If you're talking about reviews written on a review/news site, then their role is not important as far as impacting book sales. We haven't had a single sale linked from Flayrah.com after years of reviews being posted there.
I think it’s worth exploring how you can determine that. You’ve mentioned this before and always used the word “link,” and I think you may even explicitly said you’re talking about link referrals. That’s a very (ahem) fuzzy measure, though: it only counts anyone who reads the review and immediately clicks on the link from the review to buy the book right then.
If they come back a week later – or even an hour later – to the FurPlanet site directly? You can’t associate that to the review.
If they buy the ebook, even at Bad Dog Books? You can’t associate that to the review.
If they see the book at a con and think “Oh, yeah, I read a positive review of that,” and don’t happen to mention that’s how they found it when they hand their money to you? You can’t associate that to the review.
I certainly wouldn’t argue that there are dozens of sales driven by Fred’s reviews, or mine, or anyone else’s, that you’re missing. But I wouldn’t be surprised if there are at least a few sales you’ve made over the years influenced by reviews that simply aren’t that easily tracked. I know that my own book buying habits are influenced by reviews to some degree; the best book I’ve read so far this year was The First Fifteen Lives of Harry August and I found it because reviewers were talking about it. I’ve found music through reviews. If I’m on the fence about a movie, checking a few reviews may well push me to one side or another.
Between “my friend said he liked thing X” and “this guy on a blog said he liked thing X” and “this reviewer on Amazon said he liked thing X” and “the critic in F&SF said he liked thing X,” it’s “this reviewer on Amazon” who’s the least trustworthy out of that set; we’re essentially betting that if there are enough reviews on Amazon, the aggregate will be a reliable indicator. This sort of works, but only sort of. Niche stuff – like furry – is likely to have inflated ratings due to the “it has a skunk, five stars” syndrome that Kyell talked about. From a publisher/author standpoint, that’s awesome. From the standpoint of a reader who will not buy solely based on his fursona species, it’s not so helpful. I don’t actually care that it has a skunk (or even coyote). If I don’t get a recommendation directly from a friend, I’m going to want a good, well-written review – or better yet two or three. I’ll take the ones on Amazon, but I will also Google “[title of thing] review.”
Maybe I’m an outlier. (It wouldn’t be the first time, let me tell you.) But the evidence from my site analytics is that there’s a hundred regular readers or so down with following a furry-specific review site. It’s not a huge audience, but it’s still an audience – and it’s my suspicion that even a hundred people who have implicitly declared themselves to be readers of furry fiction are kind of a big deal give the sales numbers most furry authors have.
Beyond that, there’s a question about whether the sole purpose of reviews is to drive sales. If that were really what it was all about, then why are so many authors in this thread worried that reviewers are afraid to be critical? If reviews are only there to drive sales, we shouldn’t want negative reviews! A culture that supports well-written, thoughtful reviews is arguably a culture that supports elevating the overall quality of furry fiction. I’d love to start seeing books from Sofawolf and FurPlanet get noticed in Locus. I’d love to see them on non-furry recommended reading lists. And I’d like JM to eventually find a furry book that blows his socks off, and not because his socks are unusually loose that day.