Email finder - anyone run it recently? Comment on your results?

JoAnn Strommen

This summer we have run CellPhoneFinder and DeceasedRecordFinder. (AddressFinder run through NXT as well.) The next data clean up process potentially on my list is/was EmailFinder.


I've read through these old posts and see only a few recent comments but really mixed reviews.
https://community.blackbaud.com/forums/viewtopic/420/25604?post_id=90896#p90896  (started in 2015)
https://community.blackbaud.com/forums/viewtopic/143/26884?post_id=96119#p96119 (2016)
https://community.blackbaud.com/forums/viewtopic/147/33191?post_id=123047#p123047 (2017)


The comments about receiving old data, inaccurate, etc. concern me without even getting into the opt in/out issues. We currently do have platform for alums to update their info so those records are in pretty good shape IMO.


I'm trying to decide if it would be worthwhile to run or if as some commented it just muddied the waters. Would sure like to hear from anyone who has run it recently enough to know if the results were valid, worth the clean up time for what was gained, etc. We would have records without an active email and those with, so I know it would require time to compare and clean up.


Any insights are appreciated, either here or personal message.

Becky Jones

The last time we use AddressFinder, PhoneFinder & DeceasedFinder was 2-16-16.  I remember the date because it caused a giant mess.  We had used it the year before with very little problems.  In 2016, it duplicated info that was already there in the phone section.  It added a new address/or the same address we had for EVERY record....We have decided that we will not be using it until some of the problems have been worked out.  Our alums can update their info on our database so we get new info from there.  We have also sent out surveys to alum classes getting ready for reunions that year so it breaks it up into manageable amounts of information to update.  

Karen Diener 2

People seem to think that email and phone appends are somehow magical and will result in perfect matches of hidden information for your donors, and it just isn't the case.  Managing expectations and understanding the process is really important with data appends.


AddressFinder should only add duplicate addresses if you choose that setting.  I forget how it reads specifically in the wizard, but there is a way to have it only update addresses with changes.  If you don't check that box, it will "update" every address by adding a second one.


PhoneFinder will add additional phone numbers with a new type.  This for a lot of reasons - national do not call registry, differences between phone types that you are using in your database vs. what will be in the return file, etc.  I've used third-party data services as well as Blackbaud's Phone Append and the same thing happens with both.  This is an issue with data appends and not with Blackbaud specifically.


I have only run one email append and was not a fan.  It was not through Blackbaud.  I did receive quite a few emails that did not belong to the constituent on record, or received an email that was old, or a very generic one that the donor had set up for online shopping or other public-facing situations, so emails never ended up getting to the constituent.


I don't typically recommend phone or email appends, quite honestly, unless you have a strategy of how you will be using them.  I think you ultimately have a better list if you collect them because people want to hear from you.  What's the point of having 10,000 email addresses if the majority don't even make it to inboxes, or get read, or were from a data service?  Focus on quality and not quantity.  Easy for me to say, of course!!


Good luck - 


Karen


 

Elizabeth Johnson

Hi JoAnn Strommen‍, 


We ran EmailFinder at the beginning of April. Our alumni and donor email addresses are in fairly good shape however since we receive the service for free as part of NXT we wanted to give it a shot. It takes about 3 business weeks to process on Blackbaud's end. I find that Blackbaud often under promises and over-delivers. 


I haven't quite finished processing since the emails come in with their own type there was no rush and other things are clearly more important right now. I exported the results to share some insights with you and a good opportunity to reflect on was it actually worth it.


We received back 785 EmailFinder results. This number represents about 1/3 of our active database that was not emailable at that time. Of that, I still have 384 to review, the remaining people are not alumni or donors. 


Of that 401:

• I have gained 297 good matches 74%
• 88 were clearly not good matches (known emails belonging to family members or emails with someone else's name in the email address)
• 16 that looked possible have hard bounced.

I wasn't expecting it to be perfect and I'm pretty happy with the customer service and the results because it is something already included in our pricing. I would do it again. I created a new email type of EmailFinderNotLikely to make it really easy to compare the results when I run it again next year. Also, all of the emails have a comment that includes the exact phrasing "Source: EmailFinder 4/30/2020." Next time I would exclude people that have previously opted out. I included them this time since it was an opt-in service. Many of those fell into the category of not a good match and thusly was a waste of time. Live and learn.  


May you find this useful.


Best,

Elizabeth

?


 

Dariel Dixon 2

Of that 401:

• I have gained 297 good matches 74%
• 88 were clearly not good matches (known emails belonging to family members or emails with someone else's name in the email address)
• 16 that looked possible have hard bounced.

I wasn't expecting it to be perfect and I'm pretty happy with the customer service and the results because it is something already included in our pricing. I would do it again. I created a new email type of EmailFinderNotLikely to make it really easy to compare the results when I run it again next year. Also, all of the emails have a comment that includes the exact phrasing "Source: EmailFinder 4/30/2020." Next time I would exclude people that have previously opted out. I included them this time since it was an opt-in service. Many of those fell into the category of not a good match and thusly was a waste of time. Live and learn.  

 

Elizabeth Johnson‍ I wanted to ask about your the 297 "good matches".  I'm curious as to how you determined them to be good?  Does good mean simply valid, meaning they didn't bounce or look wrong, or that they were emailed to and read or opened.  I'm very worried that I wouldn't get a primary email address with these services.  I have several email addresses, but only a few that are checked.

JoAnn Strommen

Thank you for the feedback. 

Becky Jones‍ - I see the same duplicate addresses on every record here from an address finder run made 9/2014.  I believe our alum records are in pretty good shape so that's certainly a factor in our decision.

Karen Diener‍ - wouldn't it be nice if it were magical.  With CellPhoneFinder I saw the issue with receiving data that matched what was already marked as invalid in our db. Due to the results for records where we had no phone #, we felt it was worthwhile for us.  I did spend a day in a mammoth export of phone data cleaning up all the duplicates.  Agree quality so much more important than quantity.

Elizabeth Johnson‍   Thank you for sharing your results. Are you sending an Opt-in email as described in KB 54014? Or as Dariel posted, how are you determining they are valid?


Greatly appreciate your feedback as we evaluate our situation. 

Elizabeth Johnson

 Hello Dariel Dixon‍ and JoAnn Strommen‍,


Excellent questions - I am so not a questioner and appreciate it so much when others are.


Let me address the opt-in piece first. When going through the process of submitting your records to Blackbaud to execute the EmailFinder you have to supply some information that they then use for Opt-in message.


What do I consider valid and successful? It was easy to determine the ones that were not "good" emails that is true! It appears my tendency for optimism has played a roll in how I answered the question. It doesn't change the fact that I would still use the service again.


We made a conscious choice not to hit our constituents up with an ask as their first email. Notifying people of our Covid-19 Response turned out to be the first to some, followed up by DEI work for many and a thank you for helping us make our goal. Not everyone fell into a group that has received an email so not all have been tested yet. 


A deeper dive shows me that of that 297 is far less impressive than my initial fast attempt to get some stats. 

• 66 received an email and actually opened (some even went on to click)*
• 84 received and didn't open it
• 11 have only soft bounced - I don't normally give up on soft bounces but in this case I will code them as EmailFinderNotLikely
• 136 of the emails I have deemed to be valid have not yet been sent an email - but they will be sent one in October. I highly doubt I'll remember or be inclined to update my additional findings here at that time also I will have processed more of the other EmailFinder results so the waters will be muddied

*We use Online Express. I add the data to RE shortly after the email is sent and then update it again at the 1 and 2-month intervals. 


Looking at this would I do EmailFinder again? Probably, but not annually. I do think that it would be great to send an email out to the group and give them a second chance to opt-in using an OLX form. 


May this be even more helpful!

Best,

Elizabeth

?

Dariel Dixon 2

Thanks for the response Elizabeth Johnson‍.  That type of analysis is so helpful to others when considering using one of these appends.  We all know how inconsistent appends can be by nature, but I'm really glad to see that you have results that you are pleased with.  And that it was worth the cost.

Joe Moretti 3

@JoAnn Strommen I find email finder pretty useless. It searches based on address, so you can end up with the constituent's kid email or someone else in the household. Also if the people moved, but you still have the old address it will give you an email of the new people who live there. Considering that people change emails a good bit, it is not worth it. More headaches than anything.