My understanding is that these alerts are for the lender's and not to be used to force the appraiser to try to address each and every item that the lenders are seeing due to their computer printouts.
Your understanding is partially right.
The alerts are for the lender. However, there's nothing stopping the lender from using them as part of the QC process.
Did anyone really think that CU wouldn't be used as a review tool? I mean, a lender submits an appraisal to the GSEs and the GSEs come back and say,
"We have the 5 following risk flags on the appraisal report you are going to submit to us along with the loan package; we'll share with you now what we are going to see if you submit the package with the appraisal as-it-is."
What lender in its right mind is not going to review the CU risk-flags and get the appraiser to comment on those items before they submit the appraisal to the GSEs?
In regard to the first issue (different ratings from other appraisal reports), while my initial reaction would be to respond as JGrant suggests, I'd take a step back and not respond like that. Rather, I would look at my report to see if I've adequately communicated why I rated the comparables as I did. If I had adequate commentary, I'd refer the reviewer to that existing commentary. If I didn't have the commentary, I'd amend the report to include it.
In regard to the second issue (using the same comps for different assignments but with different condition ratings for those assignment) could be more problematical.
The reason could be that the comps have been updated/renovated or have experienced some negative condition influence since the last time the appraiser used them; if that is the case, the appraiser would be able to describe what new information s/he has learned and why it warrants a different condition rating than before.
The reason could be that the appraiser didn't realize the comp was used before and the condition-rating is on or near the boundary between one rating and another. In the individual appraisal, it may not impact credibility because it is a discrepancy in the rating and not in the comparison or adjustment. When evaluating multiple appraisals, it may look like a problem (appraiser is fudging the condition to fit a specific outcome).
I understand that this can be a good tool for fraud, but the application could get unnecessarily burdensome. I'd like to see Robert Parker (Wine Connoisseur) taste the same wines over and over again without the benefit of remembering what he rated them previously, and consistently rate them the same each and every time. I don't think it would happen. I think there should be an expectation that the condition rating... especially those on the margin.... have a good probability of being inconsistently rated in different reports over time.
Or, the appraiser could be fudging the condition rating (that's what this screen is designed to detect).
I'm assuming in your case, the reason for the different comparable ratings is either something had changed since the last time you used the comparables, or it is simply a case where the rating could fall either way and in this assignment, it fell differently than in the last. Either of these scenarios can be adequately explained (IMO).
Good luck!
(BTW: Northern California Appraisers- there is a Residential Symposium on August 3rd in Dublin, CA. One of the sessions is going to be a presentation by a lender and a National AMC to discuss how the CU is being used in the QC process; what are the common problems, how they can be avoided/addressed, etc., etc. See this thread for more details:
http://appraisersforum.com/forums/threads/residential-symposium-in-dublin-ca.207951/).