Time taken for clerical review

This forum is in support of all issues about Data Quality regarding DataStage and other strategies.

Moderators: chulett, rschirm

Post Reply
brendomac
Premium Member
Premium Member
Posts: 7
Joined: Tue Nov 29, 2016 5:44 pm
Location: australia

Time taken for clerical review

Post by brendomac »

Hi everyone,

We are attempting to create a timeline for a project along with resource needs.

I was hoping someone could provide some insight into the time taken for clerical review from a datastage matching job.

For example, our experience indicates clerical record review takes on average 5 minutes per record. Using a base of 1000 pairs to be checked it would take approximately 83 hours to perform the review

Thanks
Brendan
qt_ky
Premium Member
Premium Member
Posts: 2895
Joined: Wed Aug 03, 2011 6:16 am
Location: USA

Post by qt_ky »

I guess it depends entirely on the nature of your data.

Here is some food for thought. In our QualityStage matching project, the customer did not want to spend any human time doing ongoing clerical reviews, so we set the high and low thresholds equal to each other, and put them at a level that the customer was comfortable with.
Choose a job you love, and you will never have to work a day in your life. - Confucius
brendomac
Premium Member
Premium Member
Posts: 7
Joined: Tue Nov 29, 2016 5:44 pm
Location: australia

Post by brendomac »

Thanks qt..
vmcburney
Participant
Posts: 3593
Joined: Thu Jan 23, 2003 5:25 pm
Location: Australia, Melbourne
Contact:

Post by vmcburney »

Depends on what business process you want to follow out of the clerical review.

During the testing phase you will need a block of time to investigate the clerical reviews and then go back and improve your standardisation and matching rules to reduce future clerical reviews.

You have a process of adjusting your clerical review thresholds based on what you are finding.

You could have a process of going back and fixing some records back at source to get the match to work on the next batch run and this will have a support cost. You can have a low cost in the matching team to review and verify that two records should be matched. You have a higher cost in the customer support team to go and fix the source records to make them more consistent for that customer.
brendomac
Premium Member
Premium Member
Posts: 7
Joined: Tue Nov 29, 2016 5:44 pm
Location: australia

Post by brendomac »

Thankyou VM, I appreciate your input.
Post Reply