As I mentioned in the Khudawadi post yesterday, merchants simplified Sharada, presumably to let them write more quickly. In addition to dropping the vowel diacritics, they also used the same character for aspirated (“breathy”) and non-aspirated consonants, and dropped punctuation (including spaces between words). This happened in both the Khudawadi and Punjabi (AKA Multani) versions of Landa.
As late as 1916 AD, a Westerner noted that Landa was “current all over the Punjab, and is especially used by shop-keepers”. Wikipedia suggests that part of the reason to use Landa was as a code language, to keep commercial interests private. If so, that would be another reason why Landa was not particularly well-standardized: it is a language for communicating with oneself, not interpersonally, similar to Naxi Dongba and Naxi Geba.