Dynamic Subtotal of column after each row entry has been multiplied by two-way lookup

[PrettyGood_Not Great]

Hi, I have a challenge to produce a column total above the header row, where the total is the sum product of the entire row, after each entry has been multiplied against the rate table. Currently showing in row A of the data table is the double XLOOKUP to find the rate.

Thanks to @Fluff for the solution in column H where this operation is being performed for the rows. The challenge here is to have the same results for the column totals, BUT it needs to produce a subtotal when filtering on the Charge Codes (column I). Anyone know a 365 solution to this?

book1.xlsm
	A	B	C	D	E	F	G	H	I	J	K	L	M	N	O	P	Q	R	S	T	U	V	W	X	Y
1										Subtotal-->	192.5	192.5	192.5	192.5	192.5	192.5	192.5	192.5	192.5	192.5	192.5	192.5	198.3	198.3	198.3
2	RateCode	2023	2024	2025	2026	2027		SUM Result	Charge Code	Skill/Rate	Jan-24	Feb-24	Mar-24	Apr-24	May-24	Jun-24	Jul-24	Aug-24	Sep-24	Oct-24	Nov-24	Dec-24	Jan-25	Feb-25	Mar-25
3	R1	186.9	192.5	198.3	204.2	210.3		269685.2	A101	A1R1	108	120	150	114	114	138	75	75	90	71	75	75	68	75	47
4	R2	131.9	135.9	139.9	144.1	148.5		186573.3	A102	B1R2	81	120	150	114	114	138	75	75	90	71	75	75	68	75	47
5	R3	106.0	109.2	112.4	115.8	119.3		73516.2	A103	C1R3	41	45	56	43	43	52	45	45	54	43	45	45	41	45	28
6	R4	82.1	84.5	87.1	89.7	92.4		90737.4	A104	D1R4	68	75	94	71	71	86	75	75	90	71	75	75	68	75	0
7	R5	47.3	48.7	50.1	51.6	53.2		47005.4	A105	E1R5	61	68	84	64	64	78	68	68	81	64	68	68	61	68	0
Sheet1

Cell Formulas
Range	Formula
K1:Y1	K1	=XLOOKUP(RIGHT($J3,2),$A$3:$A$7,XLOOKUP(YEAR(K$2),$B$2:$F$2,$B$3:$F$7))
H3:H7	H3	=SUM(MAP($K$2:$Y$2,K3:Y3,LAMBDA(a,b,b*XLOOKUP(RIGHT($J3,2),$A$3:$A$7,XLOOKUP(YEAR(a),$B$2:$F$2,$B$3:$F$7)))))

 

[KRice]

I doubt the spilled range reference is the problem. Can you post the version of the formula you are using (overwrite the sheet name with something generic if it is sensitive), but try to duplicate as closely as possible what your formula looks like.

Some behind-the-scenes diagnostics might help. Go the the cell with the issue, and in the formula bar, use the cursor selection to highlight this part of the formula:

FILTER($P4:$P8,MAP($P$4:$P$8,LAMBDA(r,--(SUBTOTAL(103,r)=1))))

...and then hover the cursor over that highlighting to see a preview of what the returned array looks like, or alternatively, hit F9 to insert that array directly into your formula for closer examination (But do not hit Enter because you don't want to permanently commit/hardwire this array into the formula). Does it appear to be correct? You should see a trimmed down array showing only the visible Charge Codes in the P columns...those that are to be considered in the lookup table. Then hit Esc to revert back to your original formula.

Next, repeat the highlight/selection steps to investigate the array returned by:

ISNUMBER(MATCH($B$4:$B$8,FILTER($P4:$P8,MAP($P$4:$P$8,LAMBDA(r,--(SUBTOTAL(103,r)=1)))),0))

You should see an array of TRUE/FALSE and it should reflect which rows in the lookup table have Charge Codes of interest...and any FALSE's should correspond to Charge Codes that are filtered out.

Repeat the examination of:

($K$4:$K$8=R$3)

...does the Primary Criteria array appear to be correct? Any FALSE's should correspond to rows that do not meet the match criteria.

And does

($N$4:$N$8)

have all numeric values, or is there text or something else non-numeric somewhere in that range?

 

[PrettyGood_Not Great]

I doubt the spilled range reference is the problem. Can you post the version of the formula you are using (overwrite the sheet name with something generic if it is sensitive), but try to duplicate as closely as possible what your formula looks like.

Some behind-the-scenes diagnostics might help. Go the the cell with the issue, and in the formula bar, use the cursor selection to highlight this part of the formula:

FILTER($P4:$P8,MAP($P$4:$P$8,LAMBDA(r,--(SUBTOTAL(103,r)=1))))

...and then hover the cursor over that highlighting to see a preview of what the returned array looks like, or alternatively, hit F9 to insert that array directly into your formula for closer examination (But do not hit Enter because you don't want to permanently commit/hardwire this array into the formula). Does it appear to be correct? You should see a trimmed down array showing only the visible Charge Codes in the P columns...those that are to be considered in the lookup table. Then hit Esc to revert back to your original formula.

Next, repeat the highlight/selection steps to investigate the array returned by:

ISNUMBER(MATCH($B$4:$B$8,FILTER($P4:$P8,MAP($P$4:$P$8,LAMBDA(r,--(SUBTOTAL(103,r)=1)))),0))

You should see an array of TRUE/FALSE and it should reflect which rows in the lookup table have Charge Codes of interest...and any FALSE's should correspond to Charge Codes that are filtered out.

Repeat the examination of:

($K$4:$K$8=R$3)

...does the Primary Criteria array appear to be correct? Any FALSE's should correspond to rows that do not meet the match criteria.

And does

($N$4:$N$8)

have all numeric values, or is there text or something else non-numeric somewhere in that range?

Here is the actual sheet in question. The equation is in cell M8. I am looking into the details as well
EDIT
Removed as contain sensitive data.

 

[KRice]

Note that the three major terms (arrays to be multiplied) in the SUM function should have the same dimensions…yours do not.

 

[PrettyGood_Not Great]

I doubt the spilled range reference is the problem. Can you post the version of the formula you are using (overwrite the sheet name with something generic if it is sensitive), but try to duplicate as closely as possible what your formula looks like.

Some behind-the-scenes diagnostics might help. Go the the cell with the issue, and in the formula bar, use the cursor selection to highlight this part of the formula:

FILTER($P4:$P8,MAP($P$4:$P$8,LAMBDA(r,--(SUBTOTAL(103,r)=1))))

...and then hover the cursor over that highlighting to see a preview of what the returned array looks like, or alternatively, hit F9 to insert that array directly into your formula for closer examination (But do not hit Enter because you don't want to permanently commit/hardwire this array into the formula). Does it appear to be correct? You should see a trimmed down array showing only the visible Charge Codes in the P columns...those that are to be considered in the lookup table. Then hit Esc to revert back to your original formula.

Next, repeat the highlight/selection steps to investigate the array returned by:

ISNUMBER(MATCH($B$4:$B$8,FILTER($P4:$P8,MAP($P$4:$P$8,LAMBDA(r,--(SUBTOTAL(103,r)=1)))),0))

You should see an array of TRUE/FALSE and it should reflect which rows in the lookup table have Charge Codes of interest...and any FALSE's should correspond to Charge Codes that are filtered out.

Repeat the examination of:

($K$4:$K$8=R$3)

...does the Primary Criteria array appear to be correct? Any FALSE's should correspond to rows that do not meet the match criteria.

And does

($N$4:$N$8)

have all numeric values, or is there text or something else non-numeric somewhere in that range?

So for some reason after using cursor to select the portions of equations via formula bar, when I hover over the selection nothing appears. So I used the error checking to show calculation steps. For the SUM it's all long string of zeros followed by a long string of #N/As. For the ISNUBMER(MATCH() all of the charge codes are being returned.

N4:N8 is either a positive value or zeros. I can't get a read on K4:K8.

 

[PrettyGood_Not Great]

Note that the three major terms (arrays to be multiplied) in the SUM function should have the same dimensions…yours do not.

They did when I started, excel changed this when I selected Include Cell references. I have changed it back a couple times and no effect

 

[KRice]

The #N/A’s are probably due to the dimensional mismatch I mentioned. You’re multiplying a 13000+ row array by other arrays that are only about 1300 rows long.

 

[KRice]

It can be tricky sometimes to highlight the correct part of a formula. Usually, for me, the problem occurs when I don’t get the matched parentheses selections just right. And hovering sometimes doesn’t work if the cursor strays too far from the selection. Hit F9 in those cases…but be sure to hit Esc when you’re done looking.

 

[PrettyGood_Not Great]

The #N/A’s are probably due to the dimensional mismatch I mentioned. You’re multiplying a 13000+ row array by other arrays that are only about 1300 rows long.

I think your right about that, but i can't confirm. once I changed the range back to 1300 the "show calculation step" option disappeared. This is what the error box is providing for options, if this is of any help.

 

[KRice]

Those types of error messages are often not really errors. Excel tends to flag cells with green triangles suggesting that maybe there is an error if it seems as if a formula should be extended down a column or across a row.

Do you think that cell might factor into the problem? I can’t tell where that cell is located on the worksheet.

 

[PrettyGood_Not Great]

It can be tricky sometimes to highlight the correct part of a formula. Usually, for me, the problem occurs when I don’t get the matched parentheses selections just right. And hovering sometimes doesn’t work if the cursor strays too far from the selection. Hit F9 in those cases…but be sure to hit Esc when you’re done looking.

ok i'm with you now.

For Filter - all results are TRUE.

For ISNUMBER I am getting all the charge codes, plus some zeros at the end (screen shot below)

I think the problem is here: ($K$4:$K$8=R$3) as all results are FALSE