The Hagakure #63: Sunday Round-Up

Why estimates are (almost) always wrong, keeping others (and yourself) accountable, and feedback as natural as running water.

I’m an introvert who is not anti-social. And I have a tendency to go overboard sometimes in whatever I do.

Sure enough, this past week I dialed (way) up the number of new people I met both online and in person. Result: a full heart, and a depleted battery. Totally worth it. So many incredible people out there. 🙌

So, I’m using the weekend to relax, enjoy the nice weather in Berlin, and to read (savor) a really good book which I will be sharing some bits about on the Hagakure soon.

Finally, thanks to everyone who smashed that ❤️ button last week. Loud and clear. I will continue to share weekly my top 3 LinkedIn posts of the previous week, right to your inbox, or Substack app. 📥

Have a great start to your week, and stay curious (like my cat below). 🧐

Picture of a cat doing what cats do when they think they lost something.

Ever wondered why delivery estimates are (almost) always wrong?

Here’s some perspective.

⏰ Total time = 👩‍💻 Active Time + 🥱 Wait Time

Estimating when something will be done always *implies* both active time (when it is actually being worked on, aka “touch time”) AND wait time (when work is waiting for something or someone).

But when an estimate is given, we are biased to think of active time only. Without distractions, handovers, dependencies, holidays, etc. It's only human. Happy paths are alluring.

The problem is: "wait time" is virtually 𝘪𝘮𝘱𝘰𝘴𝘴𝘪𝘣𝘭𝘦 to predict. Read that again. Blew my mind, too. 🤯

Example: how can you possibly know in advance how long you'll be waiting for code review? It depends on multiple, complex, future interactions.

And the more work in progress there is, the more dependencies, the more multitasking, the more context switching... the more "wait time" you get.

What you end up with is something few teams actually make visible: 𝘸𝘢𝘺 𝘮𝘰𝘳𝘦 wait time than active time.

Is it any wonder so many businesses lose trust in engineering? Or that engineers lose trust in the business—often refusing to provide any estimates at all?

The model most companies use for product development is fundamentally broken, despite the proliferation of "agile" frameworks, masters, coaches, and consultants.

Change is hard but it's always one step at a time:

✅ 𝗦𝘁𝗼𝗽 𝘀𝘁𝗮𝗿𝘁𝗶𝗻𝗴, 𝘀𝘁𝗮𝗿𝘁 𝗳𝗶𝗻𝗶𝘀𝗵𝗶𝗻𝗴.

Keep your FOMO in check. Resist starting new work when more valuable work is yet to be finished—and shipped. It's counter-intuitive. But it's also math (see Little’s Law).

✅ 𝗙𝗶𝗻𝗱 𝘁𝗵𝗲 𝗯𝗼𝘁𝘁𝗹𝗲𝗻𝗲𝗰𝗸.

What's the biggest culprit of creating wait time in your development process? Get together as a team and find creative ways to measurably reduce it. Then rinse and repeat. (Hint: pair programming can obviate the need for code review.)

✅ 𝗕𝗿𝗶𝗻𝗴 𝗣𝗿𝗼𝗱𝘂𝗰𝘁 & 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 (𝘁𝗿𝘂𝗹𝘆) 𝘁𝗼𝗴𝗲𝘁𝗵𝗲𝗿.

Working on one thing at a time only makes sense if you consistently work on the most valuable bet. The best teams explore what that is, together, with the freshest information possible. They define success and experiment their way to it. They ship often and learn often.

Above all: start looking at your team’s development process as a whole system. And embrace the fact that estimates are not predictions.

• What's your experience with estimates and multitasking?

• What has worked and what has not worked?

---

I lost count how many times I have heard managers complain that some of their people are not “accountable”.

When I see this pattern pop up in a coaching session, I know how the story usually ends:

With my client realizing they’re not doing their part. And that it changes everything.

“Jake, we need the report by next week.”

“Yep, cool, you got it.”

Next week goes by and the report is not delivered. Disappointment. Do this enough times and you start believing Jake “isn’t accountable.” Trust erodes, and next thing you know you’re contemplating letting him go.

Here’s the thing: agreement is worthless without commitment.

An agreement is a fuzzy statement of intent. A commitment is specific. Who’s going to do what by when?

You have to move from agreement to commitment and the way you do that is by making a request, and eliciting a promise.

“Jake, we need this report done by next week, alright?”

“Yep, cool, you got it.”

“When can I expect it and how will I know?”

“Ummm, by next Tuesday EOD. I’ll share the doc over email and ping you on Slack.”

“Sounds good. Can I have your promise on that?”

“Yes, you can.”

When a promise is made, breaking it means your integrity is compromised. How can I trust you if you don’t keep your promises?

And if sh*t happens and Jake can’t honor the commitment? He can always go back to his manager before the committed deadline, explain what happened and why he can’t make it, and then agree together on a re-commitment that works for both.

If this sounds micromanage-y, think again. What you are actually doing is:

• Giving the other person an opportunity to say “No, I actually can’t commit to this. But here’s what I can do instead…”

• Creating the opening for a conversation about “having too much on my plate”

• Leading by example in the organization on how to establish true accountability.

Trust is about making promises and keeping them, consistently, over time.

Learn how to go the extra mile from agreement to commitment. Your team will thank you for it, and you won’t be wondering so much why on earth people can’t be held “accountable.”

---

5 minutes of feedback. Every 1-on-1.

That’s all it takes.

In every such interaction, reserve the last 5 minutes to give and receive feedback.

• About how this particular interaction went.

• About how we’re working together.

• About anything that is nagging me.

• About something we should do more of.

“What happened in this that you liked? And what happened that you wish were different?”

By turning this into a habit, you remove the sting out of it, while keeping all the benefits. Feedback becomes as natural as running water.

As Jennifer Garvey Berger writes in her wonderful book Simple Habits for Complex Times:

“𝘍𝘦𝘦𝘥𝘣𝘢𝘤𝘬 𝘪𝘴 𝘵𝘩𝘦 𝘭𝘪𝘧𝘦𝘣𝘭𝘰𝘰𝘥 𝘰𝘧 𝘤𝘩𝘢𝘯𝘨𝘦 𝘪𝘯 𝘢 𝘤𝘰𝘮𝘱𝘭𝘦𝘹 𝘸𝘰𝘳𝘭𝘥, 𝘸𝘩𝘦𝘵𝘩𝘦𝘳 𝘪𝘵’𝘴 𝘵𝘩𝘦 𝘬𝘪𝘯𝘥 𝘸𝘦 𝘨𝘪𝘷𝘦 𝘰𝘯𝘦 𝘢𝘯𝘰𝘵𝘩𝘦𝘳 𝘰𝘳 𝘵𝘩𝘦 𝘬𝘪𝘯𝘥 𝘵𝘩𝘢𝘵 𝘪𝘴 𝘢𝘷𝘢𝘪𝘭𝘢𝘣𝘭𝘦 𝘪𝘯𝘴𝘪𝘥𝘦 𝘢 𝘴𝘺𝘴𝘵𝘦𝘮 𝘪𝘧 𝘺𝘰𝘶 𝘨𝘰 𝘢𝘯𝘥 𝘭𝘰𝘰𝘬 𝘧𝘰𝘳 𝘪𝘵.”

Take it, acknowledge it, and then if you accept it, declare what action you’ll take and take it. It makes others feel they can actually change the way you interact with them.

I can’t think of a much better gift than changing ourselves in response to another.

And all it takes is 5 minutes. Every 1-on-1.

---

Thanks for reading. If you enjoyed this post, please consider hitting the ❤️ button, subscribing for future issues on your inbox, and sharing it using the button below.

Until next week, have a good one! 🙏