A good review is one that informs the reader accurately and honestly and fairly as to what they might expect in a work. If I was to read the review, then immediately look at the text in question, would the review have given me a good overview?
But, at the same time, it is more than a synopsis and it is more than a marketing blurb.
The key part of a review is that it is a critical evaluation. Anything that doesn't at least attempt to critically evaluate something is a review in name only. If I vote No on a review, it's because I feel there's no critical eye on it, little attempt to evaluate anything, no reasoning provided, and so on. If you want to say something is bad, explain why you feel that way. If you want to say something is good, there's maybe a little bit less explanation required. But still: show your working.
Because if you're informing the audience accurately and honestly and fairly, then you're also informing the author of what their strengths and weaknesses are. The higher your star rating, the more I expect to see.
Ideally, a review would cover the entirety of the work. This can be difficult, especially given that some serials have been in progress for years and really don't show much sign of ending, and feel like they might not actually end until the author gets bored (or, worse, burns out).
I'm not one for brutal honesty, but I'm not going to avoid what I feel is necessary honesty. Here's the thing about writing in my eyes. It's a very lonely work and it's a very personal work. We are, at the end of the day, close to what we create. We are, at the end of the day, blind to a lot of the issues with our own work. And it's probably better to have those issues brought to light now, from volunteers, before you start trying to angle for exposure or publication.
If you're making mistakes -- obvious basic mistakes -- then I think a reviewer, in the process of critically evalulating your work, has a responsiblity to at least mention them.
Some reviewers will mark based on intention or enthusiasm or effort, but I feel this is very subjective. As I mentioned in the other thread, while overwhelmingly positive feedback is nice, it doesn't actually help someone improve. Yes, the author puts in effort, yes the author is trying, yes it is probably coming from the heart -- but that doesn't mean it gets five stars, or even four.
It might not even mean they get three.
Sometimes negative feedback stings. Sometimes that's warranted, other times it stings because you can't separate the work from you. Sometimes someone does bash your work (and I've seen reviews that I'd vote helpful, just because they accurately summise the often-overlooked flaws of a work, if not for the vitriol) and sometimes people have good points that disrupt the so-called 'hug box'.
Sometimes those people are right. Sometimes they are wrong. But you won't know either way unless you hear it, acknowledge it, and reflect on it.
If an author truly doesn't think a critical review would help them improve, then I'd be curious to see if they can publish what they have without alterations. Because there are two main points that inform my reviews here. We are all working with rough first drafts, sometimes written immediately prior to update, and we are all amateurs. Talented ones, perhaps, but amateurs all the same. Some of us are fine with writing our serials, but I'm pretty sure most serial authors that pick up an audience dream of genuine publication.
So, what makes a good review?
A good review is from someone who has read as much of the story as possible. It is accurate, honest and fair. It is a critical evaluation which talks about the strengths and weaknesses of a given work, showing the working and evidence. If anything, it leans towards being critical than being glowing. It does not bash a work, but it also doesn't puff it up.
Note something I didn't mention there.
Because far, far below every other criteria within my rubric of what constitutes a good review is the final part of it: making the author feel good.
Like above, if you truly feel your first draft warrants nothing but five-star ratings, then you may need to examine your thoughts (and your work!) more closely. This also goes for the opposite, too, where if a critical review saps your motivation, perhaps you might need to wonder about why you want to write. Because if you do dream of selling your work, you're going to need to be prepared for the fact that people get far more vitriolic once they've paid for something.
And, once the audience is paying for something, those reviews matter far more than the ones you'd get here.
So, I think that covers what I personally feel makes a good review, and my reasoning behind why I review how I do. Let's look at another thing that I feel makes a good review: accurate usage of the star rating system.
.5 - 1.5 - unreadable (literally what it says, can't read, difficult to determine meaning, no redeeming qualities.)
2 - a tough slog (can get through it but it's what it says, a slog. probably had to fight the prose or make an effort to finish, serious flaws)
2.5 - almost worth a look (it has a few too many flaws to recommend. honestly, very similar to 2, but a bit kinder)
3 - worth a look (it's okay. has good points and bad. you might like it if you appreciate or are familiar with the genre or author)
3.5 - fairly solid (more good than bad. but it's only 'fairly' solid -- it has some noticable, glaring flaws)
4 - solid (generally good. if it has flaws, they aren't particularly glaring)
4.5 - compelling (see below)
5 - exceptional (for me, probably a theoretical rating. a work i'd give five stars to would be one i'd be surprised to be getting for free)
Compelling is the odd duck in this list. Something can be technically great but not compelling. Something can be compelling but technically unimpressive. Compelling is very subjective. It's very easy to say, well, did I find this story compelling and leap to 4.5, and sidestep what makes you think it is compelling in the first place.
I don't think I'll ever rate below 3, if even just the review comes with a disclaimer of 'worth a look... if you're a fan of [genre/thing]'. Because maybe my issues are really subjective and people who like [genre/thing] might like it. If my review starts getting below 3, I'll generally wonder whether I should publish it simply because the negatives are going to outweigh the positives.
However, like a lot of review systems, 4/5 is assumed to be default 'good' and everything below it is bad. A good review acknowledges that 3 is the average and works with that in mind.
Now, here are a bunch of things, presented in no particular order, that are basically things I'll probably vote No on.
Extended metaphors or vignettes, What This Serial Means To Me, My Life When I First Read This Serial, etc. This isn't a creative writing class. This tells the reader more about you than the work you're ostensibly talking about. That's the kind of thing the author might like to recieve in a nice email or message, however.
Comparing the serial you're reviewing to another work, often unfavorably. This reads to me like some form of weird guerilla marketing, especially if the review in question does not contain a star rating.
Saying that any serial is like a combination of X and Y things. This is just lazy, and means nothing if you're not familiar with X and Y.
Effusive praise and/or a rambling synopsis.
Reading more like an advertisement than a review.
Anything that can be summed up as 'read what's available, loved it, can't wait for more'.
Missing any well-known and widely available criticisms of a particular work. If you're going to go hard on praise, then you should perhaps talk about why those criticisms aren't a problem.
Typically, if I can't identify anything like this, and I'm not very familiar with the work the review is of in question, then I won't vote yes or no. If I feel it's an accurate review of a work I know, I'll vote yes. If I don't know the work, but the reviewer has provided an overview that seems fair and balanced, I'll vote yes (and maybe even check out the serial in question quickly, just to be sure).
One thing I really like on RoyalRoad is the way the site has a few scores for each work: Overall, Style, Story, Character and Grammar. In many ways, that could be better than one size fits all.