PhoneFinder Best Practices help

LINDSEY COPELAND

Hello all!  (If i'm in the wrong community, I'm sorry)


We're looking at running PhoneFinder (and Cell finder) this year. We ran it about 6 years ago before I started, so I've never done it myself.  However, we had a big issue 6 years ago with the service because you don't always know who the number belongs to that you're getting back (for instance, it could be the spouse of the alumnus or the teenager who shares the same address and account.)  


We are desperate to get new phone numbers for our constituents, but we are also desperate to not call our constituents' children while trying to solicit via our phonathon program.  After talking with Blackbaud reps/technical support, it doesn't sound like we can get straight clarity on who actually uses the number we get back - so I am reaching out to you all!


Does anyone have any suggestions for best practices on cleaning this data? How do you use Phonefinder? What are the processes that you have found to be useful?


Anything will help! Thank you!

Ryan Hyde

While others may have some good tips and tricks, I think the problem is that phone numbers are finicky things. You're absolutely going to have an error rate with PhoneFinder (or any other phone matching program that any other company offers) because people change phone numbers, people abandon phone numbers, phone numbers stay tied to addresses sometimes and family members shuffle houses, and then there's just plain 'ol human error when entering data at some point upstream from where you are in the data harvesting process. 


For sanity's sake, I think you should absolutely do two things.


1.) Import these phone numbers with a unique phone type (Finder 2019, for instance) so they are easily identifiable.


2.) When you call the number, either mark it "primary" or "inactive" depending on the results, and do NOT delete any of these numbers until after you do a final analysis. After you've called a lot of the numbers, do some analysis to see what your failure rate actually is. Your coworkers/superiors may get very agitated if they personally hit a couple of wrong phone numbers and they may start to assume that there is a very big problem in the data, but if you can determine that the error rate was, say, 5%, then suddenly the problem seems to be within the acceptable range. 


Neither of those things will help you validate the numbers before you call them though. You just have to do the work, I think. Just be prepared with an apology and a "I'll make sure this doesn't happen again" sort of statement. And if you end up getting a family member of the person you're trying to call, they may give you a legit phone number if you're nice enough on the phone.

Heather MacKenzie

Great advice, Ryan Hyde‍! 

JoAnn Strommen

Ran CellPhoneFinder (CPF) and looks like we've gotten some good results as far as number of numbers they found.  Does anyone have some suggestions on data clean up now?


I've looked at the list, I've exported all phone data, coded those who had no 'cell' number on record so I can change type from CPF to type cell. Still verifying if number is on record with another type.  


(Can old numbers with type of PhoneFinder be cell #s? In years past did PhoneFinder look for more than land lines?)


Now I've got to look for those who have CPF number on record with type of cell, home, alternate.  My goal is to change the CPF type to cell, put CPF 2020 in comment field.  I just know we don't want another active phone # type to pull into data alum's view/update.  Already questioned by them this a.m. It's just mind boggling - we do not have the cleanest data. Can't believe the # of phone types that exist in db. I'm working on it but just feeling brain's about to explode. LOL and I like puzzles and challenges. 


I created a small test file with 10 random records that seem to have most of the issues I'll be dealing with - duplicates in other fields, no cell, other active cell, etc.   


Any suggestions on processing this greatly appreciated.

Jessica Daggett

JoAnn, I'm going through a similar process with PhoneFinder (and maybe EmailFinder). We've exported a file for everyone who received PhoneFinder, each phone type in its own column. So each constituent will have Home1 with the Type, Number, Import ID (and anything else), then Home 2 with the Type, Number, Import ID, and so forth. This way I can do comparisons in Excel to identify duplicates. Then I can swap the type of PhoneFinder to Home1 if I want (though I like the idea of adding a comment that the source was PhoneFinder). Hope that helps.

Jessica Kraybill | Database Administrator

Project Angel Heart
MAIN 303.830.0202 | FAX 303.830.1840
4950 Washington St. Denver, CO 80216
www.ProjectAngelHeart.org   Follow us: Facebook | Twitter | Instagram

 

JoAnn Strommen

That's what I've done too is exported all out including primary, inactive, comment and DNC.  File is huge. Did a Match lookup so I know where the # is on record already.  Working my way through group by group.  Just completing those who already have a different cell # on record but have the CPFinder # as a home #. 


Been doing a global change to 'delete' phone type and deleting each group as I go too. Found old PeopleFinder and PhoneFinder records too that I'm trying to make a clean up plan for.

Way more work than I expected - partly due to RE structure, partly the state our our db. 

Several snags are that DNC is not an import field, phone ID not somewhere too.


Oh well, db will be in better shape when I'm done.