The story of rediscovering "the importance of reporting, contacting, and consulting" after having AI write code

The story of rediscovering "the importance of reporting, contacting, and consulting" after having AI write code

Of the three elements of "Horenso" (report, inform, consult), I believe the two that truly matter are reporting that does not demand a decision, and consulting that does demand a decision. Above all, reporting bad news as quickly as possible is what moves a project forward.
2026.07.07

This page has been translated by machine translation. View original

Introduction

Recently, I've had more opportunities to use AI coding tools like Claude Code and Codex. What surprised me after using them was that these AIs don't really keep you in the loop.

Before you even decide on a direction, they just barrel ahead with implementation on their own. When problems arise during testing, they move on without reporting them. As a result, you sometimes end up with something completely off the mark. It felt oddly human in a way. And it reminded me once again that regular reporting and communication matters, whether you're dealing with an AI or a human.

So in this article, I'll briefly share my everyday thoughts on reporting and communication.

The Two Things That Matter: Reporting and Consulting

We often talk about the three elements of workplace communication (reporting, informing, and consulting), but the two I truly consider important are reporting and consulting. These two can be distinguished by whether or not you're asking the other person to make a judgment. Reporting does not call for a judgment; consulting does.

Good reporting is reporting that conveys a coherent story.

"What has happened up until now" → "What the current situation is" → "What you intend to do going forward"

It's important that this line connects seamlessly from past to future. The essence of reporting, as I see it, is sharing the same picture with the other person and conveying exactly where you currently stand on that line.

Timing also matters. The most critical moment is reporting before you start moving. Simply saying "I'm heading in this direction" gives someone the chance to catch it early if you're off course. Rework becomes costly once you've already moved forward. The AI example at the beginning was exactly this. Because it starts running without reporting, the output ends up completely off the mark.

Consulting, on the other hand, is a somewhat weightier act, because it asks the other person to make a judgment.

What I want to avoid here is dumping the decision entirely on someone else. Simply tossing out "What should we do?" only adds to the other person's plate and rarely moves things forward. I try to frame it as: "For these reasons, I'm thinking of doing this — does that work for you?" You're presenting your own reasoning and conclusion, while ultimately leaving the final call to the other person.

In both reporting and consulting, the one thing I want to be thorough about is reducing the cognitive load on the other person. Taking care in how you communicate so the other person can understand as quickly as possible and make decisions as quickly as possible — that alone makes a remarkable difference in how work progresses.

The Worse the News, the Faster You Should Move

There's one more thing I really want to convey. The worse the news, the sooner you should raise it.

Reporting also serves to protect yourself. If you've reported something, even if you're heading in the wrong direction, someone can stop you. On the other hand, if you press on without reporting and a problem surfaces later, you'll be asked why you made that call entirely on your own.

It's embarrassing to admit, but when I was just starting out in my career, I wasn't able to do this. When problems arose from my own mistakes or misunderstandings, I felt the urge to quietly cover them up and sweep them under the rug. Looking back now, I think that was truly the wrong approach. When you hide things, you can't reflect on the failure, and it does nothing to prevent it from happening again. Problems should be brought to light as problems, and faced together as a team.

When it comes to bad news, you naturally brace yourself — worried you'll be scolded or blamed. But what matters most is that the project moves forward. The sooner bad news is shared, the sooner everyone can focus their energy on getting back on track.

By the way, this is a personal tip, but staying on good terms with your team members on a daily basis significantly lowers the psychological barrier to delivering bad news. It's important to be honest with yourself about your own personality and tendencies, and to be able to manage your own mindset.

Closing

"I really don't want to get in trouble." The moment that thought crosses your mind is exactly when you should start moving your hands and writing up a report. What matters more than protecting your own image is keeping the project moving forward. Don't hide things — face problems together as a team. In the end, that makes things easier for both you and your team.

The worse the news, the faster you should move. That is the phrase I value most when it comes to communication at work. And the more you overcome challenges together that way, the stronger your team's bonds will become.


クラスメソッドのエンジニアと1on1で話してみませんか?

選考に関係のないカジュアルな面談です。
「技術スタックや開発環境について詳しく知りたい」「実際のプロジェクト事例を聞きたい」「リモートワークや評価制度について確認したい」など、気になることを直接エンジニアに質問できます。
カジュアル面談に申し込む

Share this article