We had mailed out a survey and we pointed to a web link, but of course their are people that would rather return the paper copy.  

So I have a dilemma with spouses as the main constituent may not have filled out the survey, but the partner did.  The spouse does not have a full record but has filled in the survey.  There are fields like Gender and age group that could be different than the partner's.  We do not want to capture this data on the record if it is incorrect meaning the main record is a man and the partner is a woman.  How would you handle that so it can be imported correctly?  Create a new record for the partner?  How would you prevent mail going to the other record as we may not know that they are a "duplicate" record with regards to the mail?

Another doozy is if the partner fills it out and was never in RE and has updated the information to include the partner?

 

We are expecting about 8,000 surveys so manual correction is not an option.

Have you considered pre-printing the name of the constituent on the survey so it is very clear who the survey is for?

Also, don't know if it will help, but much improved non-constituent spouse matching/promotion to full record options coming to ImportOmatic very soon!

We did print the Addressee on the form with the current details so they could update them, and this is how we discovered that it was sent to a male and they checked female for gender. Likewise we know the guy is an investment banker and she put homemaker. Not all of our records have the male as the primary or the female. A lot the female should be the primary, but historically they entered the male in as the main constituent.

Do you know when the new version will be out? Although, the details in the survey pertain to the spouse which is not a full record.

Thanks for the thoughts Jeff!

I think the approach I would take is to try to split out the possible spouse data before the import and then enter just those manually. I would assume that the majority of the data is correctly tied to the primary record and there are only a handful that are incorrect.

So the question for me becomes: how do you figure out that a survey was filled out by the spouse instead of the primary?

You already mentioned that you have the addressee on the paper survey, and since you are going to have to transcribe them into a spreadsheet anyways you can have the data entry person make a call and flag the possible bad data.

Another clear indication is that if they do not have a spouse then you can just import them.

So really you are just looking at surveys from records where there is a spouse and then you will just have to quickly decide if the primary is M/F and then see what they responded. If it's suspicious then put it in a manual batch.

I'm not sure what else in the survey but if you are depending on gender then maybe you could output the current gender on the record and compare the two. If they are opposite then that's an indication that the survey was filled out by someone else.

Not so much an import-o-matic problem but a data sorting issue.

I'm sure you have already thought of this but next time I would recommend adding a line at the top of the survey asking the name of the person responding :)