In Excel I have a list of employees (see image) and I want to store their salary update history. For each salary update, I want to calculate the difference in their total salary. I'm unable to create a formula to calculate this and SUM() the updates per period for me.
The image below demonstrates my spreadsheet and the desired salary update total (I did calculations manually, bet need a formula for this). In reality, there will be a few dozens of employees, so the requirement for the formula is to be dynamic and not contain manual "+" for each row.
NOTE: For some people their salary can stay unchanged and there are not any values in some cells. It's the main challenge, since we have to find the value of the "previous salary" for an employee, because it's not necessary stored in the previous cell. Otherwise, I could have used a simple array formula, like:
{=SUM(G3:G6-H3:H6)}. But unfortunately, it does not work in this case (there can be no value in the H column, so we have to find where it actually is).
Also, the formula should work correctly with filtering: is some row is filtered out (person is inactive, see picture), this row/cell should be ignored.
Can anybody help me with this formula that I'm struggling with?
This will do what you ask, not my my best work.Column "I"is still =SUM. Remember to press Ctrl + shift + enter. Then drag the formula where you want it to apply.
{=IFERROR(IF(E3="",E3,E3-INDEX(F3:$I$3,MATCH(FALSE,ISBLANK(F3:$I$3),0)))
+IF(E4="",E4,E4-INDEX(F4:$I$4,MATCH(FALSE,ISBLANK(F4:$I$4),0)))
+IF(E5="",E5,E5-INDEX(F5:$I$5,MATCH(FALSE,ISBLANK(F5:$I$5),0)))
+IF(E6="",E6,E6-INDEX(F6:$I$6,MATCH(FALSE,ISBLANK(F6:$I$6),0)));SUM(E3:E6))}
This formula does give the same answers as yours on the test data as below, BUT it is making a very big assumption - that the last recorded salary is always the largest one.
=SUM(IF(D3:D6="",0,D3:D6-SUBTOTAL(4,OFFSET(E3,ROW(D3:D6)-ROW(D3),0,1,COLUMNS(E3:$J3)))))
Must be entered as an array formula using CtrlShiftEnter
If you want to ignore hidden rows, plz use the form of Subtotal that ignores hidden cells
=SUM(IF(D3:D6="",0,SUBTOTAL(104,OFFSET(D3,ROW(D3:D6)-ROW(D3),0,1,1))-SUBTOTAL(104,OFFSET(E3,ROW(D3:D6)-ROW(D3),0,1,COLUMNS(E3:$J3)))))
Related
I'm making a summary of a list of tasks, and the corresponding dates (start date, first answer date, ...).
It looks more or less like the following:
Title Start date First answer
Task1 29/06/2018 02/07/2018
Task2 09/05/2018
Task3 13/06/2018 14/06/2018
I would like to calculate the average time, needed for the first answer to be given. In case no first answer is given yet, this entry needs to be ignored in the calculation of the average.
In order to be able to understand the formulas better, I've decided to use names for the headers, like:
Name "Header_Title" has value "Title"
Name "Header_Start_Date" has value "Start Date"
Name "Header_First_Answer" has value "First answer"
Also, the number of entries, defined as COUNTA(OFFSET(Header_Title;1;0):A1048576) has a name: Total_Count.
Next to that, I've created names for the ranges of the column values:
"All_Start_Dates" is defined as =OFFSET(Header_Start_Date;1;0):OFFSET(Header_Start_Date;Total_Count;0)
"All_First_Answered" is defined as =OFFSET(Header_First_Answer;1;0):OFFSET(Header_First_Answer;Total_Count;0)
Explanation : take the first entry under the header (column title) and go to the row, corresponding to the last task.
This makes it very easy to write a formula for calculating the average difference between those date columns:
{=AVERAGE(All_First_Answered_Dates-All_Start_Dates)}
// mind the {} for showing this is an array formula
Now the problem is: how to use the AverageIf worksheet function in order not to take into account the cases where First answer is not filled in?
I already tried using ">0" and ">"&0, but this does not work, the formulas are said not to be valid:
{=AVERAGEIF(All_First_Answered_Dates-All_Start_Dates;">0")}
{=AVERAGEIF(All_First_Answered_Dates-All_Start_Dates;">"&0)}
Does anybody have an idea?
Thanks in advance
P.s.1. As you can see, I'm using cell range A1048576 as the last entry of column A, does anybody know a more elegant way to describe this?
P.s.2. One extra thing, which would make my life easier, is the possibility to see which cells have a name (I was thinking about conditional formatting, but I didn't find the way to do this). Does anybody know if there is a way to highlight individual cells, linked to a name?
So my suggested answer would be
=AVERAGEIF(All_First_Answered,">0")-AVERAGEIF(All_First_Answered,">0",All_Start_Dates)
I'm assuming here that all start dates are present but first answered dates may be missing: you could easily add an extra condition for start date if you used AVERAGEIFS. Both parts of the formula include the same conditions so they are working on the same rows.
The intermediate columns above are just included by way of explanation.
Try an array formula.
If you're not familiar with array formulas, the significant difference is you press and hold Ctrl+Shift then hit Enter instead of just pressing Enter. You will see the formula preceded and followed by curly brackets. Do not type those. Those will appear automatically.
=AVERAGE(IF(INDIRECT("C14:C"&LOOKUP(2,1/(A:A<>""),ROW(A:A)))=0,"",INDIRECT("C14:C"&LOOKUP(2,1/(A:A<>""),ROW(A:A)))-INDIRECT("B14:B"&LOOKUP(2,1/(A:A<>""),ROW(A:A)))))
I am looking to get some help with a function that I am sure is an option but I sadly have no clue on how to implement.
Basically, I'd like a formula to go from C21:C50 and look for the top two values. Based upon which two are the top, it would reference the name in B column and populate that value in the another cell (the cell the formula resides in)
If you look at the image, in the primary field, we'd have Steve. Secondary would be Alan.
Is this something anyone can help with? I simply am lost :(
Try
=INDEX($B$3:$B$7,MATCH(LARGE($C$3:$C$7,ROW(A1)),$C$3:$C$7,0))
with Bob in cell B3 and the "primary" formula in cell C9. Copy down to cell C10.
If dealing with integers, you can simply add +1/ROW([range]) to avoid doubles:
=INDEX($B$3:$B$7,MATCH(LARGE($C$3:$C$7+1/ROW($C$3:$C$7),ROW(A1)),$C$3:$C$7+1/ROW($C$3:$C$7),0))
This is an array-formula and must be confirmed with ctrl+shift+enter!
However, this may fail for numbers like 5.01 or 4.99. For that case just use it in combination with RANK.EQ:
=INDEX($B$3:$B$7,MATCH(LARGE(RANK.EQ($C$3:$C$7,$C$3:$C$7,1)+1/ROW($C$3:$C$7),ROW(A1)),RANK.EQ($C$3:$C$7,$C$3:$C$7,1)+1/ROW($C$3:$C$7),0))
This is an array-formula and must be confirmed with ctrl+shift+enter!
The steps as picture:
The first table shows the direct adding of 1/ROW which is used for LARGE and MATCH to get the row if doubles exist (so INDEX can pick the correct one)
The second table shows how the values get replaced by their rank with RANK.EQ and then are treated the same like the first table.
The third (grey) table shows, what would happen if the first formula is applied to the second table (to demonstrate how the ranks get messed up).
For Excel 2007 just replace the RANK.EQ($C$3:$C$7,$C$3:$C$7,1) with RANK($C$3:$C$7,$C$3:$C$7,1).
If you still have any questions, just ask :)
I am an excel beginner and I would like to do the following.
Let row1= (a_1 a_2 a_3) and row2= (b_1 b_2 b_3).
I want excel to calculate the largest number among the products (a_1b_1, a_2b_2, a_3b_3).
It is very difficult to look up these things for I am not sure what kind of calculation I am doing and it is hard to explain.
Take a third column, C and enter formula in C1 as $A1*$B1. Pull it down vertically to all other rows so that row number gets incremented for each.
Then in the fourth column, use the formula MAX(C:C)
The following formula, array-entered, gives you the result of the largest number among the products:
{=MATCH(A1:C1*A2:C2)}
(provided your data is in A1:C2 in the form you presented it).
For explanations on how to insert array formula in excel see e.g. this microsoft link; in short, you type the formula without the curly brackets and confirm with CTRL+SHIFT+ENTER instead of only ENTER.
If you want to find where this couple of numbers is (in your case: which column), I would try this:
{=MATCH(MAX(A1:C1*A2:C2);A1:C1*A2:C2;0)}
(also array-entered).
you can do that, or make a pivot with the raw data and get the MAX/MIN/AVG, based on the pivot options. I tend to use that instead and then vlookup the ID to the pivot to get whatever aggregate you need.
i have a little problem with final formulas in one of my column. How to start. maybe i will explain what i have a then what i want.
i have an excel worksheet with 3 sheets. i want to record goods and what are these goods made of. first is sheet called Goods where is possible to put number of goods i want to make. In this case i want to make 1x sandwich1 and at the same time 3x sandwich2. i dont want make sandwich3 this time.
Second sheet is Matrix sheet where I record every good and what it is made of. This sheet is basic sheet and all other sheets take list of goods (resp. ingredients) from this sheet. Simply when i want to make sandwich1 i look at matrix and know that i need 1x1pc of egg + 1x5g of cheese. And for 3x sandwiche2 i need 3x10g of sausages.
Final sheet is called Ingredients. It is a list of used ingredients from Matrix sheet (exactly same order) to make these sandwiches. I want to fill formulas into column B which would go through one ingredient ofter ingredient and count needed amount of it. So it would look into matrix in the same row and where there is some number it would multiply with number of items from Goods sheet. The list of goods is also in the same order as in the matrix sheet.
I hope you understand now what i want and will try to help me. I think there will be SUMPRODUCT, SUMIF and maybe INDERECT functions but i am not that skilled in excel
thanks for any suggestions
You can use MMULT function here - it's an "array formula" which you need to enter in a range. You can do that like this:
In Ingredients worksheet enter this formula in B2
=MMULT(Matrix!C2:E4+0;Goods!B2:B4+0)
[I'm assuming you have a European version of Excel where ; is used to separate arguments]
Now select the whole range B2:B4, press F2 key to select formula and hold down CTRL and SHIFT keys and press ENTER. This "array enters" the formula in the range and you should now see curly braces like { and } around the formula and also the correct results.
You cannot change part of that array now, only the whole thing
Note that I'm assuming that the contents of Goods!A2:A4 will be the same as Matrix!C1:E1 and in the same order. You can extend the ranges to be as large as you like as long as that principle still holds
I suspect that this is an issue of "when all you have is a hammer, every problem is a nail". For reasons known only to you you are using a spreadsheet to solve a problem that databases were made to do. Any solution to this problem in a spreadsheet will be entirely dependent on the integrity of your data - add another column or get things out of order and it will fail.
That said, what you have in your link is effectively a pivot table and what you need is the unpivoted version of this - the instructions for getting this are here.
When you have that, you can use the various database functions in excel to get your answer.
I am a noob in excel, hence pardon me for any mistakes made.
This question must have been answered before but I couldn't find the right string to make it work for me.
There are around 500 rows and 20 columns (Yes, it is a report)
Column A has a few values (eg: Problem, Change, Request, etc.)
Column B has ticket numbers assigned to each entry. (No, I don't work for a call center, these are Datacenter Operations tickets)
Column B has several duplicate ticket numbers, as many people worked on same ticket OR the ticket was reopened for some reason.
I wish to take a count of unique ticket numbers from Column B when the condition in Column A is Change only.
So if there are 500 ticket number 250 are duplicate for sure, and only 25% of the rest will be Change tickets.
I am not supposed to use a Pivot or filter hence asking this question.
Need a formula to retrieve the count with the condition.
I may put the formula in Sheet2 or at the extreme right column, plz don't worry about it, I will take care of those things.
Many thanks in advance.
Adding to the question,
Let me help you with some data.
Change CRQ1110001
Problem INC1110001
Change CRQ1110001
Problem INC1110001
Change CRQ1110003
Problem INC1110003
Change CRQ1110004
Problem INC1110004
Change CRQ1110004
Change CRQ1110004
Problem INC1110005
Now I wish to only consider Change here without considering duplicate values.
Maybe this helps.
Thanks again.
Based on this website count unique values and with a small change this formula should work, expand the ranges to cover your entire range.
=SUM(IF(FREQUENCY(IF(A2:A10="change",IF(LEN(B2:B10)>0,MATCH(B2:B10,B2:B10,0),""),""), IF(A2:A10="change",IF(LEN(B2:B10)>0,MATCH(B2:B10,B2:B10,0),""),""))>0,1))
Entered with ctrl+shift+enter as it's an array formula.
Note that if you do this over the entire column A:A it will take quite a bit of time to compute as it has to go through a lot of calculations in the array formula.
If your "ticket numbers" in column B are actual numbers then you can use this formula
=SUM(IF(FREQUENCY(IF(A2:A500="Change",B2:B500),B2:B500),1))
confirmed with CTRL+SHIFT+ENTER
If not numeric you need to use a version as per gtwebb's suggestion