This time we’re going bigger than ever. Fabric, Power BI, SQL, AI and more. We're covering it all. You won't want to miss it.
Learn moreDid you hear? There's a new SQL AI Developer certification (DP-800). Start preparing now and be one of the first to get certified. Register now
Hi
I have a resonably large table with the three fields shown below.
| Surveyor Name | Unit Reference | Date Completed |
| Oliver | 8088 | 08/09/2015 |
| Chris | 8088 | 28/09/2023 |
| Rizwan | 8088 | 08/09/2015 |
| Scott | 153114 | 31/08/2023 |
| Kuldip | 153114 | 07/08/2015 |
| Kuldip | 153114 | 07/08/2015 |
I require some help to how I can group the Unit Reference to display the Max Completion Date and who the Surveyor was.
I would expect this result
| Surveyor Name | Unit Reference | Date Completed |
| Chris | 8088 | 28/09/2023 |
| Scott | 153114 | 31/08/2023 |
Please note I already have some M-Code steps being performed on this table.
thankyou
Richard
Solved! Go to Solution.
So, sorry. I misread the requirement.
See below code.
let
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45W8s/JLEstUtJRsjCwsABSRgaGproGlroGFkqxOtFKzhlFmcXIskbGIFkjiGxQZlV5Yh5OzcHJ+SUlQGFDU2NDQxO4dgtdY0OwvHdpTkpmAaoCkAEWugbmRCiIBQA=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [#"Surveyor Name" = _t, #"Unit Reference" = _t, #"Date Completed" = _t]),
#"Changed Type" = Table.TransformColumnTypes(Source,{{"Surveyor Name", type text}, {"Unit Reference", Int64.Type}, {"Date Completed", type date}}),
#"Grouped Rows" = Table.Group(#"Changed Type", {"Unit Reference"}, {{"All", each Table.SelectRows(_, (x)=>x[Date Completed] = List.Max(_[Date Completed]))}})[[All]],
#"Expanded All" = Table.ExpandTableColumn(#"Grouped Rows", "All", {"Surveyor Name", "Unit Reference", "Date Completed"}, {"Surveyor Name", "Unit Reference", "Date Completed"})
in
#"Expanded All"
Insert the below step where #"Changed Type" is my previous step
= Table.Group(#"Changed Type", {"Surveyor Name", "Unit Reference"}, {{"Date Completed", each List.Max([Date Completed]), type nullable date}})
Hi thank you for responding so quickly. However the output is stillthe same os before
Surveyor NameUnit ReferenceDate Completed
| Oliver Goldsmith | 8088 | 08/09/2015 |
| Rizwan Ahmed | 8088 | 08/09/2015 |
| Kuldip Jassal | 153114 | 07/08/2015 |
| Scott Williams | 153114 | 31/08/2023 |
| Chris Donczyk | 8088 | 28/09/2023 |
Richard
So, sorry. I misread the requirement.
See below code.
let
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45W8s/JLEstUtJRsjCwsABSRgaGproGlroGFkqxOtFKzhlFmcXIskbGIFkjiGxQZlV5Yh5OzcHJ+SUlQGFDU2NDQxO4dgtdY0OwvHdpTkpmAaoCkAEWugbmRCiIBQA=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [#"Surveyor Name" = _t, #"Unit Reference" = _t, #"Date Completed" = _t]),
#"Changed Type" = Table.TransformColumnTypes(Source,{{"Surveyor Name", type text}, {"Unit Reference", Int64.Type}, {"Date Completed", type date}}),
#"Grouped Rows" = Table.Group(#"Changed Type", {"Unit Reference"}, {{"All", each Table.SelectRows(_, (x)=>x[Date Completed] = List.Max(_[Date Completed]))}})[[All]],
#"Expanded All" = Table.ExpandTableColumn(#"Grouped Rows", "All", {"Surveyor Name", "Unit Reference", "Date Completed"}, {"Surveyor Name", "Unit Reference", "Date Completed"})
in
#"Expanded All"
That works fine thank you
Maybe not as elegant but after that I would sort the date column descending and delete duplicates by Unit Reference so you only get the max date for Unit Reference and it's surveyor.
Hi
thanks for the tip. I have tried this however whenremoving duplicates it removes the wrong ones and does not leave the max date.
Richard
That happens sometimes after sorting, power query doesn't reflect it. Add an index column after the sorting that would fix it.
Thank you😀
Check out the April 2026 Power BI update to learn about new features.
If you have recently started exploring Fabric, we'd love to hear how it's going. Your feedback can help with product improvements.
| User | Count |
|---|---|
| 5 | |
| 4 | |
| 4 | |
| 4 | |
| 3 |