Not Planned

Customer would like to delete data in the MDA using Rules Engine.

  • 14 September 2016
  • 20 replies
  • 144 views

Userlevel 7
Badge
Wanting to delete older data weekly before inserting new data.

20 replies

Userlevel 6
Thanks Michael. Delete is a risky option on a system like Rules, do you think of a archive before the delete?
Userlevel 4
Sundar - We also have some similar requirement where we wanted to purge or archive data if not needed. Is there automated way of archive or purge ?
Userlevel 6
We are coming up with a feature to Export data to Your S3 bucket / Gainsight's S3 bucket, which can help in this regard.
Userlevel 5
Badge
Is there a way to delete Data within the MDA manually? I have a field i created, but want to delete the data so that i can edit the Data Type from Percentage to Number.
Userlevel 7
Badge
I think Delete should be used with caution, but should also be available. I'm creating a large number of rows for a daily calculation only to move the value back to the original table. Not being able to purge that table is a waste of space and adds some complexity to my rules. So I vote for "rule builder beware!"
Userlevel 7
Badge
And -- Johnny -- if you haven't seen this already, go to the Custom Object's Data page, filter to the right data sent, and then you can delete those records. (Or all records.)
Userlevel 4
Hi All,

Understand the risk. Can we get clarity on archive solution?

For now, manual deletion is fine but we are using MDA for automated processes and only want to keep the data we need for our daily CoPilot jobs. 

This table can get huge and we don't want to store all of that data in MDA is we don't have to.
Userlevel 4
API/automated method is good to have as we run into these situation all time
Userlevel 7
Badge +1
This is becoming a requirement for us as well. We have data loading currently on a daily basis but only need to keep the daily analysis for 12-14m. After that, we would like to aggregate the data to monthly in order to reduce the size of the file and then delete the daily records. Currently, we do not have a way to do this in mass. Any suggestions? 
Userlevel 6
Badge +1
I 100% vote this up with perhaps a large/loud disclaimer and mandatory test run and export/archive option? We have accrued over 361M records in one object and the optimal object size is 10M and we anticipate expanding our customer base rapidly. Please help!!!!

@jeffrey_boyer / @chris_mudd / @shibani_mehta Attempting to reinvigorate this need/requirement, please assist?

Userlevel 7
Badge +2
Same/ We are filling up MDA at rapid pace. Sooner than later we'll probably need this ability.

@dan_ahrens

Userlevel 7
Badge
Just wanted to call out that Administration / Data Management does support deletion of records in bulk.

https://support.gainsight.com/View_More_Categories/GainsightNXT/02Data_Management/02Managing_Data_In_Gainsight/01Gainsight_Data_Management#Deleting_Data_from_a_Custom_Object

I recorded a quick demo of this below. Just apply the filter to show the records that you want to purge and select the delete option. Takes about 30 seconds.



We realize this is not as simple as doing this via rules engine, but wanted to make sure that everyone knew that purging old records is supported today with just a little bit of manual effort.

In cases where you have huge quantities of records (hundreds of thousands or even millions) you can drop a support ticket to Gainsight Support and our tech ops team can help with those large quantities. @keith_mattes

Userlevel 7
Badge +2
Yep, aware of that. There isn't a test run function though - that would be an extra safety net in the Rules Engine.

Userlevel 7
Badge
Hi Jeff,

Can you explain a bit more as to what missing functionality you'd be looking to achieve that a test run function would give you?

Userlevel 6
Badge +1
In my "vision" this could maybe be solved with a two-step Delete-Purge "Archive" process that would initially provide a summary of the total number of records to be deleted as well provide a list of dependancies that would be impacted (this could be a reach although seems doable via analyzer tool).

Example:

XXXX number of records will be deleted that were used in the following items:

*Report Name

*Dashboard Name

*CTA Name

BONUS ADMIN FEATURE: The ability to place records in a temporary state aka "soft delete" (or table) to "test" impact" with ability to "revert" if change/impact too great or undesired.

SUPER BONUS FEATURE: Part of this Archive feature enables the automatic creation of an archive roll-up table with the multi-select option for tables for weekly/monthly/yearly rollup of the data

Badge

Long story short, “upsert” is taking too long that forces my data load rules to get aborted. Have tried different directions to get around it, but ultimately this functionality would solve it and be a huge win! Either the Rules Engine or somehow deleting in mass via the API, we would love this functionality!

Userlevel 3

This would be very useful in some of my use cases as well.

A sample is when some data is related to SFDC and we delete from SF it should be deleted on MDA as well... But in an automatic way or by rules.

Userlevel 5
Badge +1

The use case we have for deleting of records is related to data privacy.

We hold some data (including PII) that belongs to our customers and they can withdraw consent at any time, after which we need to delete the data. The consent status for all customers is included within an S3 file we get daily, and so based off that we would then want to delete relevant data for those where consent = false. Also churned customers should have the data deleted within 90 days. 

 

Currently it looks like we will need to anonymise the data instead, as deleting isn’t possible?

Userlevel 6
Badge

Just wanted to call out that Administration / Data Management does support deletion of records in bulk.

https://support.gainsight.com/View_More_Categories/GainsightNXT/02Data_Management/02Managing_Data_In_Gainsight/01Gainsight_Data_Management#Deleting_Data_from_a_Custom_Object

I recorded a quick demo of this below. Just apply the filter to show the records that you want to purge and select the delete option. Takes about 30 seconds.
 

ScreenRecording20190313at11-59-4f2ac9de-df52-4060-a490-4359d630e6e6-614912913.gif

 


We realize this is not as simple as doing this via rules engine, but wanted to make sure that everyone knew that purging old records is supported today with just a little bit of manual effort.

In cases where you have huge quantities of records (hundreds of thousands or even millions) you can drop a support ticket to Gainsight Support and our tech ops team can help with those large quantities. @keith_mattes
 

 

+1.

we have an option to delete the records from Data Management.

Userlevel 5
Badge +1

@phani_kumar I’m aware that is an option, but it’s not feasible to expect an admin to head to Data Management, apply the filters and delete the required records each day. Automation on this subject is required. 

Reply