Tarpit ideas

A tarpit idea looks attractive, keeps attracting new founders, and rarely succeeds. The market taught those earlier founders something. If your insight is not specifically why they failed and why now is different, you are walking into the same trap.

"No one has built this" is often a bad sign, not a good one. Sometimes it means the idea is hard for structural reasons.

The big idea

A tarpit is a pit of tar that looks like solid ground. Animals walk in because it seems fine. Founders walk into social networks and "Uber for X" because the apparent market is huge and the pitch writes itself.

The analogy breaks when you do have a wedge: a specific community, a distribution advantage, or a new constraint that prior attempts did not have. Then the graveyard is research, not destiny.

Walking through it

Classic tarpit shapes:

• Social networks / "the next Facebook" without a wedge into a specific community
• Yet another dating app with no distribution advantage
• "Uber for X" where X has low frequency, high trust barriers, or thin margins
• Photo sharing, note-taking, or generic productivity without a sharp user and a sharp job-to-be-done
• Marketplaces that need both supply and demand on day one with no unfair way to seed either side

Tarpits share traits: huge apparent TAM, low differentiation, and a graveyard of prior attempts that looked identical on a pitch deck.

"AI customer support" is a tarpit shape if your only insight is "LLMs can answer questions." It becomes more interesting if you have access to a vertical's ticket corpus, evals that beat generic bots on that vertical's policies, and a sales motion into that vertical. The model is the commodity. The workflow and data are the company.

Investigate failed predecessors before celebrating empty space. Then write down the why now that those attempts did not have.

Common mistakes

• Treating empty space as opportunity. Sometimes it is a graveyard.
• Borrowing a known success's shape. "Uber for X" copies the logo, not the hard part (liquidity, trust, frequency).
• A huge TAM slide with no wedge. Apparent market size is how tarpits advertise.
• Skipping the autopsy. If you cannot say why prior attempts died, you do not have an insight.

Check yourself

You want to build a new note-taking app because "Notion is cluttered." Is that enough?

No. Note-taking is a tarpit unless you have a sharp user, a sharp job, and a reason prior apps failed.

When is a tarpit shape still worth attempting?

When you can name why prior attempts failed and have a specific why-now or distribution wedge they lacked.

Key takeaways

• Tarpit ideas look huge, stay crowded, and rarely work.
• "No one has built this" can mean structural failure, not green field.
• The model or the clone is usually the commodity. Workflow, data, and distribution are the company.
• Autopsy prior attempts before you write the deck.

Further learning

A YC talk on common idea mistakes — starting with solutions, waiting for perfection, ignoring morphing:

Next: founder-market fit.