I guess there are two problems: 1/ Usually through an evolutionary process the whole CMS structure for a paper becomes rather cumbersome with many patches and

I guess there are two problems: 1/ Usually through an evolutionary process the whole CMS structure for a paper becomes rather cumbersome with many patches and suboptimal services that nobody really knows about in details. Especially if the tech team rotates. And this leads to the next point. 2/ Bad administrative control of who put what on a page--sales, marketing, editors etc., and this results in a crappy experience with 50-100 different tracking garbage on every page. As a result, almost every paper becomes a pretty ugly monster that nobody wants to touch not to ruin everything. Any potential change could lead to the whole system collapse.