Only Action Counts

The author contrasts a long list of preparation and performance-adjacent behaviors with actually doing the task. They emphasize that even imperfect or partial attempts count as real work, whereas planning and consuming content do not. The piece ends with a prompt to stop stalling and get back to work.

Key Points

• Most preparatory or adjacent activities (planning, researching, talking, buying tools) are not the same as doing the work.
• Consuming content about the task (podcasts, tutorials, threads) does not equal action.
• Perfectionism and elaborate systems can become procrastination disguised as productivity.
• Imperfect action—failing, doing it badly, doing a small part—still counts as real progress.
• The takeaway is to stop stalling and start doing the thing now.

Sentiment

The community is broadly supportive of the core message about overcoming procrastination through imperfect action. The tone is more 'yes, and...' than hostile disagreement — even those who push back generally agree with the spirit while adding caveats about the value of thoughtful preparation in professional settings. The main criticism targets the article's originality rather than its substance.

In Agreement

• Doing it badly is still doing it — imperfect first attempts unlock clarity and momentum that planning alone never provides
• Analysis paralysis is pervasive in tech: teams that over-plan for months often disband without shipping, while scrappy teams that start building deliver real products
• The article's self-awareness about blogging not being 'doing the thing' resonated, with commenters noting the meta-irony of discussing productivity on HN instead of working
• People with ADHD connected deeply with the message, recognizing the infinite preparation loop as a personal struggle
• Prototyping is cheap and educational — even failed attempts teach more about requirements than months of abstract planning

Opposed

• Planning and design ARE part of doing the thing — skipping them in software engineering causes compounding problems that cost far more to fix later
• The advice is tailored to solopreneurs and doesn't translate to team settings where coordination, code review, tickets, and meetings are necessary parts of collective execution
• In the GenAI era, 'just do it badly' has become trivially easy, making deliberate counterweights like design docs and review processes more important, not less
• The article is strikingly similar to an earlier StrangestLoop essay, with some calling it derivative or a rip-off rather than original thought
• Some important endeavors are too costly to 'just do' — thinking carefully before acting is sometimes more valuable than the action itself