Problem: Regex extract only returns the first match
The REGEXEXTRACT function will extract the first matching substrings according to a regular expression, not all the substrings.
Solution: Use regex replace function
The REGEXREPLACE function can be used as a workaround to extract all substrings according to a regular expression. By replacing the string with the capture group instead of other text, we can extract all the substrings that match the regular expression.
REGEXREPLACE(A5,".?((?i)coffee)|.", "$1 ")
Note that in the regex above, the (?i) is for case insensitive
Problem: Match entire word, not part of a word
I wanted to match for the word "tea", but not words that include "tea" in it like "tears"
Solution: Use word boundaries
The regex for word boundaries is \b