Person/Jordan Lord/Blog/26/04/07/3 constraints before I build anything

• 3 constraints before I build anything

  • Author: Person/Jordan Lord
  • Published: 2026-04-07 (on jordanlord.co.uk)

  • Summary

    • Person/Jordan Lord argues constraints enable creativity: they shrink the search space so you can find sharper solutions. After ~10 years building, he uses three gates before starting anything; if an idea fails any of them, he does not build it.

  • The three constraints

    • 1. One page or it doesn’t get built — Write a single one-pager per idea: north star, non‑negotiable, precise, ambitious, lean. Use it for investors, collaborators, and yourself. If something matters in a dispute, it belongs in the one-pager or the one-pager changes; otherwise it is not the fight. If you cannot fill one page without fluff, you are not ready—research, plan, prototype, rewrite. If it needs more than one page, it is too complex to build (yet).
    • 2. The core tech must be separable from the product — Ship a core capability (method, skill, tool, or extractable library) that supports the product but survives without it—reusable IP that compounds while products pivot. Examples in the post: git for Linux, HCL for HashiCorp, Kubernetes at Google; yours might be smaller (a library or a methodology). If the idea does not enable core tech, it is not high enough leverage.
    • 3. One defining constraint must shape the product — Pick one obvious constraint users see constantly; it is the product’s identity and limits feature creep (e.g. Minecraft’s blocks, IKEA’s flat-pack). A good constraint reduces decision space so design can “fall out” of it; weak or missing constraints yield bloated “do everything” products. That constraint should also be front and centre in the one-pager.

  • Closing rule (from post)

    • If it fails any of the three, do not build it.