Persistence Is the Only Variable That Matters
There is a well-worn list of things people credit for success in software. Intelligence. Taste. The right tools at the right moment. Being early to the right technology. Knowing the right people. Having a strong team.
None of these are nothing. But none of them are the variable that actually separates teams that ship from teams that don’t.
Persistence is.
This is not a motivational claim. It is an observation about how software actually gets made.
Building something real takes longer than you expect. It encounters problems you didn’t anticipate. It requires decisions to be made with incomplete information. It involves stretches where nothing is working and the path forward is not obvious. These are not exceptional circumstances. They are the normal texture of building anything significant.
The teams that get through them are not necessarily the most talented. They are the ones that keep going. They are the ones that, when the approach doesn’t work, try a different approach. When the feature takes three times as long as estimated, ship it anyway. When the launch is disappointing, iterate. When the client changes the requirements, absorb the change and keep moving. Persistence is not stubbornness - it is the willingness to keep working on the problem long after the initial enthusiasm has worn off and the difficulty has become clear.
Talent is visible early and fails late. Persistence is invisible early and wins late.
The talented developer who gets bored and moves on to the next interesting problem leaves behind unfinished work. The less talented developer who stays with the problem until it’s solved has something to show. Over a long enough timeline, completion beats brilliance.
This is uncomfortable because the industry is built around signals of intelligence - the hard interview, the impressive technical background, the strong opinions about architecture. These signals are real. Intelligence matters. But intelligence without persistence is a pattern of impressive starts and disappointing finishes. Persistence without exceptional intelligence still produces shipped software, maintained relationships, and compounding institutional knowledge.
The organizational version of this is equally important.
Products that succeed are rarely the first version of the idea. They are the version that survived long enough to find the right form. The initial launch is not the product - it is the beginning of the process of discovering what the product should actually be. Teams that treat the first launch as the endpoint don’t make it to the second iteration. Teams that treat every launch as the starting point for the next one build things that work.
This requires persistence at the organizational level. The willingness to absorb negative feedback without abandoning the direction. The discipline to fix what’s broken instead of replacing it with something new. The patience to compound small improvements over time rather than bet everything on a single reinvention.
Nothing takes the place of it. Not talent. Not genius. Not education. Not timing. Not resources.
The work that gets done is the work that someone kept doing. That’s the whole principle.
Keep going.