Microsoft Excel Tutorial: Solving a Text Comparison Problem: 3 AI Bots versus 3 Humans.
To download the workbook from today: Excel Three Bots Versus Three Gurus Episode 2672 Sample Files - MrExcel Publishing
The battle is on! In Episode 2672, we’re pitting three AI bots (Deep Seek, Copilot, and Beta Copilot with Python) against three (and a half) human Excel gurus (Rico, XLLambda, and Geert). Who will come up with the best solution for an Excel text comparison problem?
This challenge comes from Brian, who wanted to compare two text strings and extract the differences. Seems simple, right? Well, not quite! My initial formula had some issues, so I turned to AI for help. But the AI solutions… let’s just say, they were interesting. Some worked, some didn’t, and some made things even worse!
So, I turned to three Excel experts—and WOW, did they deliver! Rico crafted a brilliant formula, XLLambda pulled off an advanced Lambda function, and Geert worked his Power Query magic. Their solutions were faster, smarter, and more reliable than anything the AI bots came up with.
In this video, you’ll see:
The original problem and my not-so-great solution
How Deep Seek, Copilot, and Copilot with Python tackled the challenge (with mixed results)
Rico’s clever recursive Excel formula, XLLambda’s dual Lambda approach, and Geert’s Power Query approach
A final verdict: Who won—AI or human experts?
All formulas & Power Query code are in the workbook—try them out yourself! Download the workbook from:
Don’t forget to like, comment, and subscribe for more Excel challenges!
#Excel #AIvsHumans #Copilot #PowerQuery #ExcelFormula #ExcelChallenge
Buy Bill Jelen's latest Excel book: MrExcel 2024 Igniting Excel
This video answers these search terms:
Excel AI vs Human
Compare text strings in Excel
Excel formula to find differences in text
Copilot vs Deep Seek vs Python in Excel
Power Query text comparison Excel
Excel AI formula challenge
Recursive Lambda function Excel
Best way to compare text in Excel
Excel AI vs human experts
Advanced Excel formulas for text analysis
To download the workbook from today: Excel Three Bots Versus Three Gurus Episode 2672 Sample Files - MrExcel Publishing
The battle is on! In Episode 2672, we’re pitting three AI bots (Deep Seek, Copilot, and Beta Copilot with Python) against three (and a half) human Excel gurus (Rico, XLLambda, and Geert). Who will come up with the best solution for an Excel text comparison problem?
This challenge comes from Brian, who wanted to compare two text strings and extract the differences. Seems simple, right? Well, not quite! My initial formula had some issues, so I turned to AI for help. But the AI solutions… let’s just say, they were interesting. Some worked, some didn’t, and some made things even worse!
So, I turned to three Excel experts—and WOW, did they deliver! Rico crafted a brilliant formula, XLLambda pulled off an advanced Lambda function, and Geert worked his Power Query magic. Their solutions were faster, smarter, and more reliable than anything the AI bots came up with.
In this video, you’ll see:
The original problem and my not-so-great solution How Deep Seek, Copilot, and Copilot with Python tackled the challenge (with mixed results) Rico’s clever recursive Excel formula, XLLambda’s dual Lambda approach, and Geert’s Power Query approach A final verdict: Who won—AI or human experts? All formulas & Power Query code are in the workbook—try them out yourself! Download the workbook from: Don’t forget to like, comment, and subscribe for more Excel challenges!#Excel #AIvsHumans #Copilot #PowerQuery #ExcelFormula #ExcelChallenge
Buy Bill Jelen's latest Excel book: MrExcel 2024 Igniting Excel
This video answers these search terms:
Excel AI vs Human
Compare text strings in Excel
Excel formula to find differences in text
Copilot vs Deep Seek vs Python in Excel
Power Query text comparison Excel
Excel AI formula challenge
Recursive Lambda function Excel
Best way to compare text in Excel
Excel AI vs human experts
Advanced Excel formulas for text analysis
Transcript of the video:
Oh, we've got something really exciting today. Episode 2672 is three AI Bots versus three (and a half) human gurus.
Head-to-head. Alright, so in the last video it was Deep Seek versus Copilot in Excel. Today it's a whole new level.
Okay, we're going to start off with the original problem.
And my not-so-great solution. I realized it wasn't a great solution.
So I went out for help.
I went to Copilot, I went to Deep Seek, I went to Copilot in Python. Then I also sent it to my three friends.
Who were my three smartest Excel people, Rico, XLLambda, and Geert.
All of them came back with different answers. Some of them worked.
If you want to jump to one specific point.
I'll put the timestamps here where you'll find all of them. Let's go.
Hey, this is a great question from Brian.
“Is it possible to compare two text strings and return the differences in a new column”? He has two lists of car make and model.
One might say Toyota Avensis and the other might say Toyota Avensis 2D.
Is it possible to return the difference here? In other words, 2D?
He's thinking that Power Query might be a way to do it.
So I try to come up with some examples here. Including the one that Brian gave me.
But then “Mary Ellen Jelen” and “Mary Ellen Jelen, RN”.
Or “Nurse Mary Ellen” and “Mary Ellen”. “123 Jump Street” versus “Jump St”..
Okay, but here's the one that gets me. This is where my formula is not going to work.
So A-C-E is in this text, but it's not in consecutive letters.
Or “Bill Jelen” and “William Jelen”. What's the difference there?
Okay, so the formula that I'm using here. Is figure out which of these two is shorter.
So in this case, A2 is shorter. But in this case it's column B that's shorter.
And then actually take the longer one and remove all characters that are in the shorter one.
“Toyota Avenesis” from “Toyota Avenesis 2D”. And we're left with space 2D.
Here we are getting space rn.
So I'm really happy with the way that it's working here. I'm even happy with the way it's working here.
It gets the 123, takes out the “Jump St” and gives me the “reet”.
In my opinion, it's not working well here.
And that's when I decided to go bounce this off some experts. Let's take a look.
Just a note from Post-Production.
This is one where you probably really want to download the workbook.
The responses from the various AI bots are very wordy. And I am not going to read all of those.
But I pasted them all in this workbook. All of the examples.
All of the formulas from the AI Bots and our human experts are available.
The link is down in the YouTube description.
So I created this simple workbook that I'm going to upload to Deep Seek.
First thing I'm going to do is select the file and upload. And then the prompt.
So we'll try this prompt.
“This Excel workbook has two columns of text strings starting in A2 and B2. “In each row, the strings are similar.
“But one of the two strings has more text.
There might be extra characters in the beginning, the ending or in the middle.
Can you suggest a formula for C2 that will identify the extra characters in the longer string?
All right, so it can actually see the text that's in my workbook.
We'll copy this.
All right, so it didn't work here where the extra text is before.
So this feels very similar to the method that I came up with.
We will ask it if it can fix C6. Okay, this is great.
So it came up with an even more complicated formula.
I feel like. Okay, while it got the B, D, F, right.
It is worse. Like nurses in here.
1, 2, 3, we're missing the space in re. And I don't know how it came up with WAM.
I don't feel like Deep Seek is a significant improvement over what I had.
Alright, let's try the same task with Copilot. I've Auto Saved - turned that on.
And I'm going to ask Copilot.
We use the exact same prompt, that we sent to Deep Seek.
[ The formula does not work ] [ Typing a new prompt for Copilot ] All right, we'll copy that formula.
No, that's not it at all. I feel like we got a lot further away there.
All right, so here in this one It should be nurse and it's coming up with NUS.
I'm going to give this one a “Fail” for Copilot.
Okay, so we've straight formulas from Copilot, Deep Seek.
I think the actual answer here is going to be Python. Python probably has a way to deal with this.
So to get into Python, choose “Get deeper analysis results using Python”.
And then over here we have to say “Start advanced analysis”.
So in lands the data from the sheet. And then it automatically starts doing something.
Even though I haven't given it a prompt.
And it calculates something called a similarity ratio. So then my first prompt.
Identify the extra characters in the longer string.
And it does that. Let's take a look over here.
So there's the original text. The new text.
Similarity. And then the extra characters.
First thing I noticed is there's a “space 2D” there.
And then look at this. So the R in nurse.
Because there's an R in Mary Ellen. It's not seeing that as an extra character.
So even with Python, it's still not getting exactly what we want.
Alright, here's a beautiful formula from Rico.
Now Rico pointed out that with Bill Jelen and William Jelen. This has an extra character.
And this has an extra character. So Rico actually has two different formulas here.
One where we say remove B7 from A7. And another one where remove A7 from B7.
And he said, of course you can HSTACK these back together if you needed them.
But he suspects that in real life there's going to be one column that's better than the others.
More correct than the others. Recursive Lambda.
You can see it over here in Excel Labs. And I've copied it here.
It's pretty slick.
It substitutes the first instance in Text One, of the first character in Text Two.
And then recursively passes it back to the function.
With the leftmost character of Text Two dropped for the next iteration. What does that really mean?
Okay, so here's Text One. Here's Text Twi.
First character in Text Two is A. So we substitute A with nothing.
That A goes away. And we're left with B, C, D, E, F.
And then the function calls itself again. But this time stripping off the first character.
So we're left with C-E. The first character C.
Substitute C with nothing. We get B, D, E, F.
Call the thing again.
And you end up substituting E and we get that correct result of B, D, F. It's pretty slick.
And it gets the first instance only.
Which is preventing some of the problems that the AI bots were coming up with.
Although it does kind of do this weird thing like nurse is extra. But that “r” is found in “Mary Ellen”.
So this time we're calling the function 11 times.
And it's like right here when it's trying to remove the R. It should be removing that R from Mary Ellen.
But it's removing the first one. Same thing here.
Its removing this E instead of that E. But it's still the extra characters.
I think in principle, this is still really correct.
Okay, now this awesome formula from XL Lambda. And it's in his name< “XL Lambda”.
So there's two Lambdas in here using REDUCE and MAP.
He has two different sets here. Words in B that are not an A.
Or in A that are not in B. Or if you really need what I asked for.
The shorter one, subtracted from the longer one, we have that.
Interesting enough. That since he's working on a word level.
He removes jump from 123 Jump Street.
Removes from 123 St and it ends up with a 123 reet. So that's just kind of interesting.
Although I can see how that's happening.
And then he had an asterisk here for A, B, C, D, E, F, and A, C, E. He's like, Nope, that's not relevant.
The granulation level here is words.
If you really wanted to compare letters, that'd be a different scenario.
And he points out that he has a lot of other algorithms that could do words versus sentences.
The sky's the limit and it can be as complex as regex. There's no magical solution to cover them all.
But I think these are awesome formulas that cover the spirit of the original problem.
Let's whisper this. Like most formulas that XL Lambda sends me.
I can't understand what it's doing.
Let's see if Deep Seek can, This is really good. This is a great explanation.
I'll leave it there on the screen. Now if you remember.
This question was sent in by Brian.
And Brian thought that it was going to be a power query solution. And we have our solution here from Geert.
And it is Power Query. Geert actually has an awesome query right here.
That is showing the difference.
Extra words on the left side, then a pipe, and then extra words on the right side. So there's no extra words on the left side.
But the pipe, and then 2D. On the right side, RN.
On the left side, it has an extra word nurse. 123 Street and “st”..
And then it converts over to character by character, if it's a single word.
Isn't this awesome?
And here to text the Bill is extra on the left side. William is extra on the right side.
Jelen in common between them. And then identical just returns nothing.
Which is all explained down here. I'll put this code out in the YouTube description.
Actually all the formulas from today. So you can try them out and compare 'em.
Really cool trick here we have group and outline mode.
And this one is collapsed, right?
So I'm going to hold on the shift key and scroll the wheel mouse away from me. So here it has several different queries in here.
That work on a character level.
But this one here is the one that combines them all. So really, really cool method there.
Okay, so here's the recap. I'm giving myself a C.
I got that formula first, but I knew that it wouldn't handle all the cases.
Now Deep Seek did get the exact same formula as me, but it took three tries.
I'm giving it a C minus. Copilot actually got the same formula as me.
But then when I tried to ask it to get better. It just gave me stuff that didn't work.
And then also didn't work.
Copilot Python gave me stuff that didn't work, but it got it in one try. And probably someone with more Python skills.
Hey Owen, where are you? Could probably get that Python to work.
I really think that this is probably a good way to go.
So for these AI bots:. The advantage?
Quick, they were all really quick. I didn't have to go bother my friends.
I got answers right away. But they're all wrong.
I'll give a bonus.
Deep Seek was able to explain the M code and XL Lambda formula.
But then when I asked Deep Seek how to pronounce Geert, it gave me the Dutch pronunciation instead of the Flemish.
And what's kind of funny. Is left out here is Chat GPT.
Who will have to analyze this transcript to come up with the SEO.
And they'll Chat GPT be like, “what the heck? Why are you asking the three of them?
“And you didn't ask me”? Oh, well.
Over here - the gurus.
A's across the board, Rico, what an amazing formula.
And actually analysis of “Really in real life, one of those columns is going to be better than the other.
“And so it's not as complicated as the question seems.
XL Lambda an amazing formula with a couple of lambdas inside.
And then Geert, just an amazing Power Query that actually handles either character by character or word by word depending on it. So, the quick recap.
Bots: Fast, but mostly wrong. The human solutions all worked.
Three innovative methods.
[ Angela ] That AutoSave feature – Bill what are your thoughts on that? [ Bill ] Thanks Angela!
None of the humans require me to turn on AutoSave. The most dangerous feature in Excel.
And of course for Copilot, we have to have AutoSave on, if only temporarily.
A shout-out to the person on YouTube who pointed out.
That just turning on AutoSave. And then opening a file.
Changes the Modification date in Windows Explorer.
I hate AutoSave So, you know, what a great test here. Three bots, three humans.
I am going to call it three and a half humans. I knew that my solution was not the best solution.
There had to be something better. And so when I turned to the bots.
Very quickly got answers that were not any better than my answer.
When I turned to the people who actually are subject matter experts.
I got three answers that were all better than my answers.
And three very different innovative ways to go. So for right now, here we are, February 2025.
Humans still win. And a shout-out to Rico.
I was going back through the e-mails from this project.
And five days ago, Rico said, “I expect a very convincing, well-written response by the AI guys.
“Completely inaccurate but seems plausible to the average user”.
And at the moment I read that, I really thought that I had a good solution from one of the AI Bots.
I guess I was playing the role of the average user.
I was fooled.
I am like, “Rico, you are too skeptical!” But it turns out, you were exactly right, Rico.
Well I want to thank our three human Subject Matter Experts.
Geert, Rico, and XLLambda for their time in solving this. I want to thank you for stopping by.
We'll see you next time for another netcast from MrExcel.
Head-to-head. Alright, so in the last video it was Deep Seek versus Copilot in Excel. Today it's a whole new level.
Okay, we're going to start off with the original problem.
And my not-so-great solution. I realized it wasn't a great solution.
So I went out for help.
I went to Copilot, I went to Deep Seek, I went to Copilot in Python. Then I also sent it to my three friends.
Who were my three smartest Excel people, Rico, XLLambda, and Geert.
All of them came back with different answers. Some of them worked.
If you want to jump to one specific point.
I'll put the timestamps here where you'll find all of them. Let's go.
Hey, this is a great question from Brian.
“Is it possible to compare two text strings and return the differences in a new column”? He has two lists of car make and model.
One might say Toyota Avensis and the other might say Toyota Avensis 2D.
Is it possible to return the difference here? In other words, 2D?
He's thinking that Power Query might be a way to do it.
So I try to come up with some examples here. Including the one that Brian gave me.
But then “Mary Ellen Jelen” and “Mary Ellen Jelen, RN”.
Or “Nurse Mary Ellen” and “Mary Ellen”. “123 Jump Street” versus “Jump St”..
Okay, but here's the one that gets me. This is where my formula is not going to work.
So A-C-E is in this text, but it's not in consecutive letters.
Or “Bill Jelen” and “William Jelen”. What's the difference there?
Okay, so the formula that I'm using here. Is figure out which of these two is shorter.
So in this case, A2 is shorter. But in this case it's column B that's shorter.
And then actually take the longer one and remove all characters that are in the shorter one.
“Toyota Avenesis” from “Toyota Avenesis 2D”. And we're left with space 2D.
Here we are getting space rn.
So I'm really happy with the way that it's working here. I'm even happy with the way it's working here.
It gets the 123, takes out the “Jump St” and gives me the “reet”.
In my opinion, it's not working well here.
And that's when I decided to go bounce this off some experts. Let's take a look.
Just a note from Post-Production.
This is one where you probably really want to download the workbook.
The responses from the various AI bots are very wordy. And I am not going to read all of those.
But I pasted them all in this workbook. All of the examples.
All of the formulas from the AI Bots and our human experts are available.
The link is down in the YouTube description.
So I created this simple workbook that I'm going to upload to Deep Seek.
First thing I'm going to do is select the file and upload. And then the prompt.
So we'll try this prompt.
“This Excel workbook has two columns of text strings starting in A2 and B2. “In each row, the strings are similar.
“But one of the two strings has more text.
There might be extra characters in the beginning, the ending or in the middle.
Can you suggest a formula for C2 that will identify the extra characters in the longer string?
All right, so it can actually see the text that's in my workbook.
We'll copy this.
All right, so it didn't work here where the extra text is before.
So this feels very similar to the method that I came up with.
We will ask it if it can fix C6. Okay, this is great.
So it came up with an even more complicated formula.
I feel like. Okay, while it got the B, D, F, right.
It is worse. Like nurses in here.
1, 2, 3, we're missing the space in re. And I don't know how it came up with WAM.
I don't feel like Deep Seek is a significant improvement over what I had.
Alright, let's try the same task with Copilot. I've Auto Saved - turned that on.
And I'm going to ask Copilot.
We use the exact same prompt, that we sent to Deep Seek.
[ The formula does not work ] [ Typing a new prompt for Copilot ] All right, we'll copy that formula.
No, that's not it at all. I feel like we got a lot further away there.
All right, so here in this one It should be nurse and it's coming up with NUS.
I'm going to give this one a “Fail” for Copilot.
Okay, so we've straight formulas from Copilot, Deep Seek.
I think the actual answer here is going to be Python. Python probably has a way to deal with this.
So to get into Python, choose “Get deeper analysis results using Python”.
And then over here we have to say “Start advanced analysis”.
So in lands the data from the sheet. And then it automatically starts doing something.
Even though I haven't given it a prompt.
And it calculates something called a similarity ratio. So then my first prompt.
Identify the extra characters in the longer string.
And it does that. Let's take a look over here.
So there's the original text. The new text.
Similarity. And then the extra characters.
First thing I noticed is there's a “space 2D” there.
And then look at this. So the R in nurse.
Because there's an R in Mary Ellen. It's not seeing that as an extra character.
So even with Python, it's still not getting exactly what we want.
Alright, here's a beautiful formula from Rico.
Now Rico pointed out that with Bill Jelen and William Jelen. This has an extra character.
And this has an extra character. So Rico actually has two different formulas here.
One where we say remove B7 from A7. And another one where remove A7 from B7.
And he said, of course you can HSTACK these back together if you needed them.
But he suspects that in real life there's going to be one column that's better than the others.
More correct than the others. Recursive Lambda.
You can see it over here in Excel Labs. And I've copied it here.
It's pretty slick.
It substitutes the first instance in Text One, of the first character in Text Two.
And then recursively passes it back to the function.
With the leftmost character of Text Two dropped for the next iteration. What does that really mean?
Okay, so here's Text One. Here's Text Twi.
First character in Text Two is A. So we substitute A with nothing.
That A goes away. And we're left with B, C, D, E, F.
And then the function calls itself again. But this time stripping off the first character.
So we're left with C-E. The first character C.
Substitute C with nothing. We get B, D, E, F.
Call the thing again.
And you end up substituting E and we get that correct result of B, D, F. It's pretty slick.
And it gets the first instance only.
Which is preventing some of the problems that the AI bots were coming up with.
Although it does kind of do this weird thing like nurse is extra. But that “r” is found in “Mary Ellen”.
So this time we're calling the function 11 times.
And it's like right here when it's trying to remove the R. It should be removing that R from Mary Ellen.
But it's removing the first one. Same thing here.
Its removing this E instead of that E. But it's still the extra characters.
I think in principle, this is still really correct.
Okay, now this awesome formula from XL Lambda. And it's in his name< “XL Lambda”.
So there's two Lambdas in here using REDUCE and MAP.
He has two different sets here. Words in B that are not an A.
Or in A that are not in B. Or if you really need what I asked for.
The shorter one, subtracted from the longer one, we have that.
Interesting enough. That since he's working on a word level.
He removes jump from 123 Jump Street.
Removes from 123 St and it ends up with a 123 reet. So that's just kind of interesting.
Although I can see how that's happening.
And then he had an asterisk here for A, B, C, D, E, F, and A, C, E. He's like, Nope, that's not relevant.
The granulation level here is words.
If you really wanted to compare letters, that'd be a different scenario.
And he points out that he has a lot of other algorithms that could do words versus sentences.
The sky's the limit and it can be as complex as regex. There's no magical solution to cover them all.
But I think these are awesome formulas that cover the spirit of the original problem.
Let's whisper this. Like most formulas that XL Lambda sends me.
I can't understand what it's doing.
Let's see if Deep Seek can, This is really good. This is a great explanation.
I'll leave it there on the screen. Now if you remember.
This question was sent in by Brian.
And Brian thought that it was going to be a power query solution. And we have our solution here from Geert.
And it is Power Query. Geert actually has an awesome query right here.
That is showing the difference.
Extra words on the left side, then a pipe, and then extra words on the right side. So there's no extra words on the left side.
But the pipe, and then 2D. On the right side, RN.
On the left side, it has an extra word nurse. 123 Street and “st”..
And then it converts over to character by character, if it's a single word.
Isn't this awesome?
And here to text the Bill is extra on the left side. William is extra on the right side.
Jelen in common between them. And then identical just returns nothing.
Which is all explained down here. I'll put this code out in the YouTube description.
Actually all the formulas from today. So you can try them out and compare 'em.
Really cool trick here we have group and outline mode.
And this one is collapsed, right?
So I'm going to hold on the shift key and scroll the wheel mouse away from me. So here it has several different queries in here.
That work on a character level.
But this one here is the one that combines them all. So really, really cool method there.
Okay, so here's the recap. I'm giving myself a C.
I got that formula first, but I knew that it wouldn't handle all the cases.
Now Deep Seek did get the exact same formula as me, but it took three tries.
I'm giving it a C minus. Copilot actually got the same formula as me.
But then when I tried to ask it to get better. It just gave me stuff that didn't work.
And then also didn't work.
Copilot Python gave me stuff that didn't work, but it got it in one try. And probably someone with more Python skills.
Hey Owen, where are you? Could probably get that Python to work.
I really think that this is probably a good way to go.
So for these AI bots:. The advantage?
Quick, they were all really quick. I didn't have to go bother my friends.
I got answers right away. But they're all wrong.
I'll give a bonus.
Deep Seek was able to explain the M code and XL Lambda formula.
But then when I asked Deep Seek how to pronounce Geert, it gave me the Dutch pronunciation instead of the Flemish.
And what's kind of funny. Is left out here is Chat GPT.
Who will have to analyze this transcript to come up with the SEO.
And they'll Chat GPT be like, “what the heck? Why are you asking the three of them?
“And you didn't ask me”? Oh, well.
Over here - the gurus.
A's across the board, Rico, what an amazing formula.
And actually analysis of “Really in real life, one of those columns is going to be better than the other.
“And so it's not as complicated as the question seems.
XL Lambda an amazing formula with a couple of lambdas inside.
And then Geert, just an amazing Power Query that actually handles either character by character or word by word depending on it. So, the quick recap.
Bots: Fast, but mostly wrong. The human solutions all worked.
Three innovative methods.
[ Angela ] That AutoSave feature – Bill what are your thoughts on that? [ Bill ] Thanks Angela!
None of the humans require me to turn on AutoSave. The most dangerous feature in Excel.
And of course for Copilot, we have to have AutoSave on, if only temporarily.
A shout-out to the person on YouTube who pointed out.
That just turning on AutoSave. And then opening a file.
Changes the Modification date in Windows Explorer.
I hate AutoSave So, you know, what a great test here. Three bots, three humans.
I am going to call it three and a half humans. I knew that my solution was not the best solution.
There had to be something better. And so when I turned to the bots.
Very quickly got answers that were not any better than my answer.
When I turned to the people who actually are subject matter experts.
I got three answers that were all better than my answers.
And three very different innovative ways to go. So for right now, here we are, February 2025.
Humans still win. And a shout-out to Rico.
I was going back through the e-mails from this project.
And five days ago, Rico said, “I expect a very convincing, well-written response by the AI guys.
“Completely inaccurate but seems plausible to the average user”.
And at the moment I read that, I really thought that I had a good solution from one of the AI Bots.
I guess I was playing the role of the average user.
I was fooled.
I am like, “Rico, you are too skeptical!” But it turns out, you were exactly right, Rico.
Well I want to thank our three human Subject Matter Experts.
Geert, Rico, and XLLambda for their time in solving this. I want to thank you for stopping by.
We'll see you next time for another netcast from MrExcel.
