Append to existing sheet add time stamp when unique

[kmccleary]

I have a workflow that tags rows of data with a piece of metadata. Each row has a unique ID.

The data source never deletes the unique IDs or rows, they are just updated.

I want to be able to monitor when a new item hits the output and print a timestamp in an adjacent column.

So for example the first time it is run it will show:

ID123	1/1/2020
ID456	1/1/2020
ID789	1/1/2020

The second time, if all unique ID are the same:

ID123	1/1/2020
ID456	1/1/2020
ID789	1/1/2020

The next time, if a new ID hits the table

ID123	1/1/2020
ID456	1/1/2020
ID789	1/1/2020
ID345	2/7/2020

Basically, tag the "discovery date" based on if the ID is new or not.

[TomWelgemoed · Accepted solution]

Hi @kmccleary ,

I may be oversimplifying, but I've attached an example for you. I'm assuming you're running the same workflow multiple times, so you actually only need the first part of this workflow (highlighted in red). I've added a "day 3" to simulate how first, nothing changes and then an additional record is added.

In essence, it's just about doing an "right outer join" on the right fields (ID & Date?) and then ensuring the data that wasn't in the dataset originally gets a timestamp. Let me know if I'm off the mark here.

Best,

Tom

[kmccleary]

This is close but I have all of my data input in one tile. The one input passes through the workflow, some hits are there period after period (unique ID is present), when a new hit pops into the output, tag it with the date it arrived. So I can plot on a line graph the continuous flow of these hits over time.

[TomWelgemoed]

Hi @kmccleary ,

I think the same solution applies, the only thing is that you need to store your data in the end - ideally in a database, but Excel would also work.

When you run again, you're using your stored excel together with the one file in which these records usually show up. Let me know if that is unclear.

Regards,

Tom

[kmccleary]

You  are correct! I see now! Thank you. So simple I just couldn't picture it. Thanks again.