The Tiny Core Principle
Every enduring product has one tiny thing that is a superpower — find it, protect it, and stop bolting on features.
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 3
- Confidence
- 90%
Schoening's theory of what makes products great: it is never the suite of features or even the best design or engineering, but one tiny core that is so exceptionally good it becomes a superpower. Everything else flows from that core. The corresponding failure mode is the feature death-spiral — believing that adding just one more thing will finally make the product great. The discipline is to identify the core, resist expanding surface area, and even push back on features that dilute it.
Origin
Schoening's own synthesis from contributing to GitHub, Heroku, and Notion, illustrated with outside examples (Dropbox, the iPhone, Snapchat).
Core principles
- 01Great products have one tiny core that is exceptionally good
- 02The core is part luck (you stumbled on it) and part market validation
- 03Feature count is a vanity metric, like lines of code or tokens consumed
- 04Adding 'one more thing' to fix a weak product never works
- 05Fewer, really good features where the combinatorics let you do everything beats a big suite
How to run it
- 1
Identify the one tiny superpower
Name the single smallest capability that makes the product intoxicating. Schoening's examples: GitHub = the pull request; Heroku = 'git push heroku master'; Dropbox = a menu-bar icon that synced so reliably it beat the OS at detecting connectivity; Notion = blocks and slash commands.
Pro tip Test it by asking what tiny thing users keep coming back to no matter how bad everything else is.
Watch out If you can't point to a tiny core, the product may not have one yet — adding features won't create it.
- 2
Protect the core from surface-area creep
When teams try to expand the product's surface area around the core, push back. Dropbox spent years trying to grow beyond sync; Schoening's instinct was 'no, no, push it back — this is the only job I want from you.'
Pro tip Get out of the way: let the core do its one job perfectly rather than surrounding it with adjacent jobs.
Watch out Growing surface area to look more valuable usually degrades the very thing that made the product loved.
- 3
Don't give up on a strong core that tests badly
A genuinely good core can still fall flat in early user studies because people don't yet know how to use it. Hold the core (roughly 80%) and relentlessly iterate the remaining 20% until it clicks.
Pro tip Separate 'the core idea is wrong' from 'the core idea is right but not yet legible to users' — only the former justifies abandoning it.
Watch out Killing a strong core because early users are confused is as costly as polishing a weak one forever.
In the wild
Dropbox's tiny core was a menu-bar icon so good at syncing that people used it as a proxy for whether they had internet — it detected connectivity better than the Mac itself. For years the company tried to increase surface area and Schoening kept thinking it should be pushed back to that one job.
→ The single 'my files are always there, get out of the way' job was what made the product great; expansion added little.
His 2014 startup poured effort into polishing the editing experience — markdown folding and features now in Obsidian — believing that was what mattered. Notion's first editor was terrible by comparison (all blocks, you couldn't even select between two blocks), yet it didn't matter because Notion had the right tiny core in blocks.
→ He spent years working diligently on the wrong thing; the polished editor never became a core superpower, and the startup didn't win.
Common mistakes
The 'one more feature' death spiral
Believing that if you just add one more thing the product will finally be great. Schoening says this never works and is one of the biggest pitfalls — a weak core is not fixed by additional features.
Chasing feature count as a metric
Feature count is the same silly metric as lines of code or tokens consumed. Optimizing for it produces a bloated suite instead of a sharp, exceptional core.
Is it for you?
Best for
Product leaders and founders deciding where to focus scarce build effort, especially when tempted to expand a product's surface area
Not ideal for
Platform or infrastructure plays where breadth and integration genuinely are the value, or very early exploration before any core has emerged
From the transcript
“if I really look at the the truly great products they all have one tiny core that is so”
“all the great products have something tiny that is a superpower”
“one of the biggest pitfalls is if you get into the loop of if I just add one more thing to the product it'll be…”
From the episode
Why cultivating agency matters more than cultivating skills in the AI era
Max Schoening (Head of Product, Notion)