Best practices for data review?

Options

I hope I'm not asking something that's already been answered, but I didn't immediately see a post on this, although I haven't been a consistent user of the forum so I could have been looking in the wrong place.

Our development staff recently expanded, with an additional layer between data entry and the director. I have moved from being primarily responsible for data entry and database administration/ data health to being responsible for database administration and supervising the data entry process. This is a needed change, as our database has expanded drastically over the past couple of years, and having a position that is responsible for data integrity is overdue.

My question is, how do other organizations review their data? Having not had a formal plan in place before, I have been thinking of the ways in which we pull, use and report on our data. Creating a training manual to compliment our database protocols has been valuable in organizing my thoughts on our needs, but I'd love to hear from other organizations tackle database health. Are there any best practices out there?

Tagged:

Comments

  • Nicole Holt:

    I hope I'm not asking something that's already been answered, but I didn't immediately see a post on this, although I haven't been a consistent user of the forum so I could have been looking in the wrong place.

    Our development staff recently expanded, with an additional layer between data entry and the director. I have moved from being primarily responsible for data entry and database administration/ data health to being responsible for database administration and supervising the data entry process. This is a needed change, as our database has expanded drastically over the past couple of years, and having a position that is responsible for data integrity is overdue.

    My question is, how do other organizations review their data? Having not had a formal plan in place before, I have been thinking of the ways in which we pull, use and report on our data. Creating a training manual to compliment our database protocols has been valuable in organizing my thoughts on our needs, but I'd love to hear from other organizations tackle database health. Are there any best practices out there?

    Nicole - conratulations on your move.

    I am not exactly sure how to answer your post as I am not sure exactly what you mean by "data review". The data in a database gets reviewed in any number of ways - with different goals. Do you mean what I call "data auditing"? Checking to be sure that things are being put in properly, completely, and accurately?

    I have presented at BBCON on data auditing before and have some slides on the topic. The basic premise is querying - what can you check by query and how often should you check it? I have query folders for daily, weekly, monthly, quarterly and annual audits. Every day after gift entry is complete - we run the daily audits. These look for common errors made in gift entry which would result in an incorrect receipt or report to the team. Every week we query on other things that should get addressed quickly for reporting purposes but would not have resulted in a problem with a gift receipt. We do most of our constituent audits monthly - errors in addresses that were changed, etc.

    There will certainly be things you can not query on. The donor indicated on a reply device they wanted only one solicitation a year - did your gift processor see it and mark the record properly? No query would catch this - these would need to be double checked from the original backup after gift entry, or if not a huge issue, minimally spot check records to be sure there are not issues with these being missed.

  • JoAnn Strommen
    JoAnn Strommen ✭✭✭✭✭
    Ancient Membership Facilitator 4 Name Dropper Photogenic
    Nicole Holt:

    I hope I'm not asking something that's already been answered, but I didn't immediately see a post on this, although I haven't been a consistent user of the forum so I could have been looking in the wrong place.

    Our development staff recently expanded, with an additional layer between data entry and the director. I have moved from being primarily responsible for data entry and database administration/ data health to being responsible for database administration and supervising the data entry process. This is a needed change, as our database has expanded drastically over the past couple of years, and having a position that is responsible for data integrity is overdue.

    My question is, how do other organizations review their data? Having not had a formal plan in place before, I have been thinking of the ways in which we pull, use and report on our data. Creating a training manual to compliment our database protocols has been valuable in organizing my thoughts on our needs, but I'd love to hear from other organizations tackle database health. Are there any best practices out there?

    Nicole, there have been some past discussions on this but, yes, can be hard to find using search - get lots of job postings.  Some words to try might be: database cleansing, audit.  Most are probably under data or admin sections of forum. 

    I start with "Where do mistakes happen?"  Build a query to check for that.  Can be missing data - like no constit code (even though it's a required field)zip code, no title, no gender, coding/notation for invalid addresses/address updates, address blank + 'no valid address' unchecked...etc. To me, you're definitely starting correctly by looking at how you pull, use and report on your data

    I did find this earlier post - wasn't the one I was looking for but might give you some ideas on keeping data integrity: http://forums.blackbaud.com/forums/p/15375/145596.aspx#145596

    Just a couple of thoughts

  • Melissa Graves:

    Nicole - conratulations on your move.

    I am not exactly sure how to answer your post as I am not sure exactly what you mean by "data review". The data in a database gets reviewed in any number of ways - with different goals. Do you mean what I call "data auditing"? Checking to be sure that things are being put in properly, completely, and accurately?

    I have presented at BBCON on data auditing before and have some slides on the topic. The basic premise is querying - what can you check by query and how often should you check it? I have query folders for daily, weekly, monthly, quarterly and annual audits. Every day after gift entry is complete - we run the daily audits. These look for common errors made in gift entry which would result in an incorrect receipt or report to the team. Every week we query on other things that should get addressed quickly for reporting purposes but would not have resulted in a problem with a gift receipt. We do most of our constituent audits monthly - errors in addresses that were changed, etc.

    There will certainly be things you can not query on. The donor indicated on a reply device they wanted only one solicitation a year - did your gift processor see it and mark the record properly? No query would catch this - these would need to be double checked from the original backup after gift entry, or if not a huge issue, minimally spot check records to be sure there are not issues with these being missed.

     Thanks so much, Melissa. "Data auditing" would be a more precise term. Your schedule of daily/weekly/monthly queries looks like a great start! And, as you said, the items that are not captured are likely to be the more difficult issues. Since we have new staff, daily reconciliation of data provided vs. data entered will be vital, at least for the near-term.

  • JoAnn Strommen:

    Nicole, there have been some past discussions on this but, yes, can be hard to find using search - get lots of job postings.  Some words to try might be: database cleansing, audit.  Most are probably under data or admin sections of forum. 

    I start with "Where do mistakes happen?"  Build a query to check for that.  Can be missing data - like no constit code (even though it's a required field)zip code, no title, no gender, coding/notation for invalid addresses/address updates, address blank + 'no valid address' unchecked...etc. To me, you're definitely starting correctly by looking at how you pull, use and report on your data

    I did find this earlier post - wasn't the one I was looking for but might give you some ideas on keeping data integrity: http://forums.blackbaud.com/forums/p/15375/145596.aspx#145596

    Just a couple of thoughts

     Thanks so much, JoAnn! And thank you for the suggested search terms. 

     

    Best,

    Nicole

     

Categories