How to match cell values to columns based on their inital characters

[Strixx]

Hello,

I am parsing information off a content curation tool into Excel for further processing in MS Power BI.
I am parsing titles, given tags, a summary and the time stamp of articles. The parsed tags causing my trouble.

I use a bi-hierachicial taxonomy for tagging my content with one topic tag and one describing tag and a company tag with no leading numeration. Something like this:
1 Topic
1.1 Topic description
1.2 Topic description
1.3 …
…
2 Topic
2.1 Topic desciption
2.2 …
…
Company AB
Company CD
Company EF

Unfortunately my content curation tool has no feature to display the related tags in alphabetical order. So I get rows like this when parsing the tags column, where the company tag could be anywhere:

1 topic, 1.5 topic description, Company AB
5.3 topic description, Company YZ, 5 topic
…

Out of this single column i want to create columns three columns:

TIER 1 Tag (#) | TIER 2 Tag (#.#) | Company (abc) | title | comment | timestamp

1 topic | 1.5 description | Company AB | … | … | …
5 topic | 5.3 description | Company XY
…

So I am looking for a way to match the tags row by row to the right column by their initial characters # / #.# / abc, since a “simple” separation by text to column will not let me filter properly by company.

Every help is much appreciated!
Thanks in advance.