If science is a relay race, when you don’t publish, the baton goes to someone else. If science is a quilt, when you don’t publish, someone else’s square will go in the place saved for yours. If science is a community garden, when you don’t publish, someone else will till your soil. If science is development, when you don’t publish, another way will be found around that stage. If science is a great saga, if you don’t publish, your bit in it will be left out.
When to publish is sometimes a matter of taste. Do you tell the story bit by bit, or wait until it is complete? Do you do one more experiment? Is it decided by years or by figures? My preference is to see a story brought to a fairly complete level, with several experiments or studies that answer a question quite completely.
Another important factor in when to publish depends on whose work it is. If a student completed a story to a certain point and then moved on, there will be a bias towards publishing that story without too much more. The story matters, but so do the authors.
Since science builds upon itself, a study done in one year will not fit years later without revision. Other work will have taken place that at best means your study must be reinterpreted and at worst means your story is now irrelevant. Publishing promptly when the research is done will avoid these problems.
It was one of the surprises when I became head of a group that others might not be as anxious to get their work out as I was. Or if they were as anxious, writing blocks, or failures of confidence kept the work under wraps. I have lists of papers, now probably irrelevant, that never got written, never fulfilled our obligation to the taxpayers that funded the work, never helped someone’s career advance. Try to keep this from being your story. Finish your work and write it up. It will never be truly done, but putting it out there will let others join in building the next story.