This kinda feels too little too late to me. I wrote it as a UDF five years ago and already includes most of what everyone's asking for here ie case sensitivity and a few things they aren't like Nth match. Splitting the look-in range and the return value range into two arguments is a bit of an interesting approach, and I am wondering if it'll work when you throw two or higher dimensional references at it or maybe even row to column pivoting. However the real questions are where are the new features that'll keep Excel alive and kicking in the data space for the next generations to come? REGEXLOOKUP(), tree, orbital, or hex grid data tabs, relational joins, histograms, non-linear regression lines, high-dim clustering analysis, CRYPTORAND(), density plots, B2B data pipelining, ... the list goes on. The whole selling point of Excel is that it packages these tools into an easyish to use and experiment with environment where your average person can poke around with it in an interactive way; however when other tools like Python, R, and the various web based visualization suites are easier to use then Excel, it loses some of its curb-appeal.
Microsoft