[fundamental]

In an ideal world you start off with loose highly manual code/processes when validating an idea and as you get closer to a publication components are restructured/rewritten to make it easy to replicate results.

Other researchers care about the final code which is used to generate the results. So, in my book it's ok if there's a large gap between the code that lead to the initial idea and the code which was used to show the idea in practice (i.e. the code used to generate all graphs/tables within a submitted publication).