Showing results for 
Search instead for 
Did you mean: 

Best Practice for Ongoing Collection Task-CW7.1.3

Level 3



Thanks for reading this.  I was wondering what my best option would be in the following scenario:


I have 21 mailboxes i am collecting.  The collection task is scheduled to run weekly so i can get any changes in the mailboxes. i am wondering if i am better off creating a new collection set for each task so i can ingest data each week or if i should just create one collection set and ingest all the data at the same time.  What do you all think?  Am i missing something?  This is the first time i have had to deal with ongoing collections.







Level 5
Partner Employee Accredited Certified

Hello Mark,

I've linked below an answer given by Liam Finn on another thread, although it's not the same question, I think the answer is probably going to be very similar :

Best of luck.

Level 3
Partner Accredited

You can certainly create one collection task for all 21 mailboxes to run weekly.  At the end of the each task you will have to create a collection set prior to processing the data.  I recommend identifying each collection set and processing batch with a date or week number so that you can easily identify these collections during review.

Alternatively, you can create an individual task for each of the mailboxes to run weekly.  This makes it easier to add/remove a mailbox collection task should you need to.  This also helps in troubleshooting in the event that you run into any issues collecting from a mailbox.  If you create an individual collection task for each mailbox, you will still be able to create one weekly collection set that includes all 21 mailbox collection tasks.


Level 6
Employee Accredited Certified

I think this comes down to your preference. For ease of management a single task is easier to manage but if you want to follow the advice from Miguel seperate tasks and a collection set to process them in a single step can also work

My preference is to keep it as easy to manage as possible. You can collect the data as a single task and then in Preprocessing sort the data so it can be processed in batches in different collection sets (if thats what you want) or process the complete collection set as a single processing job.

On another note, weekly mailbox collections may not be the best move if you want to decrease the chance of missing data. Items deleted between collections will be missed. The ultimate solution is that email is Journaled so that nothing is missed and all BCC and excpanded distribution list membership is captured