This is common in ERC appraisals, too. It can be enlightening to see what your competitors are up to, and does provide an opportunity to assess and defend your own processes. I recently did an ERC appraisal and apparently was the third appraiser, along with at least 2 brokers, to opine an anticipated selling price (in this case, everyone was to suggest the price the subject would sell for within 120 days from the effective date). The initial response to my report was that I was outside their 5% differential (as I understand it, if the first two are more than 5% apart, a third appraisal is commissioned). Among the differences they noted was in GLA, and I did find where I had missed that my software had classified a porch as GLA. After correcting that (less than a 1% change in value), I was still about 15% above the high end of the crowd. The same questions were asked of the other appraisers, and among the questions I got were about several sales that were not in my report.
Most of those were in my work file, having been adjusted to the subject and fully considered. Others were simply irrelevant. During the course of the conversation, I asked the client for grids from the other appraisals (so I could explain what they did wrong). I only got snippets for the comps in question, but that was telling. The subject was odd in that it had about 20% of its GLA in a bonus room. I treated it as GLA as the stairway to the area was from the intersection of the foyer, living room, and kitchen dining areas. The glaring difference between their appraisals and mine was in GLA adjustments. My adjustment was $90/sf. Theirs were either $77 and $61 if they gave the large bonus room no value, or $26/sf if they included the bonus room in the GLA. That feature is not common in the broader market area, but is much more prevalent in one subdivision where other aspects of homes (age, quality, location, etc) are much less variable than in the broader market area. When I developed another model from just sales in that subdivision, it suggested GLA including bonus room area added $100/sf, but when separated, GLA added $106/sf and bonus room area added $69/sf, or about 65% of GLA value.
For I think the first time ever, I provided the regression results for those scenarios so they could see the basis for my statements, and told them even if I conceded that the subject bonus room should be devalued by 35%, my value would only drop by about 2%. Somewhere in all of that, the relocation company and transferee agreed to list it for a little over 97% of my estimate, and they had an offer in 11 days in a market where many sales are contracted after nearly 90 days on the market. It didn't pay well, but it was an interesting and enjoyable exercise that confirmed I don't have glaring flaws in my process.
