The AI Learning Hub Journal

Prototype, Something Friends Can Try, Real Product

Three very different things people call "an app"knowing which one you have is the difference between confidence and delusionPROTOTYPEexists to answer a questionruns on your machineallowed to breakone user: youfragile is fine here≈10×workFRIENDS CAN TRYlives on the internetsurvives ordinary usea little real data, frompeople who will forgive yousmall and real≈10×workREAL PRODUCTused by strangers whowill not forgive youstays up · protects datafixable at midnighta serious commitmentSTEP UP: PROTOTYPE → FRIENDSit leaves your laptop and goes onlineother people's data starts to arrivepasswords and keys come out of the codeSTEP UP: FRIENDS → STRANGERSyou know it broke without being tolda spare copy of the data existsa cap on what it can cost youa plan for when it fails while you sleepNONE OF THIS WORK IS VISIBLE ON THE SCREEN — WHICH IS WHY IT GETS SKIPPEDa finished-looking app can be zero percent of the way to a real product — looks and readiness are separate"THIS IS A WORKING PROTOTYPE — NOT READY FOR CUSTOMERS YET"that sentence costs nothing, buys credibility, and keeps your own estimate of the remaining work honestSAY WHICH ONE YOU HAVE — OUT LOUD, ESPECIALLY TO OTHER PEOPLEthe gap between a finished-looking demo and a product is where reputations are lost
A prototype, something friends can try, and a real product are three different things — each roughly ten times the work of the last.

Three Different Things

People say "I built an app" about three very different objects. Confusing them causes most of the pain in this field. A prototype exists to answer a question. It runs on your machine, it breaks if you look at it oddly, and that is fine, because its job is to inform a decision. Something friends can try lives on the internet, survives ordinary use, and holds a little real data from people who will forgive you. A real product is used by strangers who will not forgive you. It stays up, it protects data, it handles the odd cases, someone can fix it at midnight, and it does not quietly bankrupt you. Each step is roughly ten times the work of the last. Knowing which one you have is the difference between confidence and delusion.

  • Prototype: answers a question, allowed to be fragile, one user — you
  • Friends-can-try: online, survives normal use, small amount of real data
  • Real product: strangers, uptime, security, cost control, someone on call
  • Each stage is roughly ten times the work of the one before it

What Changes at Each Step

Moving between stages is not about polish. Prototype to friends-can-try means it lives somewhere other than your laptop. Other people's data appears, and your keys and passwords must stop sitting in the code. Friends to strangers asks more again. You need to know when it breaks without being told. You need a copy of the data if something is deleted, and a cap on what it can cost you. You need a plan for the day it goes wrong while you are asleep. None of that is visible in the interface, which is exactly why people skip it. The user-facing part of your app may be entirely finished while the product part has not been started. Those are separate axes of progress.

  • Prototype to friends: it leaves your laptop and real data arrives
  • Friends to strangers: monitoring, backups, cost caps, and a plan for failure
  • None of this work is visible in the interface, which is why it gets skipped
  • A finished-looking app can be zero percent of the way to a product

Say Which One You Have

Be precise about this out loud, especially with other people. Tell an investor, a client or your own team that you "have an app" when you have a prototype. You have just set an expectation you will pay for later, usually in public. The accurate version costs you nothing and buys credibility. "This is a working prototype. It proves the flow, and it is not ready for customers yet." Nobody sensible is disappointed by that sentence. It is a genuinely impressive thing to have made in a weekend. What damages trust is a demo that looked finished, followed by three months of silence. Naming the stage also keeps you honest with yourself about how much work remains. That is the number people most consistently underestimate.

  • Say "working prototype" out loud rather than letting people assume "product"
  • The gap between demo and product is where reputations are lost
  • Accurate framing costs nothing and protects you when timelines slip
  • Naming the stage keeps your own estimates honest too

Prefer slides, quizzes, and saved progress? Read this lesson in the library — free, no sign-up.