the first version only needs to teach you.
founders often ask an early product to do too much. it must impress investors, silence competitors, delight every customer, justify the last six months, and look polished enough to protect the founder from criticism.
that is a heavy assignment for version one.
the first version has a simpler job. it should answer a question that matters.
will someone use this without being chased? where do they hesitate? what do they misunderstand? which part creates relief? what did the team believe that reality immediately disproved?
you cannot learn those things from another internal meeting.
polish is useful after direction is earned. before that, polish can become expensive camouflage. a beautiful workflow may hide the fact that the customer does not care. a complete feature set may delay the one interaction that would expose the weak assumption.
choose the riskiest belief and build the shortest honest test.
do not remove so much that the product becomes meaningless. a weak test creates weak evidence. give the customer enough value to make a real choice, then watch the choice without explaining it away.
the most valuable result may be rejection. clear rejection saves months of building around a fantasy. confusion is also useful when it points to the exact place where your thinking and the customer's world separate.
version one should leave evidence behind. recordings. questions. repeated behavior. abandoned steps. willingness to return. willingness to pay.
those traces are more valuable than compliments from people who never had to choose.
the founder's pride is not part of the measurement.
build something real enough to be judged and small enough to be changed. then let what you learn earn the second version.
