Visualizing Commonwealth games performance – Interactive chart

Share

Facebook
Twitter
LinkedIn

The 2018 edition of Commonwealth games are on for a week now. Both of my homes – India and New Zealand have done so well. Naturally, I wanted to gather games data and make something fun and creative from it. Here is my attempt to amuse you on this Friday.

Looks interesting? Want to know how to make something like this on your own? Then read on…

1. Gather data thru live connection to gc2018.com

We want to set up a refreshable visualization. So the data will be fetched thru PowerQuery. All we need to know about medal standings, medalists and country participation data is available at gc2018.com website.

Latest Medal Standings: This is available at https://results.gc2018.com/en/all-sports/medal-standings.htm page as an HTML table (the first table on the page).

So we can get the data using below M:

= Web.Page(Web.Contents("https://results.gc2018.com/en/all-sports/medal-standings.htm")){0}[Data]

This is extracting data column of first row of the Web.Page.

Number of participants by country: There is no one page where this information is available. Instead you need to visit each country’s page on gc2018.com to get the data from participants table. For example, the page for Bermuda (available at https://results.gc2018.com/en/all-sports/entries-bermuda.htm) looks like this:

Fortunately, all URLs follow the same pattern. https://results.gc2018.com/en/all-sports/entries-<country name>.htm

So given a URL in column [URL], we can use this custom column formula to get the total number of participants.

=Table.SelectRows(Web.Page(Web.Contents([URL])){0}[Data], each ([Discipline] = "Total")))

This is extracting the first table on URL and then filtering it for Total Row. 

Arranging everything in one table:

We can use a bit of built-in operations in Power Query to arrange all the necessary data in one tidy table. I am not going to explain all steps, but here is the final output. Try to come up with this on your own.

Now that our medal data is in Excel, in a table named medal_standings, let’s go ahead to next step.

2. Calculations to show medal standings by any criteria

The calculation engine for our little medal standings has few key things:

  • Fetching and sorting by a column
  • Slicer selection for sort options (Gold, Silver, Bronze, Total Medals, M/P)

Let’s go thru them:

Fetching and sorting by a column:

We would like to see countries by Gold, Silver, Bronze, Total medals and Medals per participant.

As per our medal_standings table, these are 3,4,5,6 and 9 respectively.

Assuming the column we want to sort is given by a named range – sort.option.num, we can use INDEX formula to fetch values, like this:

=IFERROR(INDEX(medal_standings,<row number>,sort.option.num),-10)

The -10 ensures that if we poll for a row that doesn’t exit, we get a negative value rather than 0. As some countries have 0 medals, having negative ensures that when sorting such rows are always at bottom.

Once we fetch a column, you can use LARGE() to re-order them top to bottom.

As there will be ties (few countries getting same number of medals), we can use de-duplication logic. This is when you add a very small unique fraction to each row before calling LARGE(). It is an elegant way to deal with ties and overcome Excel’s lookup formula limitations of returning only first match. See this decade old post by Robert discussing deduplication technique.

After this, just re-arrange original data (only columns needed for output) using another set of INDEX formulas.

A slicer to allow user to pick sort option

Now let’s just link up sort.option.num to a slicer so user can tell calculation engine what the sort order should be.

Start by making a pivot from a range like this:

But when you add a slicer on sort option, you realize the folly of your plan. The slicer buttons are out of order.

Technically, they are in order – alphabetical. But that is not what we want. We want them in the order – Gold, Silver, Bronze, Total Medals and M/P.

So what now?

Simple, we can ask Jackie Chan to karate chop the slicer and re-arrange it.

Alas, my summonJackie() macro was subscript out of ranging. So we need something else.

So we cheat Excel. We can pre-fix empty spaces – CHAR(129) to the slicer items. Since these are empty spaces, we just add 1 space for Gold, 2 for Silver etc. and make a slicer from these new values.

Note: In Power BI, you can simply order the sort option column by index number and that will fix the slicer problem. In fact, we wouldn’t bother with a slicer as Power BI tables are sortable by clicking on header.

That is better. Now simply style it and give it a buzz cut and you get this.

Related: Comprehensive guide and tutorial on all things Slicer – MUST READ

 

3. Preparing the viz

Now that everything we need is ready, simply bring calculated table to a blank worksheet (using Copy, Paste links) and arrange it in a neat table. Add Conditional formatting > Databars on medal and M/P columns. Position slicer neatly where these columns headers should be and you are gold.

Every now and then press Ctrl+Alt+F5 and go make a cuppa. When you are back, the medal table would be updated. Of course, come 16th of April 2018, there is no need to refresh it as the games would have ended.

Download the Commonwealth 2018 games medal tracker

Click here to download the Excel file. Play with it to learn more. Examine the query definition, control sheet and viz sheet to understand how it is put together. Make changes to the query (but duplicate it first, otherwise you will break the calculations) to fetch other data and make your own charts.

Ways to enhance this – adding past performance etc.

You can use the data from https://thecgf.com/ (Commonwealth Games Federation) to see historical performance and contestant data. They do not yet have 2018 values (as the games are ongoing) but you can see how countries have done in 2014 or 2010. Or you could combine this with performance in Olympics. How about combining this with demographic and well-being data (Gini scores or HDI ranks)? There are several ways you can mash-up this.

Love Games and Excel? Check out these visualizations

Me too. I like sport and I like data. Guess what, I build a lot of charts and cool visualizations on sport. Check out below and have fun.

Want to learn how to build awesome worksheets – Check out 50 ways to analyze data course

If you liked this, you are going to love our 50 ways to analyze data online class. This program helps you analyze and visualize business data in myriad ways. Learn all about statistical analysis, financial analysis, analytical modeling, data sciency stuff (clustering, outlier detection, optimization etc.) from the comfort of your office or home. Next batch enrollments will begin soon.

Visit 50 ways course page to know more and join the waiting list.

 

Facebook
Twitter
LinkedIn

Share this tip with your colleagues

Excel and Power BI tips - Chandoo.org Newsletter

Get FREE Excel + Power BI Tips

Simple, fun and useful emails, once per week.

Learn & be awesome.

Welcome to Chandoo.org

Thank you so much for visiting. My aim is to make you awesome in Excel & Power BI. I do this by sharing videos, tips, examples and downloads on this website. There are more than 1,000 pages with all things Excel, Power BI, Dashboards & VBA here. Go ahead and spend few minutes to be AWESOME.

Read my storyFREE Excel tips book

Overall I learned a lot and I thought you did a great job of explaining how to do things. This will definitely elevate my reporting in the future.
Rebekah S
Reporting Analyst
Excel formula list - 100+ examples and howto guide for you

From simple to complex, there is a formula for every occasion. Check out the list now.

Calendars, invoices, trackers and much more. All free, fun and fantastic.

Advanced Pivot Table tricks

Power Query, Data model, DAX, Filters, Slicers, Conditional formats and beautiful charts. It's all here.

Still on fence about Power BI? In this getting started guide, learn what is Power BI, how to get it and how to create your first report from scratch.

60 Responses to “Excel to the Next Level by Mastering Multiple Occurrences”

  1. Desk Lamp says:

    My most often used variation of this is to remove blanks from a list.

    Suppose column A contains information but some of the rows are blank. I want to return a continuous list of information without the blanks so I do...

    Your original formula looks like this:
    =IFERROR(INDEX($B$1:$B$20,SMALL(IF($A$1:$A$20=$E$2,ROW($A$1:$A$20)),ROW()-2),1),"")

    I want to look for non-blanks and all my data is in column A so I change it to:
    =IFERROR(INDEX($A$1:$A$20,SMALL(IF($A$1:$A$20"",ROW($A$1:$A$20)),ROW()-2),1),"")

    Ctrl+Shift+Enter, fill down and ta-da! A nice continuous list of information without any blanks.

    • Desk Lamp says:

      =IFERROR(INDEX($A$1:$A$20,SMALL(IF($A$1:$A$20"",ROW($A$1:$A$20)),ROW()-2),1),"")

      The original post chopped out my 'does not equal' for some reason. This is how it should look

    • Ajit says:

      Hi Sir,

      I am not able get any value by using below formula.
      =IFERROR(INDEX(DeliveriesMaster!$H:$H,SMALL(IF(Criteria!$A$3=DeliveriesMaster!$A:$A,ROW(DeliveriesMaster!$H:$H)-7,""),ROW()-3)),"")

      I want try

      help me

  2. Jeff Weir says:

    Great stuff. I laughed. I cried. I hurled.

    Personally I would use a PivotTable and Gordon Ramsay. But hey...as long as we cook the books, then each to their own, I'd say.

  3. Elias says:

    I won’t recommend the use of ROW()-2 because everything gets mess if you insert a row(s) before the row 2. The alternative would be ROWS(E$3:E3).

    Regards

    • Husain says:

      Hi Elias,
      I tried doing what you have suggested here.
      Ading any additional row messes up everything like you siad. But using the formula that you have suggested, shows only one value for the entire array. Would you please help me undersatand your method. I feel I may not be doing it correctly.
      Regards

  4. Denice says:

    Thank you Sohail. Great post. The comments are also very helpful.
    PS! Jamie Oliver was a great choice.

  5. Mando says:

    I've been using data with multiple occurrences for awhile now, and was glad to see the question I've been trying to ask and don't know how finally got answered. Now if I can be brave enough to use this, is another question.
    What I usually do is just add another column to the end of my data =IF((COUNTIF($B$2:B2,B2))=1,1,"") where B is my unique identifier and then just do multiple COUNTIFS with it.
    For multiple Occurrences and Criterias, I just add another column to Concatenate my unique identifier and the other criteria =$B2&" "&$C2, then add another column using the same =IF((COUNTIF($B$2:B2,B2))=1,1,"") but this time use the column where I placed the concatenated data.
    Any ideas how to lessen the number of columns I use without using any Arrays or VBA's?

    • Sohail Anwar says:

      Hi Mando,
      Are you pretty much asking for an alternative way to do this without VBA/Array Formulas? If so, I would recommend not doing that, Arrays make things a bit easier. The method you wrote looks like it will increase work, I'm always in search of efficiency in the long term 🙂

  6. XOR LX says:

    It's both illogical and unnecessary to use a construction for SMALL's (or LARGE's) k parameter which consists of the ROW function (either in its unqualified form, i.e. ROW(), or with a reference, e.g. ROW(A1)) +/- some constant.

    Not only is such a construction necessarily dependent upon the row number in which the user decides to place the initial formula in the series, but it is also susceptible to error upon row insertions within the sheet.

    ROWS (i.e. ROWS($1:1), or ROWS(A$1:A1) if you prefer) gives precisely the same results, though suffers from neither of these two drawbacks:

    http://excelxor.com/2014/08/25/row-vs-rows-for-consecutive-integer-generation/

    Regards

  7. […] Excel to the Next Level by Mastering Multiple Occurrences: Microsoft Excel cleverness. […]

  8. Sohail Anwar says:

    @Elias and XOR LX, great point and while I use the construct you mentioned in other things, I never really gave it too much thought since I owuldn't readily insert rows in this sort of thing.

    I love the rule of ROW(A1) +/- constant being illogical! Any time I can eliminate something from my arsenal due to redundancy is good. Much appreciated and once again this sort of exchange is precisely why we love Chandoo 🙂

  9. Jonathan James says:

    Great post, love this way of retrieving lists of items. Will certainly be giving this a go.

  10. Mr J says:

    I like this technique a lot and *will* be using it. However how can it be done in 2D. E.g I have a 3 by four table (12 items) and each items is either an "Apple" or an "Orange". I want to get the row and column position of each occurrence of "Apple" and of "Orange"? How would I do this?

  11. XOR LX says:

    @Mr J

    When you say "row and column position", do you mean relative positions or absolute? For example, if your table was in A10:D12, and the first occurrence of "Orange" was in cell B11, would you want 11 (absolute) or 2 (relative) returned for the row position?

    Regards

  12. Shakil says:

    The master database contain name, designation, salary, passport no, expiry date, joining date, project no. camp name, floor no., flat no., room no., around 20 more column, and this is more than 500 staff member.

    i want to make report for the camp and i want use the employee ID to transfer their name, designation, flat no., and their room no only to other sheet using VBA code.

    Please help me.

    Thanks

  13. Gk says:

    Great post, love this way of retrieving lists of items.

  14. BeEman says:

    This was a great post and I learned a lot. i am attempting to do exactly what this post was about with the exception of direction, i want to go across not down. is this possible?

  15. Skrattoune says:

    To summarize for those who will not take the time to go through the whole comments list (and who therefore will avoid some brain overload and save some grey cells), use at the end of your array formulas

    ROWS($1:1) instead of ROW()-2

    it additionally is more intuitive for understanding the formula:
    ROWS($1:1) => displays 1st result
    ROWS($1:3) => displays 3rd result
    ...

    Thanks all for this posts & comments

    Skrattoune

    • shehan says:

      in the Multiple Occurrences fomula, we couldnt get the second line since its not appear, but when we check your file, i saw there is {} brackets before equal but when we extract it we couldnt see it. how to do that?

  16. Srinivasan says:

    Difficult to understand
    But I am sure it will be of immense use to me

  17. CJS says:

    Very useful post. I worked with the downloadable workbook and did some experimenting to see how each part of the formulas worked. Although I understood most of it, I have a question. What if I wanted the results of my search for each person to be listed by column instead of by row?

  18. Mia says:

    Hi all,
    thanks for the contribution, it helped a lot.

    But what if I need to get the average of the multiple values I get?
    Is there a way to get the average of these multiple values directly (without listing them beforehand...my sheet is already busy)?

    thanks a lot.

  19. Matt says:

    What changes would you make to allow these multiple values to be horizontal rather than vertical, as shown?

  20. Diane says:

    Mr. Doo, you are so funny! I did not know the multiple occurrences could be done without a (trial and error) macro.
    You make it fun to make a complicated task a Can - Do ! Thanks!

  21. Julia says:

    Hi,
    It looks super helpful.
    However, whatever I do it feels I'm almost there... but every time it's a mirage.
    I've a (very) big data table consisting of multiple parameters (about 10) for every value in column A. A problem - same A value may (or may not) appear multiple times in my big table. Luckily, the repetition is always in clusters - one after another (and after the cluster ends, there is no more same A).
    The goal - I've a subset of data consisting of arbitrary values of column A (each one repeats only once), and I want to get all the parameters for all them (including for the as much as there is same A values). With you function, it fills nicely automatically for only the first A, but only once (without considering multiple occurrence), and then jumps to the next one.
    Is there a way to solve this (without tediously manually inserting N rows number for N A's)? I prefer not using macro's.
    Thank you,
    Julia

  22. JB says:

    Does anyone know how to summarise the following data to return the record vertically under the expected result?

    Much appreciated ...

    Data is from A1 to D3
    Name "Asset Name#1","Asset Name#2","Asset Name#3"
    ABC Asset 1 Asset 2
    ZXY Asset 1

    Expected Result:
    Name: Asset Name
    ABC Asset 1
    ABC Asset 2
    ZXY Asset 1

  23. HKM says:

    Hi

    What if I have multiple criteria I need to do this for? So in your example, instead of just "Tom Yorke", I had a list of first and last names I needed to identify all instances of in a larger file. How would I go about doing that? Thanks!!

  24. Prasand says:

    Hi,
    I have 2 sets of name lists in a spreadsheet and need to find whether the same set of names repeat in the consecutive rows. can anyone please help me.

  25. gopal sharma says:

    hi dear
    i have a list of persons(First name space last name) in column A. multiple values are equal to first name and last name. ie. A kumar, b kumar alok das, alok ranjan. now i want multiple entries of all matching first name or second name as per my choice, what is the solution.

  26. Arvind Kumar says:

    Hi,
    I have 10 rows. in row 1 there are multiple columns. in few colums some values are present. just i wants to count the coulmn number of first record. how do i get it ?

    example

    A B C D E F G H I J
    10 13 19 12 --> here number 10 position is 3
    11 2 5 8 --> here number 11 position is 1
    23 45 48 --> here number 23 position is 2

    • Hui... says:

      @Arvind
      Try:
      =INDEX(COLUMN(A1:E1),MATCH(TRUE,INDEX(A1:E1<>0,),0)) Ctrl+Shift+Enter

      Copy down

      Change Column E to match the last column of your data

  27. Samy MREJEN says:

    Hi
    I wonder if you have any tutorial (preferably in video format) concerning your technique of sorting a data table in a dashboard based on user choice control button
    Thank you

  28. David says:

    I am trying to subscribe, but I not getting the confirmation email.
    I have tried it few times but its not working.
    My email is muntoo76@hotmail.com

  29. Saskia says:

    Great post! Thanks for presenting a solution to a problem I had. However, how do I expand this to search across multiple worksheets? Thanks!

  30. Well done says:

    Just to say that you have been the only person I've found to bother explaining the rationale behind your function choices. There were other articles on the internet where people didn't bother to make the effort. Many thanks.

  31. Peter says:

    Is there a text character limit to this formula? It works when I enter a few sentences, but not when I have 10 sentences.

    • Hui... says:

      @Peter
      I don't believe so
      There maybe with pre-2007 versions of Excel
      Can you post a sample data

      • Peter says:

        this is the formula I'm running:

        =IFERROR(INDEX(Input!$A$1:$R$201,SMALL(IF(IFERROR(SEARCH($E$2,Input!$D$1:$D$201)>0,FALSE),ROW(Input!$D$1:$D$201)),ROW()-5),COLUMN()),"")

        and when I have this text paragraph on the sheet I'm pulling from, it won't pull in:

        "We do need a fair amount of analysis in advance of the meeting. Let's start with a sensitivity analysis at plan value under various assumptions in terms of what lenders take - say 50% up to 100% in 5% increments. Need to understand dilution at various points to each side as we negotiate. If we can get that in the next hour or so, we can figure out what else would be helpful to negotiations. "

        But when I shorten it to:

        "We do need a fair amount of analysis in advance of the meeting. Let's start with a sensitivity analysis at plan value under various assumptions in terms of what lenders take - say 50% up to 100% in 5% increments."

        It works then..

  32. Matthew Edwards says:

    I like your work. the tread has been very informative.
    What I am trying to do get the multiple occurrences fill in columns not rows. AKA while you example has results in a the following format:
    Thom Yorke
    3
    8
    10
    12
    18

    I want the result to be
    Thom Yorke 3 8 10 12 18

    Can you assist with this change?

  33. DJ Pinguim says:

    Great work in this article! Very well explained!

    But i need some help...

    I want to use the Multiple Occurrences and Multiple Criteria with the Partial Text Search.

    Example:
    1st criteria: G11
    2nd criteria: Varnish
    3rd criteria: 1503/5

    And i want to use in the 3rd criteria only the "1503" to seeach 1503/5, 1503/6 and 1503/7.

    Can you help me with this issue?

  34. Hi chandoo, thanks for your wonderful work.

    I am in stuck to find a solution to extract multiple rows (by using index+ small+ if) and extract the multi columns to its rows.(multicolumn data should be combined as single).
    I repeated the index function three time to get three column's data and combine it with wild character and got the required answer. But feel this can be done in better way. so Could you please help to simplify the below formula in alternative way.

    {=IFERROR(INDEX(Table1,SMALL(IF(Table1[Tag trim]=LEFT(F75,8),ROW(Table1[Tag trim])-1),1),COLUMN(Table1[MAX. LENGTH (mm)
    (22)]))&" X "&INDEX(Table1,SMALL(IF(Table1[Tag trim]=LEFT(F75,8),ROW(Table1[Tag trim])-1),1),COLUMN(Table1[MAX. WIDTH (mm)(24)]))&" X "&INDEX(Table1,SMALL(IF(Table1[Tag trim]=LEFT(F75,8),ROW(Table1[Tag trim])-1),1),COLUMN(Table1[HEIGHT (mm)
    (23)])),"")}

  35. Jose says:

    Hi. Your help in excel is great. It has being very helpfull in a project I am working on.

    I got a question about Multiple Occurrences: I am trying to get all different values from the a same date and return values horizontally.
    It ls like this:

    Date provider
    June 2 A
    June 2 A
    May 3 A
    May 3 A
    May3 B
    April 4 B
    April 4 B
    April 4 B
    April 4 C
    April 4 C
    April 4 A

    Could you please help me with the formula?

  36. Jarek says:

    I've got a lot of hints from this post and was able to get almost there with my task but there is one problem - string length. I have a long list of stuff given in consequtive columns. I need to peak certain type of data (long string) and put them together in one cell. The text type comes after the text, so schematically one raw of the data looks like this (where Ty My Wy Oni etc is the Type and it repeats):
    Text_A Ty Text_B My Text_C Wy Text_D Oni Text_E Ja Text_F Ty Text_G My Text_H Wy Text_I Oni Text_J Ja Text_K Ty Text_L My Text_M Wy Text_N Oni Text_O Ja Text_P Ty Text_R My Text_S Wy

    What I want is "Text_A, Text_F, Tekst_K, Text_P" if the search="Ty"
    The following works if the string in Text_X is <256; if logner -forget it
    =TEXTJOIN(", ";TRUE;IF($C$4:$AL$4="Ty";$B$4:$AK$4;""))
    same with error handling
    =TEXTJOIN(", ";TRUE;IFERROR(IF($C$4:$AL$4="Ty";$B$4:$AK$4;"");""))

    Most of the Index - Small etc solutions take up several cells to work and that is not an option this time. Any hints, please?

  37. Ratish says:

    Hi Chandoo,

    I have been brainstorming this from past couple of months. I work in reporting team and during month end I pull all incident report which has changed priority from P1-P2-P3-P4, P2-P3-P4 or P3 to P4. Currently, I am performing it manually (4000+ count). Below is the sample excel where I would highlight in a different color if priority changes from P1-P2-P3-P4, P2-P3-P4 or P3 to P4. So basically I want to check column A if it has more than 2 similar value it should check the final priority in column B based on Column C's updated time and it should return value as P1-P2-P3-P4, P2-P3-P4 or P3 to P4 in Column D.

    Number Priority Start time
    INC0281369 Priority 2 2017-07-03 13:01:07
    INC0281369 Priority 4 2017-07-03 13:04:29
    INC0281696 Priority 3 2017-07-26 21:20:16
    INC0281696 Priority 4 2017-07-27 00:06:21
    INC0281962 Priority 3 2017-07-01 01:13:41
    INC0281962 Priority 4 2017-07-01 04:21:12
    INC0281974 Priority 3 2017-07-01 01:35:41
    INC0281974 Priority 4 2017-07-01 03:25:14
    INC0281976 Priority 3 2017-07-01 01:40:25
    INC0281976 Priority 4 2017-07-01 03:26:29
    INC0281985 Priority 2 2017-07-01 02:03:38
    INC0281985 Priority 3 2017-07-04 18:29:34
    INC0281987 Priority 2 2017-07-01 02:06:38

    Any help would be appreciated

  38. Nadeem says:

    You have done a great job, Bravo!
    I want the same result but my "Das hoff" is in multiple sheets. Can you please be kind enough to give me the formula to have the same output but the searches are in different sheets.

    Thanks in advance.

    Nadeem

  39. Christina says:

    Hi! Your instruction is great on this however I am still stuck with my formula. I revert back to INDEX/MATCH but I know my data is skewed. I really hope you can help!

    I am working with two worksheets, CREDIT _MEMO_ACCRUAL_MASTER & CM_12 - I will reference them as WS A& WS B.
    WS A is the master where my formula starts in column 15, row 2. My index/match is based on multiple criteria, Invoice # & Sku, to lookup the Original Invoice Date from Index sheet WS B. WS B only contains original invoice date, sku, credit date and amount.

    WS A:
    INVOICE# SKU RESULT FROM WS B
    139591 XYZ (BLANK)
    139612 ABC 12/11/2017

    Currently in "RESULT FROM WS B"
    =IFERROR(INDEX(CM_12!$B$2:$B$602,MATCH(CREDIT_MEMO_ACCRUAL_MASTER!B2&CREDIT_MEMO_ACCRUAL_MASTER!F2,CM_12!$D$2:$D$602&CM_12!$F$2:$F$602,0)),0)

    The trouble is this:
    WS B has reoccuring original invoice date and sku. In other words - invoice 139612 on credit date 11/30/2017 may have several different "original invoice dates" and 10 returned skus, therefore show up in 10 different rows.
    WB S:
    Invoice # Original invoice date Credit date SKU
    139612 08/08/2017 11/30/2017 1234
    139612 08/21/2017 11/30/2017 5678
    139612 08/30/2017 11/30/2017 1234

    I need a formula that will recognize the exact original invoice date for an invoice # and sku. Currently my index/match as you know only results in the first instance.

    I tried your index/small/if formula but it didnt work for me. index/small/if is very new to me so I am sure i was doing it wrong somewhere.

    I really hope you can help!
    Happy New Year!

  40. Geert says:

    Hi All,

    Great post, which I come back to multiple times !!

    Can anyone explain to me how to amend the formula when you want to either exclude (e.g. all the lines NOT concerning DAS HOFF) rather than select a certain value, or when you want to allow more than one value (e.g. the lines where DAS HOFF is linked to US or UK)

    Thanks for your help.

    Geert.

  41. Matthew Buhagiar says:

    Great post!

    How do I get the output of the multiple occurrences into another coloum instead of on the same row?

    Thanks

  42. Neerav Gupta says:

    Thanks for the aide. I have been using this formula but the step by step explanation you have given makes me understand now completely the inside chemistry as to what is happening. Keep it up.

  43. Robert says:

    Hi Chandoo

    I've replicated your exact spreadsheet and it works perfectly, thanks! For my actual application, I'm using a Named Table where:

    $B$1:$B$20 = Chandoo[PointlessThing]
    $A$1:$A$20 = Chandoo[Person]

    Replacing the fixed cell references with the Table[Column] values the array formula produces an output that is one cell below what the actual value is. For example, if my lookup value is Das Hoff with the named table I get Amnesiac, Raging, Limb King, Krautrock, Erasing. When I just use the cell references I get Talented, Knightrider, Baywatcher, SpongeBob, Krautrock. As you can see, outputs when using the named table are actually one row below the intended output.

    I've varied the formula, from completely deleting the -2 in ...ROW()-2, to trying 0-3. I can never get the named table formula to output the same results as the cell reference formula.

    I've noticed the lateral distance doesn't matter, only the relative horizontal distance, so for that reason my named table formula starts in cell E3, referencing E2 as the lookup value, and my cell reference formula starts in cell G3, referencing G2 as the lookup value. The Person/PointlessThing columns begin at A1 and B1. The table is named "Chandoo." So my named table references are Chandoo[Person] and Chandoo[PointlessThings].

    As a final note, I'm using data validation, referencing the Person column of the named table as my lookup values in cells E2 and G2.

    • Robert Lawson says:

      So I retried the formula with dragging ranges (which automatically populates the range name) and I got this:

      =IFERROR(INDEX(Chandoo[[#All],[PointlessThing]],SMALL(IF(Chandoo[[#All],[Person]]=$F$3,ROW(Chandoo[[#All],[Person]])),ROW()-2),1),"")

      And it works!

      Originally I was hand typing it to make sure I got it all right and was entering this:

      =IFERROR(INDEX(Chandoo[PointlessThing],SMALL(IF(Chandoo[Person]=$F$3,ROW(Chandoo[Person])),ROW()-2),1),"")

      As you can see, I was missing [#All] preceding the column reference.

      That said, this also works when referencing another sheet in the workbook, as long as the relative positions stay the same.

      What I've run into now is this: Where I want the multiple occurrences to appear are 'Visit Tear Sheet!F12:F16'

      The drop-down data validation is Visit Tear Sheet!F8

      The table location is 'Visit Log'B49:C148

      I've kinda buried the table at the bottom of a spreadsheet because I don't want non-tech saavy users to easily find it and screw it up. I know I could let it rest on a separate sheet starting at A1 like our sample data set, but I'm trying to keep the number of sheets to a minimum to keep the weight of the file down.

  44. Jimbob says:

    Instead of the results being in downward rows is it possible to put them in the next columns?

  45. Jim says:

    Have you ever had to do this using Power Query? Or, know of a way to do something similar, but using Power Query? I have a huge workbook that uses a method similar to yours, but it's way to slow using the SMALL and ROW formula so I'm trying to speed it up, but by using PQ. Thank you so much in advance for any help!

  46. Christian says:

    Can this method work with wildcards? I tried but I could not find a way to make it work.

Leave a Reply