**Please Note: **
This article is written for users of the following Microsoft Excel versions: 97, 2000, 2002, and 2003. If you are using a later version (Excel 2007 or later), *this tip may not work for you*. For a version of this tip written specifically for later versions of Excel, click here: Concatenating Values from a Variable Number of Cells.

Pam has two columns of data. In column A there are simple identifiers, such as A, B, C, etc. In column B there are a series of integer values. She can sort the data by the identifier and, secondarily, by the integer values. Now she wants, in column C, to have a formula that will concatenate all the integer values for a particular identifier. Thus, if A1:A4 all contain the identifier A, then in cell C1 she would like to have all the values in B1:B4 concatenated and divided by commas, such as "11, 17, 19, 25". Since the number of rows for each identifier can be different, Pam isn't sure how to go about the concatenation.

The easiest way to accomplish this is to use a macro, which can be created as a user-defined function. Here's an example:

Function CatSame(c As Range) As String Application.Volatile sTemp = "" iCurCol = c.Column If iCurCol = 3 Then If c.Row = 1 Then sLast = "" Else sLast = c.Offset(-1, -2) End If If c.Offset(0, -2) <> sLast Then J = 0 Do sTemp = sTemp & ", " & c.Offset(J, -1) J = J + 1 Loop While c.Offset(J, -2) = c.Offset(J - 1, -2) sTemp = Right(sTemp, Len(sTemp) - 2) End If End If CatSame = sTemp End Function

This function basically takes a value that is passed to it (a cell reference) and verifies that the cell reference is for column C. If it is, then it starts to concatenate values from column B based on the values in column A. It only returns the string of concatenated values if the value is column A is different than the value in the row above it. Assuming your identifiers are in column A and your values to be concatenated are in column B, you could place the following in column C:

=CatSame(C1)

Copy this down as far as necessary in column C and you end up with exactly what Pam wanted.

A more versatile function would be one that would function somewhat like VLOOKUP, but bring back a concatenated list of values that match whatever you are looking up. Consider the following function:

Function VLookupAll(vValue, rngAll As Range, _ iCol As Integer, Optional sSep As String = ", ") Dim rCell As Range Dim rng As Range On Error GoTo ErrHandler Application.Volatile Set rng = Intersect(rngAll, rngAll.Columns(1)) For Each rCell In rng If rCell.Value = vValue Then _ VLookupAll = VLookupAll & sSep & _ rCell.Offset(0, iCol).Value Next rCell If VLookupAll = "" Then VLookupAll = CVErr(xlErrNA) Else VLookupAll = Right(VLookupAll, Len(VLookupAll) - Len(sSep)) End If ErrHandler: If Err.Number <> 0 Then VLookupAll = CVErr(xlErrValue) End Function

This function takes up to four arguments. The first is the value you want to match in your lookup. In Pam's instance, this would be the identifier you want, such as A, B, or C. The second argument is the range of cells in which to look for the matches (column A in this case). The third argument is an offset (from the range in the second argument) that represents the values you want concatenated. You can use the function in this manner:

=VLookupAll("B",A1:A99,1)

If you want to specify a different delimiter between values, you can do it using the optional fourth argument. For instance, the following returns a string where a dash separates each value:

=VLookupAll("B",A1:A99,1,"-")

The solutions so far have focused on using macros. The reason for this is relatively simple: There isn't a formula-based solution that can do what Pam needs. Using nested IF statements to evaluate what is in column A won't work well because you are limited in how deeply IF statements can be nested.

You could use a formula and an intermediate result if you don't mind having the concatenated values be at the last instance of an identifier in column A. Start by putting this formula in cell C1:

=B1

This formula should go into cell C2:

=IF(A2=A1,C1 & ", " & B2, B2)

Copy this formula down as many rows as necessary. What you end up with is an increasingly long series of concatenated values in column C, with the longest in each run being on the same row as the last sequential identifier in column A. You can then put the following in all the applicable cells of column D:

=IF(LEN(C2)>LEN(C1),"",C1)

This formula only displays the longest strings from column C, which is what Pam needed to begin with.

*ExcelTips* is your source for cost-effective Microsoft Excel training.
This tip (9197) applies to Microsoft Excel 97, 2000, 2002, and 2003. You can find a version of this tip for the ribbon interface of Excel (Excel 2007 and later) here: **Concatenating Values from a Variable Number of Cells**.

**Solve Real Business Problems** Master business modeling and analysis techniques with Excel and transform data into bottom-line results. This hands-on, scenario-focused guide shows you how to use the latest Excel tools to integrate data from multiple tables. Check out *Microsoft Excel 2013 Data Analysis and Business Modeling* today!

When working with large amounts of data, it is a good idea to make sure that the data all consistently follows a pattern. ...

Discover MoreNeed to figure out if a value is outside of some arbitrary limit related to a different value? There are a number of ways ...

Discover MoreWith a long list of items in a worksheet, you may want to determine the last time a particular item appeared in the list. ...

Discover More**FREE SERVICE:** Get tips like this every week in *ExcelTips,* a free productivity newsletter. Enter your address and click "Subscribe."

Got a version of Excel that uses the
menu interface (Excel 97, Excel 2000, Excel 2002, or Excel 2003)?
**This site is for you!** If you
use a later version of Excel, visit
our *ExcelTips* site focusing on the ribbon interface.

**FREE SERVICE:** Get tips like this every week in *ExcelTips,* a free productivity newsletter. Enter your address and click "Subscribe."

Copyright © 2018 Sharon Parq Associates, Inc.

## Comments