I think there won't be never a solution which could fetch each kind of changes in the business logic and/or within the data and their quality. There is always the need that the results of the reports are evaluated by capable and interested people. Without this there will be always a risk that the output is wrong. And this isn't only related to Qlik else to all tools.
The best approach to minimize the efforts on the needed adjustments by changes is to use/create well-designed business-processes. This means the processes don't require much hard-coded stuff, well maintained core-data and so on - and are quite flexible so that new / changed / not more active products, article-groups, stores, sales-reps and so on don't need mandatory a touch.
Of course there will be always some exceptions which require some adjustments but it makes a big difference if the main-parts works smoothly and only a few things need to be changed or if it's reversed. The real nightmare is on the business-side and not on the technically point of view - whereby it becomes a nightmare on the technically-side, too …
Beside from this there is a technically method to keep more control to what is used where and this is to maintain all fields, measures, variables, comments and so on externally in a database and each application looped through it and load it's belonging parts. Everything would be on a single place and it's quite easy readable and changeable. But the efforts to implement such an approach is rather enormous.
In regard to your example it might be helpful to add some in between-layer so that you don't query directly specific product categories else something like a "focus-category" which contained the monthly/regular changing product categories and/or with this layer are extra [Sales] or [Flag] fields created - means also to transfer as much as possible of the logic from the UI into the datamodel.
- Marcus
