Conditional formatting based on previous row value

[IuteJV]

Hello

I have a problem I have not been able to crack. I have an audit table, where I need to display different data points and highlight those data points (different columns) that have been changed, when compared to a previous row. The example data I have provided, let's sa that it consists of 4 different regions and within each of those regions there are two different products. Both of the products will be updated in the system at the same time, nevertheless if one of them was not changed (nothing can be done regarding that, the system was built like that).

Row Number (example purpose only)	Country	Valid from	Valid until	Product	Measure1-age	Measure2 - percentage	Measure3-amount
1	Country1	Fri, 02 Feb 2024 15:10:28		Product1	18	0%	5000
2	Country1	Fri, 02 Feb 2024 15:10:28		Product2	20	0%	300
3	Country2	Fri, 02 Feb 2024 15:10:28		Product1	18	0%	15000
4	Country2	Fri, 02 Feb 2024 15:10:28		Product2	20	0%	90000
5	Country3	Fri, 26 Jan 2024 09:16:40		Product1	18	0%	400
6	Country3	Fri, 26 Jan 2024 09:16:40		Product2	20	0%	1000
7	Country4	Thu, 25 Jan 2024 14:21:44		Product1	18	0%	3000
8	Country4	Thu, 25 Jan 2024 14:21:44		Product2	20	0%	400
9	Country1	Sat, 17 Sep 2022 21:53:47	Fri, 02 Feb 2024 15:10:28	Product1	18	10%	5000
10	Country1	Sat, 17 Sep 2022 21:53:47	Fri, 02 Feb 2024 15:10:28	Product2	18	15%	300
11	Country3	Mon, 18 Sep 2023 11:14:43	Fri, 26 Jan 2024 09:16:40	Product1	18	0%	500
12	Country3	Mon, 18 Sep 2023 11:14:43	Fri, 26 Jan 2024 09:16:40	Product2	18	10%	1200
13	Country2	Tue, 09 Jan 2024 12:52:01	Fri, 02 Feb 2024 15:10:28	Product1	18	5%	10000
14	Country2	Tue, 09 Jan 2024 12:52:01	Fri, 02 Feb 2024 15:10:28	Product2	18	0%	70000
15	Country4	Mon, 01 Jan 2024 00:00:00	Thu, 25 Jan 2024 14:21:44	Product1	18	0%	3500
16	Country4	Mon, 01 Jan 2024 00:00:00	Thu, 25 Jan 2024 14:21:44	Product2	18	0%	400
17	Country4	Tue, 31 Jan 2023 16:25:35	Mon, 01 Jan 2024 00:00:00	Product1	18	0%	3500
18	Country4	Tue, 31 Jan 2023 16:25:35	Mon, 01 Jan 2024 00:00:00	Product2	16	2%	400
19	Country2	Fri, 24 Nov 2023 10:45:53	Tue, 09 Jan 2024 12:52:01	Product1	16	5%	15000
20	Country2	Fri, 24 Nov 2023 10:45:53	Tue, 09 Jan 2024 12:52:01	Product2	16	0%	90000
21	Country1	Sat, 17 Sep 2022 21:56:25	Sat, 17 Sep 2022 21:53:47	Product1	18	10%	8000
22	Country1	Sat, 17 Sep 2022 21:56:25	Sat, 17 Sep 2022 21:53:47	Product2	16	12%	400
23	Country2	Thu, 04 May 2023 22:02:51	Fri, 24 Nov 2023 10:45:53	Product1	20	4%	15000
24	Country2	Thu, 04 May 2023 22:02:51	Fri, 24 Nov 2023 10:45:53	Product2	20	2%	90000

The final outcome should be something like this. All of the highlighted values, are those values that had been changed from the pervious time.

Can you help a guy out to achieve this?

Best regards!

[lbendlin · Accepted solution]

Here's a general idea how you can do this with OFFSET

The sorting was done to check that the formula works, and can be changed back to the row number (or whatever else you are planning to sort this by.

Age Change = 
var v = sum('Table'[Measure1-age])
var vp = CALCULATE(sum('Table'[Measure1-age]),OFFSET(-1,ALLSELECTED('Table'),
                                              ORDERBY([Row Number (example purpose only)]),
                                              PARTITIONBY('Table'[Country],'Table'[Product]),
                                              MATCHBY('Table'[Row Number (example purpose only)])))
return if(COALESCE(vp,v)<>v,"yellow")

[IuteJV]

Thank you @lbendlin This works just great!