -

• Whenever we are communicating numbers, we are actually NOT communicating THE number. Pay attention to the thing we are trying to do. E.g., “tumor as big as a peach” rather than “a 3 inch wide tumor”
• Principal rule of communication is to be short. See the reference above about human bandwidth — 10 bits-per-second. We do not want to flood the channel. However, since numbers above 3 or 4 are not “innate”, it is always a good idea to widen the translation when making a number relatable. People intuit “about the number of breaths in a human lifetime” much better than “2 billion”.
• Lincoln did not use ‘90 years ago’, we remember him for “four score and 10 years ago” — always widen the number to translate. It sustains our 10 bps data channel!
• We especially have a problem interpreting very large numbers. Just look up our national debt!
• Humans are instinctively much better relating to time, especially up to a calendar year. If a number scale is translated to time — say, over a day — it sticks. There is a famous example of if the big bang was at midnight, humans just arrived after 11:59:59 PM. That sticks!

One thing from the book

Offer an Encore — “surprise” from the book above — establish a pattern and then break it with numbers. When introducing MacBook Air, Steve Jobs shared the then thinnest laptop’s (Sony TZ) thickest and thinnest (0.8 inch) parts. He then revealed the MacBook Air Profile — and simply said “the thickest (0.76 inch) part of MacBook Air is still thinner than the thinnest part of the TZ”.

That is also a nifty rhetorical trick — note the “th”s in that sentence — “our thickest is thinner than their thinnest”.

Similar
Powers of Ten is an incredible video to see this in action.

5. Fundamentals of Data Visualizations

Why

Metaphors visualize ideas. Good charts visualize data. As a leader, you would often have insight from data. Often the next step is to influence others with data and try to have alignment on an action founded on that insight. Good visuals are the best influencing tool in such cases. A while ago, at a place we were growing really fast — grew almost 10x traffic in a year, some of it internationally — we realized we needed to stop our prolonged “monthly scheduled maintenance window”. A colored heat map, where we showed the amount of money($) moving every hour for all 168 hours a week, ensured buy-in to significantly invest into reducing, if not entirely eliminating, that downtime. Charts changed the world — from Florence Nightingale’s to Iraq WMD!

This book is a recent, framework agnostic summary of key methods, principles and great full-color examples of awesome data visualizations -

1. Offers a comprehensive taxonomy of “what” and “when” — from stacked bar-charts to choropleths with BOTH good and bad examples.
2. Concisely lays out a set of intuitive and agreeable principles — especially, on choice of colors — e.g., dark yellow is almost absent in nature; dealing with multi-dimensional data; representing correlations etc
3. Analyzes top-down on the type of data set — time series, multiple frequency distributions, trends, geographical, outcomes, x-y etc — and concretely recommends appropriate visual techniques for each. Very useful structure!

One thing from the book

Inferring probability is difficult from pure numbers! However, we humans have an innate sense of probability when mapped to our “fear domain”. Visuals, like below, work so much better. Especially, on negative probability — imagine telling someone the lighter blue boxes have mines and which one would be the best minesweeper board to win!

Similar
“Beauty of Data Visualization” is a funny and good TED talk.