**Hi Shrivallabha,**

You are too kind! But there is no need for any "hat doffing"... we are all in this to learn something new... I have certainly learned a ton going through these challenges.

(I don't typically care too much about formula efficiency either, probably because of the small data sets I typically work with. However, it is good to learn about them. Thanks for sharing the link to the decisionmodels site.)

In some respects, I am not sure if the drive to shorter formulas is a good thing from a learning perspective since it probably discourages others from submitting new / other ideas. In case anyone is holding back on ideas, here is my request to readers to submit ideas for solving the challenges, regardless of the length of such solutions.

Hi Jeff,
By the way, one thing that should probably get included in the writeups is a limitation of a given approach. For example, the following formula would not return the desired results on a string such as "First12345Second6789Third4567". While this was not explicitly called out as a requirement in the challenge description, it is good to understand the limits of a given approach.

=MID(A1,MATCH(1,-LEFT(TEXT(MID(A1,ROW(OFFSET(A$1,,,LEN(A1))),2),"A"),1))+1,6^6)

to get the desired results, one will need to tweak it as follows:

=MID(A1,MATCH(0,0*LEFT(TEXT(MID(A1,ROW(OFFSET(A$1,,,LEN(A1))),2),"A"),1),)+1,6^6)

(Of course, this is because of the way MATCH works when the last parameter is omitted.)

Perhaps you can devote a whole article to describe the limits of the various formulas! Just ensuring you have a steady stream of topics to write about!

Regards,

Sajan.