Opportunity Brief — 2026-04-05 23:22

Name

Pocket Search

One-Line Wedge

Good-enough site search for teams too small for Algolia pricing and too sane for Elasticsearch theater.

Problem

Small sites, stores, and docs portals keep running into the same wall:

  • users expect typo tolerance and decent relevance
  • existing search SaaS is priced for companies bigger than them
  • the self-hosted options are either overkill or annoying to operate

This creates a stupid market gap: search is important enough to matter, but not important enough to justify a whole infra story.

Why Now

In 2026, small teams are cutting tool sprawl hard. The market mood is not “give me a better platform”. It is “stop making me buy enterprise software for basic UX”.

Search sits right in that zone. It is essential, visible, and frequently overpriced relative to value for small installs.

MVP

Build only this:

  • sitemap / URL crawler ingest
  • SQLite FTS5 index
  • typo-tolerant fallback search
  • synonyms config file
  • embeddable search widget
  • tiny dashboard for reindex + query analytics
  • JSON API for custom frontends

Brutal Scope Cut

Do NOT build in v1:

  • vector search
  • AI answers
  • semantic reranking
  • multi-tenant billing
  • team permissions circus
  • Elasticsearch compatibility layers
  • enterprise connectors

If the core search UX is not excellent on a tiny dataset, nothing else matters.

Why Open Source Wins

This category is vulnerable because the buyer already knows the problem should be cheaper. Nobody feels proud paying a recurring bill for basic site search.

Open source wins here by offering:

  • predictable cost
  • self-hosting for privacy and control
  • enough features for 80% of small use cases
  • easy embedding into static sites, docs, and simple stores

Who Buys / Uses It

  • docs site owners
  • bootstrapped SaaS teams
  • WooCommerce stores
  • self-hosted hobbyists
  • dev tool makers
  • content sites with search-heavy navigation

What It Replaces

  • Algolia on tiny installs
  • hosted Typesense / Meilisearch for low-volume usage
  • overpriced search plugins
  • brittle DIY search hacks

Why This Could Spread

This is easy to explain in one sentence:

“It gives you decent search without an enterprise bill.”

That is a strong word-of-mouth wedge in self-hosted, indie-hacker, and docs-tool circles.

Suggested Stack

  • Node.js + TypeScript
  • SQLite FTS5
  • Fastify or Express
  • Web Components for embeddable UI
  • simple crawler using sitemap + fetch
  • optional static deploy helper / Docker image

Go-To-Market Angle

Lead with:

  • Algolia alternative for small sites
  • self-hosted search for docs and stores
  • cheap typo-tolerant search

Ship a one-command Docker setup and a demo docs site. If install takes more than 10 minutes, the wedge gets weaker.

Risks

  1. Search quality may disappoint
    • Mitigation: focus on small datasets, synonyms, typo tolerance, strong defaults
  2. Category already has open-source players
    • Mitigation: compete on simplicity + embed UX + zero-bullshit setup, not raw engine sophistication
  3. Teams may still choose hosted for convenience
    • Mitigation: provide optional lightweight managed mode later, but only after the self-hosted wedge proves real

Scores

  • Severity: 4/5 — broken search hurts conversion and trust fast
  • Frequency: 4/5 — repeated across self-hosted/small-team contexts
  • Solvability: 5/5 — MVP is very buildable with boring tech
  • OSS Displacement: 5/5 — directly targets paid recurring search spend
  • Distribution: 4/5 — clear problem, clear buyer, clear comparison page potential

Total: 22/25

Status

🔥 shortlisted

Decision

Worth keeping near the top of the list because it is painful, specific, explainable, and realistically shippable.