This is a guest post by Daniel Ferry of Excelhero.com.
Have you ever wanted to fetch live stock quotes from excel? In this post we will learn about how to get stock quotes for specified symbols using macros.
One method that has worked well for my clients can be implemented with just a few lines of VBA code. I call it the ActiveRange.
An ActiveRange is an area on a worksheet that you define by simply entering the range address in a configuration sheet. Once enabled, that range becomes live in the sense that if you add or change a stock symbol in the first column of the range, the range will automatically (and almost instantly) update. You can specify any of 84 information attributes to include as columns in the ActiveRange. This includes things such as Last Trade Price, EBITDA, Ask, Bid, P/E Ratio, etc. Whenever you add or change one of these attributes in the first row of the ActiveRange, the range will automatically update as well.
Sound interesting, useful?
In this post, you can learn how to use excel macros to fetch live stock quotes from Yahoo! Finance website. It is also going to be a crash course in VBA for the express purpose of learning how the ActiveRange method works so that you can use it yourself.
Download Excel Stock Quotes Macro:
Click here to download the excel stock quotes macro workbook. It will be much easier to follow this tutorial if you refer to the workbook.
Background – Understanding The Stock Quotes Problem:
The stock information for the ActiveRange will come from Yahoo Finance. A number of years ago, Yahoo created a useful interface to their stock data that allows anyone at anytime to enter a URL into a web browser and receive a CSV file containing current data on the stocks specified in the URL. That’s neat and simple.
But it gets a little more complicated when you get down to specifying which attributes you want to retrieve [information here]. Remember there are 84 discreet attributes available. Under the Yahoo system, each attribute has a short string Tag Code. All we need to do is to concatenate the string codes for each attribute we want and add the resulting string to the URL. We then need to figure out what to do with the CSV file that comes back.
Our VBA will take care of that and manage the ActiveRange. Excel includes the QueryTable as one of its core objects, and it is fully addressable from VBA. We will utilize it to retrieve the data we want and to write those data to the ActiveRange.
Before we start the coding we need to include two support sheets for the ActiveRange. The first is called “YF_Attribs”, and as the name implies is a list of the 84 attributes available on Yahoo Finance along with their Yahoo Finance Tag Codes. The second sheet is called, “arConfig_xxxx” where xxxx is the name of our sheet where the ActiveRange will reside. It contains some configurable information about the ActiveRange which our VBA will use.
All of the VBA code for this project will reside inside of the worksheet module for the sheet where we want our ActiveRange to be. For this tutorial, I called the sheet, “DEMO”.
Writing the Macros to Fetch Stock Quotes:
Press ALT-F11 on your keyboard, which will open the VBE. Double click on the DEMO sheet in the left pane. We will enter out code on the right. To begin with, enter these lines:
Private rnAR_Dest As Range
Private rnAR_Table As Range
Private stAR_ConfigSheetName As String
Always start a module with Option Explicit. It forces you to define your variable types, and will save you untold grief at debugging time. In VBA each variable can be one of a number of variable types, such as a Long or a String or a Double or a Range, etc. For right now, don’t worry too much about this – just follow along.
Sidebar on Variable Naming Conventions
Variable names must begin with a letter. Everyone and their brother seems to have a different method for naming variables. I like to prefix mine with context. The first couple of letters are in lower case and represent the type of the variable. This allows me to look at the variable anywhere it’s used and immediately know its type. In this project I’ve also prefaced the variables with “AR_” so that I know the variable is related to the ActiveRange implementation. In larger projects this would be useful. After the underscore, I include a description of what the variable is used for. That’s my method.
In the above code we have defined three variables and their types. Since these are defined at the top of a worksheet module, they will be available to each procedure that we define in this module. This is known as scope. In VBA, variables can have scope restricted to a procedure, to a module (as we have done above), or they can be global in scope and hence available to the entire program, regardless of module. Again we are putting all of the code for this project in the code module of the DEMO worksheet. Every worksheet has a code module. Code modules can also be added to a workbook that are not associated with any worksheet. UserForms can be added and they have code modules as well. Finally, a special type of code module, called a class module, can also be added. Any global variables would be available to procedures in all of these. However, it is good practice to always limit the scope of your variables to the level where you need them.
In that vein, notice that the three variables above are defined with the word Private. This specifically restricts their scope to this module.
Every worksheet module has the built-in capability of firing off a bit of code in response to a change in any of the sheet’s cell values. This is called the Worksheet_Change event. If we select Worksheet from the combo box at the top and Change in the other combo box, the VBE will kindly define for us a new procedure in this module. It will look like this:
End Sub
Notice that by default this procedure is defined as Private. This is good and as a result the procedure will not show up as a macro. Notice the word Target near the end of the first line. This represents the range that has been changed. Place code between these two lines so that the entire procedure now looks like this:
The Heart of our Excel Stock Quotes Code – Worksheet_Change()
ActivateRange
If Worksheets(stAR_ConfigSheetName).[ar_enabled] Then
If Intersect(Target, rnAR_Dest) Is Nothing Then Exit Sub
If Target.Column <> rnAR_Dest.Column And Target.Row <> rnAR_Dest.Row Then
PostProcessActiveRange
Exit Sub
End If
ActiveRangeResponse
End If
End Sub
That may look like a handful but it’s really rather simple. Let’s step through it. The first line is ActivateRange. This is the name of another sub-procedure that will be defined in a moment. This line just directs the program to run that sub, which provides values to the three variables we defined at the top. Again, since those variables were defined at the top of the module, their values will be available to all procedures in the module. The ActivateRange procedure gives them values.
Next we see this odd looking fellow:
If Intersect(Target, rnAR_Dest) Is Nothing Then Exit Sub
All this does is check to see if the Target (the cell that was changed on the worksheet) is part of our ActiveRange. If it is the procedure continues. If it’s not, the procedure is exited.
The next line checks to see if the cell that was changed is in the first column or first row of the ActiveRange. If it is, the post processing is skipped. If the change is any other part of the ActiveRange, another sub-procedure (defined below) is run to do some post processing of the retrieved data, and then exits this procedure.
If the cell that changed was in the first column or the first row, the program runs another sub-procedure, called ActiveRangeResponse, which is also defined below. ActiveRangeResponse builds the URL for YF, deletes any previous QueryTables related to the ActiveRange, and creates a new QueryTable as specified in our configuration sheet.
That’s it. The heart of the whole program resides here in the Worksheet_Change event procedure. It relies on a number of other subprocedures, but this is the whole program. When a change is made in the ActiveRange’s first column (stock symbols) or its first row (stock attributes), ActiveRangeResponse runs and our ActiveRange is updated.
Understanding other sub-procedures that help us get the stock quotes:
So let’s look at those supporting subprocedures. The first is ActivateRange:
stAR_ConfigSheetName = “arConfig_” & Me.Name
Set rnAR_Dest = Me.Range(Worksheets(stAR_ConfigSheetName).[ar_range].Value)
Set rnAR_Table = rnAR_Dest.Resize(1, 1).Offset(1, 1)
Worksheets(stAR_ConfigSheetName).[ar_YFAttributes] = GetCurrentYahooFinancialAttributeTags
End Sub
Again, all this does is give values to our three module level variables. In addition it builds the concatenated string of YF Tag Codes required for the URL. It does this by calling a function that I’ve defined at the very bottom of the module, called GetCurrentYahooFinancialAttributeTags.
The next subprocedure is ActiveRangeResponse:
Dim vArr As Variant
Dim stCnx As String
Const YAHOO_FINANCE_URL = “http://finance.yahoo.com/d/quotes.csv?s=[SYMBOLS]&f=[ATTRIBUTES]”
vArr = Application.Transpose(rnAR_Dest.Resize(rnAR_Dest.Rows.Count – 1, 1).Offset(1))
stCnx = Replace(YAHOO_FINANCE_URL, “[SYMBOLS]”, Replace(WorksheetFunction.Trim(Join(vArr)), ” “, “+”))
stCnx = Replace(stCnx, “[ATTRIBUTES]”, Worksheets(stAR_ConfigSheetName).[ar_YFAttributes])
AddQueryTable rnAR_Table.Resize(UBound(vArr)), “URL;” & stCnx
End Sub
Notice that here we have variables defined at the top of this procedure and consequently their scope is limited to this procedure only. This means that we could have the same variable names defined in other procedures but those variables would not be related to these and would have completely different values.
Next notice that we have defined a constant. This is good practice, as it forces us to specify what the constant value is by naming the constant. I could have just used the value where I later use the constant, but then the question arises as to what is this value and where did it come from. Here I have named the value, YAHOO_FINANCE_URL, removing all doubt as to its purpose.
The next line is this:
vArr = Application.Transpose(rnAR_Dest.Resize(rnAR_Dest.Rows.Count - 1, 1).Offset(1))
and it deserves some explanation. Let me back up by saying that whenever we write or read multiple cells from a worksheet we should always try to do it in one go, rather than one cell at a time. The more cells involved the more important this is. Otherwise we pay a massive penalty in processing time. One of the best optimization techniques available is to replace code that loops through cell reads/writes and replace it with code that reads/writes all the cells at once. It can literally be hundreds to thousands of times faster.
Here we are interested in getting the list of all of the stock symbols in the first column of the ActiveRange. So how do we get them in one shot? We use something called a variant array. Notice that we defined vArr at the top of this procedure. A variant array is a special kind of variable that holds a list of values and it DOES NOT CARE what variable types those values are. This is important when retrieving data from a sheet because the data could be numbers, text, Boolean (True or False), etc. Variants are powerful, but they are much slower than other variable types, such as a Long for numeric data for example. However, in the case of retrieving or writing large chunks of data from/to a sheet the slight penalty of the variant is dwarfed by the massive increase in the speed of data transfer.
It’s very simple to retrieve range data (regardless of the size) into a variant array. All you do is:
v = range
where v is defined as a variant and range is any VBA reference to a worksheet range. And magically all of the values in that range are now in v. Note that v is not connected to the range. A change in any of v’s values does not propogate back to the range, and likewise a change to the range does not make it’s way to v all by itself. v will ALWAYS be a two-demensional array. The first dimension is the index of the rows, the second dimension is the index of the columns. So v(1,1) will refer to the value that came from the top left cell in the range. v(6,9) will hold the value that came from the cell in the range at row 6 and column 9.
For most circumstances this two-dimensional format is fine. But we are only retrieving one column of stock symbols. The procedure will still give us a two-dimensional array, with the column dimension being only 1 element wide. This is a shame because VBA has a wonderful function called Join that allows you in one step (no loop) to concatenate every element of an array into a string. You can even specify a custom string to delimit (go in-between) each element in the output string. The problem is that Join only works on single dimensioned arrays 🙁
But there’s always a way, right? We can use the Application.Transpose method on the 2-D array and presto we get a 1-D array. The rest of the line just specifies what range (the stock symbols) to grab.
The next two lines are:
stCnx = Replace(YAHOO_FINANCE_URL, "[SYMBOLS]", Replace(WorksheetFunction.Trim(Join(vArr)), " ", "+"))
stCnx = Replace(stCnx, "[ATTRIBUTES]", Worksheets(stAR_ConfigSheetName).[ar_YFAttributes])
Again a handful, but all we are doing here is replacing the monikers, [SYMBOLS] and [ATTRIBUTES] in the YAHOO_FINANCE_URL constant with the list of stock symbols (delimited by a plus sign) and the string of attributes.
In the final line of the procedure:
AddQueryTable rnAR_Table.Resize(UBound(vArr)), "URL;" & stCnx
we are running another subprocedure called, AddQueryTable and we are telling it where to place the new QueryTable and providing the connection string for the QueryTable, which in this case is the YF URL that we just built.
Nothing unusual happens in the AddQueryTable sub. It just deletes any existing AR related QueryTables and adds the new one according to the options in the configuration sheet.
The PostProcessActiveRange sub is interesting:
If rnAR_Dest.Columns.Count > 2 Then
Application.DisplayAlerts = False
rnAR_Table.Resize(rnAR_Dest.Rows.Count).TextToColumns Destination:=rnAR_Table, DataType:=xlDelimited, Comma:=True
Application.DisplayAlerts = True
Worksheets(stAR_ConfigSheetName).[ar_LocalTimeLastUpdate] = Now
End If
End Sub
Processing Yahoo Finance Output using Query Table & Text-Import Utility:
As mentioned before the data from YF comes back as a CSV file. The QueryTable dumps this into one column. If you were only retrieving one attribute for each stock this would be fine as is. However, two or more attributes is going to result in unwanted commas and multiple attribute values squished into the first column of the QueryTable output. Unfortunately this is poor design by Microsoft, especially when you consider that the QueryTable does not behave like this when it is retrieving SQL data or opening a Text file from disk. You can actually specify this operation to be a text file and it will properly spread the output over all of the columns. To do so, you specify the disk location as being the URL of the YF CSV file, but as Murphy would have it, it’s unbelievably slow and pops up a status dialog as it slowly retrieving the CSV. Using the URL instruction instead of the TEXT instruction at the beginning of the connection string is incredibly fast in comparison, but dumps all of the data into the first column.
So what to do? We’ll just employ Excel’s built-in TextToColumns capability and bam, our data is where we want it.
Our finalized stock quotes fetcher worksheet should look like this:
Download Excel Stock Quotes Macro:
Click here to download the excel stock quotes macro workbook. It will be much easier to follow this tutorial if you refer to the workbook.
Final Thoughts on Excel Stock Quotes
The ActiveRange technique is quite versatile. It can be implemented with other data sources such as SQL, or even lookups to other Excel files, or websites.
In this example it provides a nice way to easily track whatever stocks you may have interest in and up to 84 different attributes of those stocks. You can enable and disable the activeness of the ActiveRange on the fly. You can set the AR to AutoRefresh the data at periods that you set or to not refresh at all.
This is a basic implementation. For example, changing the AutoRefresh setting will have no effect until a new QueryTable is built. That won’t happen until you also add or change a stock symbol or add or change an attribute. An easy enhancement would be to add a little code to the arConfig_DEMO code module to respond to changes to the ar_AutoRefresh named range cell.
Another enhancement would be to eliminate the slight flicker of the update by moving the QueryTable destination to the arConfig_DEMO and then doing the TextToColumns with the destination set to the DEMO sheet. In an effort to simplify this tutorial I have left these easy enhancements as an exercise for you to implement.
Have a question or doubt? Please Ask
Do you have any questions or doubts on the above technique? Have you used ActiveRange or similar implementations earlier? What is your experience? Please share your thoughts / questions using comments.
I read Chandoo.org regularly and will be monitoring the post for questions. But you can also reach me at my blog:
Further References & Help on Excel Stock Quotes [Added by Chandoo]
- Fetching Stock Quotes using Research Pane
- Stock Portfolio Tracker using Google Docs
- QueryTable Object Model & Properties
- Using QueryTable to Generate Dynamic Reports
- Yahoo Finance API Documentation & Example
Excel Hero is dedicated to expanding your notion of what is possible in MS Excel and to inspiring you to become an Excel Hero at your workplace. It has many articles and sample workbooks on advanced Excel development and advanced Excel charting.
67 Responses to “How to use XLOOKUP in Excel?”
Sure it's a nice new command. It would be useful if everyone had access to it. But if there is any chance you will be sharing the file with someone who has a onetime payment Office license, or an older version of Office you can't use it.
That is my biggest gripe with many new features MS is launching. With such vast userbase and existing spreadsheet "systems", all of these formulas are going to create more trouble than imagined. That said, we should learn new things, especially if you move to a new job chances are you will be using a different version of Excel there.
I love to learn new things, like this new command. But I can't afford, literally don't have the money, to keep paying for 365.
This is the thing that especially offends me about the Office 365 pricing scam/scheme. Sure, if they want to milk more money from users using the rental scam, fine I know I don't have to fall for it. But restricting new "features", like new commands to 365 is offensive. It makes one-time payment users "second class" customers, especially anyone who has paid for Office 2019. At least in the past new features/commands came only came out every few years, with new versions so there was some logic to the separation. But now the new features are coming every few months and there is no real separation between 2019 and 365, but still they limit the new features to 365. Even 2016 is close enough. MS "accidentally" pushes a few new features to 2016, when they feel like it or when they are too lazy to do the extra work to prevent them from going to 2016.
I agree with Ron I have MS Office 2019 which I used for Charity work but a pensioner I find the cost of the MS365 unaffordable. Perhaps there is some way for a Ms Guru to perhaps create 3rd party update for the stand alone versions.
I will however continues with Ms 365 this year as I have just renewed the subscription
thanks very much for keeping us abreast of latest developments and also the excel community for their useful feed back
regards Brian 18/03/2024
Good point. I suggest using the free MS Office online (you just need onedrive account) to maintain old files and work on them. The only limitation is that it is browser based, so you won't be able to do many advanced things. But it is better than the alternative of shelling out $100+ every year.
Yes, of course this is the latest and excellent update from Microsoft but this feature will take years to come in the market because most of the people or offices are still using Office 2007 or 2013.
Dear Chandoo Sir
Thank you for updating latest idea this idea is centralized lookup formula all about.
this idea is realy impressive and samart
I couldn't observe any benefit, over MATCH+INDEX.
Hmm, the base scenario is similar to index+match, but XLOOKUP makes life simple with single formula and default "exact match" setup. Plus I find the "lookup from last" and "less than" "greater than" options very useful and less cryptic than MATCH options.
Thanks for sharing, it added some excitement to my Friday morning! I don't have 365 but am still excited to be aware of the existence of these features! I know that vlookup on larger sets of data can really take up some resources--it makes sense, it's performing a lot of operations for us while we sit and sip on coffee. 😉 However, I'm wondering if you've you noticed a difference in performance with xlookup? Is it slower, faster, or pretty much the same in terms of calculation speed?
I haven't tested it against VLOOKUP or INDEX+MATCH. If anything, I would guess that the performance should be similar as they could all use same logic internally. I will try this and share some outcomes later.
I would love to know the results. We're crunching a ton of data and I love the simplicity of XLOOKUP, but we can't handle the sluggishness of VLOOKUP. I hope XL is faster!!!
I believe XLOOKUP has been written to deliver exact matches at the same speed as a binary (vlookup's approximate) search.
Here is a nice overview of differences in performance of different lookup formulas. Unexpected, but XLOOKUP is not always fastest.
https://professor-excel.com/performance-of-xlookup-how-fast-is-the-new-xlookup-vs-vlookup/?amp#What_is_the_8220binary_search_mode8221_of_XLOOKUP
You can use an if logic to wrap around a vlookup with a TRUE argument to speed up lookups.
A nice addition to the function list. Very usefull and easier to use then INDEX + MATCH.
Since XLOOKUP is in beta testing, it would be great if Microsoft development team added a 5th. argument: if_na. That is: if XLOOKUP returns #N/A, an alternate value could be returned instead. Therefore, it wouldn't be necessary to do =IFNA(XLOOKUP(...), value_if_na).
Good idea. But I feel this can be a dangerous precedent as no other formula in Excel has fail-safe option (other than IFERROR and IFNA ofcourse). So may be leave it to return error.
Don't overlook the new FILTER function. That has a final [if_empty] setting.
Although I don't have and expecting to be around soon in EXCEL 2019, my question is there a way to work around the new function "xlookup" but not the old ones.
However it is appreciated tip,thanks
Chandoo
You can also use XLookup like
=Sum(xlookup():Xlookup())
Refer the example 4 at:
https://support.office.com/en-us/article/xlookup-function-b7fd680e-6d10-43e6-84f9-88eae8bf5929?ui=en-US&rs=en-US&ad=US
This makes it hugely powerful as it is returning an address like Index can do
Great point Hui. I am yet to find a practical use case for summing between lookups, but I am pretty sure others will find this useful.
Here is an idea.
If you wish to analyse data for a given month, the relevant portion of the Sales table (sorted by date) is given by
= XLOOKUP( EOMONTH(month,0), EOMONTH(+sales[Date],0), sales,0,1 ) :
XLOOKUP( EOMONTH(month,0), EOMONTH(+sales[Date],0), sales,0,-1 )
which can be referred to as a named formula 'selected'. Being a reference to the original table, range intersection with columns works. Hence
= XLOOKUP( MAX(selected sales[Net Sales]),
selected sales[Net Sales], selected sales[Sales Person] )
provides an answer to
Who had most sales for February?
Caution: The formula requires 7 separate searches of the data but they are very fast.
I use VLOOKUP a lot with named ranges, are you able to reference those in XLOOKUP?
@Hamish... you should be able to use any reference styles that work with other formulas in XLOOKUP. So yes for names, structural, cell and references to other sheets / workbooks.
Hamish, Yes it all works perfectly. That includes cases in which the data table does not comprise raw data but rather is made up of dynamic arrays. Naming the anchor cell of each dynamic array allows expressions such as
= XLOOKUP( MAX(selectedNetSales#), selectedNetSales#, selectedSalesPerson# )
Conversely, if the returned field is comprised of anchor cells for separate dynamic lists (e.g. employment data for the specified salesman) then the list can be returned by adding '#'
=XLOOKUP(0,sales[Net Sales],EmployeeInfo,1)#
Since the documentation says it returns a reference array, could you write formulas that could answer questions that need to perform a function upon a result set that contains multiple rows such as:
1. What is the total Profit/Loss for SalesPersons named [Jamie]?
2. What is the MAX/MIN Net Sales for SalesPersons named [Jamie]?
3. What was the Average Net Sales for everyone that had exactly [8] Customers?
I think the answer to your question is 'no' unless you are willing to sort the table so that the records you wish to aggregate form a continuous range. That is, the formula
= SUM(
XLOOKUP(salesPerson,sales[Sales Person],sales[Profit / Loss],,,1):
XLOOKUP(salesPerson,sales[Sales Person],sales[Profit / Loss],,,-1))
only works if the data is sorted by Sales Person.
Otherwise it looks like SUMIFS (and similar) offers the best solutions with FILTER a close second.
= SUMIFS( sales[Profit / Loss], sales[Sales Person], salesPerson )
= SUM( FILTER(sales[Profit / Loss], sales[Sales Person]=salesPerson ) )
XLOOKUP allows us to look for a variable in a column and return a value from a row: combining VLOOKUP ad HLOOKUP in essence.
I watched a video last night in which the presenter showed an example that returned an error. The solution that the presented was using is this: =XLOOKUP(A4,B7:B9,C6:E6)
To see the problem in action, put a b c in the range B7:B9 and 1 2 3 in the range C6:E6 and in A4 enter a or b or c
I solved this problem in this way:
=XLOOKUP(A12,B15:B17,TRANSPOSE(C14:E14))
I have also set up a financial analysis example in which I wanted to find, for every line item in an income statement, which month was exactly equal to the mean of that row or which was immediately below the mean or immediately above it. Or Median, or Standard Deviation ...
I used XLOOKUP() and IFS() together with Data Validation (although that is optional) and while the formula is a little unwieldy, again I am effectively combining vertical and horizontal lookups.
Excellent find and tip Duncan 🙂
Hi,
Can you please tell me if there is any way to return multiple values with a single match.
Thanks in Advance
when will be in excel 2019
Thanks
Never.
"New features" like the XLookUp() command are only added to Office 365. They will never be added to Office 2019. They may show up in Office V-Next, when ever it comes out, in the near future. MS has not yet announced a new version. If they follow the pattern in the last few versions that would be fall 2021. But that is only a guess.
I have it now in office 2021
I downloaded your sample spreadsheet and three of your first seven examples are incorrect. Then I stopped.
Which version of Excel are you running? XLOOKUP doesn't work in any version except Office 365.
Hi, Chandoo.
Great tips, thanks!
In example #11, "What is the 'net sales' for Johnson? = 1540" the formula only takes into account the first match for Johnson (D10)?
In row 21 Johnson appears again so the correct answer should be 4192 (D10 + D21).
Imagine a DB with hundreds of records!
How can we deal with duplicates using XLOOKUP?
Thanks.
Is there an easy way to handle if the cell is blank in the data table to prove the result of a blank? With VLOOKUP, previously to get this result, I had to do:
=IF(VLOOKUP($B2,data,6,FALSE)="","",VLOOKUP($B2,data,6,FALSE))
I am hoping that I don't have to resort to the same lengthy format. I did try the "Value Not Found" example you provided (love it). However that is when the search value is not listed, not when the search value is found and the result value is a blank cell.
Thanks for everything you do!!!!
Hi Sherry,
Are you using the IF formula to show "" instead of 0 ?
If so, you can use this structure
=XLOOKUP($B$2, data[col1], data[col6]) & ""
This will force 0 to convert to empty space. It won't impact other results though, (assuming column 6 is text)
column 6 is a date.
A bit longer, but to force the 'value not found' you could remove the entry from the lookup array
= XLOOKUP(lookupValue,
IF(data[col6]"", data[col1]),
data[col6], "Missing data")
Hi Chandoo,
I've been waiting for this function for months so that I could replace all my INDEX / MATCH / MATCH statements. However, I have hit a snag with using nested XLOOKUPs as replacements. If the inner XLOOKUP can't find a value, then whatever value I specify as the [if not found] value causes the outer XLOOKUP to fail and return #VALUE. So the [if not found] functionality works if a single XLOOKUP can't find the search value, but it causes nested XLOOKUPs to fail. Can you see any way around that?
Thanks
Hey Stuart... Can you share an example of what result you are expecting in nested case? One option is to use a single IFERROR outside all the nested functions.
@Stuart
Do not limit yourself to thinking of [if_not_found] as being a text string, e.g. "Oops"; it can be a formula in its own right, returning a default row from the original table or even a lookup from an alternative table.
What it must return is an array in order to form a valid parameter for the outer XLOOKUP.
Hi Peter,
You've got it! As you suggest, by setting the inner XLOOKUP to return an array full of zeroes (or whatever) solves the problem. The outer XLOOKUP can of course just have 0, or whatever, stated its if_not_found value.
I am surprised that I haven't come across this issue or solution anywhere else. There are lots of blogs / videos which mention using nested XLOOKUPs as a replacement for INDEX / MATCH / MATCH. I can't say I've read or watched them all, but the ones I have don't mention this issue. I suspect there are / will be a lot of people getting #N/As or, worse, #VALUES depending on what they specify as the inner function's if_not_found.
Thanks for your help!
I am trying to lookup a date and name and return the number of hours from another worksheet? If I'm mixing text and dates, will this still work?
Great article. But,...two questions:
1) I do have Office 365. Yet, the XLookup is not recognized by Excel. Your sample file displays a #NAME? Why?
2) In your samplefile you have a leading '_xlfn.' in front of the formula. Why is that?
Hi Michael...
Can you confirm what is your current version of Excel is? Also see if you can update to newer version. You can do both from File > Account.
Great Job..
My values that I want to join are not exact, i.e.
000025868 and 0000258 68 Total
Is there a way to join the data?
Interesting. Assuming the space is in the lookup column, try this:
=xlookup("000025868", substitute(lookup_col, " ",""), result_col)
Getting a #N/A as the results.
Is there a way to convert "0000258 68 Total" to 000025868 (or visa versa) before I run the =XLOOKUP?
If you just want to remove the word "total" at the end, use SUBSTITUTE for that. If there can be other words, you are better off first running the data thru Power Query so you can clean it.
One thing that is possible is to take a numeric lookup value and convert it to text before searching a text lookup array. For example
= XLOOKUP(TEXT( value, "0000000\?00\*" ), array, return, , 2 )
will perform a search with wildcards that allow "Total" to be appended or any character to be inserted two digits before the end of the number.
That would pick up
"0000258 68 Total"
but you would need an alternative test to match the number 25868, itself.
Check the reference, while selecting data the xlookup function automatically starts from new line. Try changing it to the first row and it would work.
YOU ARE THE EXCEL KING!
Thank you
Hi Chandoo,
I have 2 sheets with 5 columns. data in columns A:C is similar except that changes are made in columns A and C. I want to lookup in column C in Sheet2 and update Sheet1 columns A:C.
for example
Sheet1
ColA ColB ColC
123 AB12 One
234 BC23
323 CB22 Six
Sheet2
ColA ColB ColC
123 AB12 One
234 BB22 Two
323 CB22 Six
I don't think we can claim that XLOOKUP "replaces" INDEX+MATCH. Yes, it provides a suitably powerful alternative, and is absolutely a full replacement for VLOOKUP and HLOOKUP, but it can't easily play some of the "math" games that are possible with INDEX+MATCH and sometimes even necessary when the data isn't in a convenient layout.
What if you needed the row above or below the match or if the data was laid out in repeating sections where you first needed to know the location of the section header and then the location of a given item within each section? Both of those problems can be solved with plus/minus shifting of the number returned from the MATCH.
So I would argue that INDEX+XMATCH are the true replacement for INDEX+MATCH, thus taking full advantage of the X -- defaulting to exact matches, virtual sorting, and so on -- while preserving the ability to "shift" the match as needed.
I'm looking for a price in a multiple column price list. With Vlookup, I specified the entire table and for the column, looked at the user selected model/column. In Xlookup, how to specify the column number and the range up and down or can I just specify the column number only?
One advantage that VLOOKUP retains over XLOOKUP is the ability to supply a lookup column number dynamically, as a purely numerical result of a calculation. To replicate this functionality using XLOOKUP, you would need seperate logic to calculate the column reference (i.e. the column's number, range name or range address) and pass it to the XLOOKUP formula. You could do this inside the XLOOKUP function by setting up the 3rd param of XLOOKUP to be based on your "user selected model/column".
Using Xlookup with "match mode" = -1 and "if not found" = "ABC"
Now if the lookup value is not found in the lookup_array excel gives the the highest value from the return_array.
This is not what I expect from xlookup.
It should return "ABC"
Can you explain why?
Chandoo,
I am having trouble with XLookUp. How do I get it to return multiple values such as employees with salary greater than $45,000 or to sum all the sales in the East region? Are these more pivot table inquires?
Is XLOOKUP more useful for finding one record than multiple records?
Thank you,
Jennifer Jeffords
Hi Chandoo,
Is it possible to use XLOOKUP to return a status such as "Checked" and "NoCheck"(something similar to IF stmt)
Thank you.
I used the index and match to look up the hourly rate for a job classification as a part of a drop down. Now, I want to calculate the hourly rate multiplied by hours worked and the cell will not calculate. What might be the problem? The results cell of the look-up is formatted to be currency?
You show return array can be more than 1 column but what about Look up array? What if I want to find a value than can be in 1 of 3 columns and then return one value from another column.
You can use XLOOKUP for such things too.
For example, if you have three columns: home phone, cell phone and email address
and a column with customer name
and you want to lookup the name of the customer when you specify any value from one of those 3 columns,
you can use the below XLOOKUP.
=XLOOKUP(TRUE,BYROW(C3:E22=I2,LAMBDA(a, OR(a))), B3:B22, "No record found!")
Here I2 contains the search criteria (either home phone, cell phone or email)
B3:B22 have names
C3:E22 have the home / cell / email values
Hi my name is Musawir Rasool i am from India in a state of jammu and Kashmir I love watching your videos and lot from your videos
Thanks
And one more can u teach me full power bi?
Hi Chandoo,
I was referring to your xlookup-examples file, and in that I saw your formula for Sl. 8 - Who has least sales? You wrote formula =XLOOKUP(0,sales[Net Sales],sales[Sales Person],,1) but I think a more better way would be to write =XLOOKUP(MIN(sales[Net Sales]),sales[Net Sales],sales[Sales Person],,1). This is because your formula would not reliable unless you're specifically looking for a salesperson who has exactly 0 in sales, which is not the same as the least sales — unless 0 happens to be the lowest. Also, the 1 as the last argument means "approximate match in ascending order," which could return wrong results if 0 isn't found.