Paper boat
Paper boats are usually folded by children for fun. They decide on a goal in their head, that is, let's say they want that paper boat to float on top of the water in the bathtub. That's why they ask people they think can fold a paper boat. They show them, teach them, and fold a paper boat or two.
Each of their folds might be a little different, and they begin to observe which type can stay on the surface of the water for how long. As it develops and more and more boats are folded, they realize the features, the folding techniques, that they need to pay more attention to if they want it to last a long time on top of the water.
A point may then be reached when they are satisfied with the water's ability to survive and start, for example, to generate waves in the bathtub with their hands. The first pair of types, even if they stayed on top of the water for a long time in wave-free weather, will not be sufficiently resistant to waves. Then they start producing types that are likely to be more resistant. Here, of course, there will be many types that either have a weaker ability to stay afloat or cannot withstand the waves. So it is up to the dear designer to consider which features of the boat he prefers.
If you think about it, this kind of experimentation is very typical for the discovery and development of new things. Ideally, every discovery would be a series of playful experiments during which, despite changing requirements, a new product is always created. As a result of this process, paper ships, products, and inventions are formed that will be suitable for certain purposes in certain circumstances.
If all small children's paper boat folding experiments were collected and all their related tests were documented, then a robust set of knowledge would be created, based on which anyone, knowing the requirements and based on the current knowledge base, would be able to select the most ideal paper boat.
However, the critical prerequisite for this is that each experiment and each condition and each test result must be recorded in a form that is later accessible to everyone. If this step is not taken, then we will deprive all the paper boat makers of the world from folding effective paper boats without wasting much time.
Even though such thorough documentation seems like a lot of administration, think about where humanity would be today if every single test result, every single test result of every physical experiment ever performed was accessible without falsification, even if the conclusion drawn from it was that the test was unsuccessful, since it did not meet the requirements set by them.
You never know who will have modified requirements, but which the given test perfectly meets.
Try to record the conditions of the measurement as thoroughly as possible if you find yourself in such a situation. Try to make as many measurements as possible with the same settings and the same configuration. As soon as you change a single parameter, the new measurements will no longer be perfectly comparable with your previous measurements.
Take the time and energy to develop those protocols, those parameters that you think can lead to results in the long term, and which can possibly clearly indicate the movement in the direction of your goal. Be careful when drawing conclusions, and if possible separate them from the experiments themselves and the results measured during the experiment.
It will be difficult to measure how many people these experimental results reach, but still, if it reaches only a very narrow group for whom this kind of experimental result is critical for moving forward, it was already worth putting so much work and energy into it.
Let this be today's lesson.

