I don't really know any hard and fast rules of how to beta-read a manuscript.
Thought I was gonna say something sagely and wise there, didn't you.
But anyway, I HAVE learned a thing or two about how to beta my way, and I figured I can share them with you.
Thought Number One: Be sure to establish an understanding of your duties as a beta reader with the author.
Does the author need your notes and impressions by a certain date? What kind of feedback do they need? Are they interested in grammar and punctuation nitpicks, or do they just want comments on the story? Are they trying to get something in shape for consideration by a publishing professional, or is this story for fun? Are they going to be receptive to very large or very critical edits? Sometimes you can upset people or sour your relationship with them if the comments you offer are of an unexpected nature in some way. So go ahead and ask your author what kinds of comments they can use.
And between you and me (she says, to the Internet), if you're betaing for a newer author and you feel this is more an instructional situation than an exchange of opinion between career equals, you might consider being a little less critical even if it kinda needs it. There is a LOT you can't teach a writer when they're new, and you might as well not kick their heart out from under them. They will learn in time the longer they keep at it, and if you murder their manuscript the first time they ask someone to look at it, they may really be shocked and disheartened.
As an extra note here: Please, please don't volunteer to beta for someone and then just disappear without explanation. We understand that betaing is generally a free/volunteer service we perform for each other, but if you thought you would have time and you didn't, or you're just not able to complete what you said you would, let the author know!
Thought Number Two: Give the author some sense of what they're doing right.
The most important function of a beta reader is to iron out kinks and give impressions, of course, but giving compliments can help a LOT with perspective even as it also functions to keep morale up during a potential pounding. Sometimes if you help a writer see what you're enjoying, they'll get some understanding on how to reinvent the parts of their story that need more massaging (or massive overhauls). And you don't have to do it in a patronizing way, like "this is good, this is the correct way to do scenes like this"; just do stuff like laugh at the funny parts, tell them when you're excited or anticipating the next plot point, yell at the characters when you get caught up in their lives, quote and praise pretty turns of phrase, and definitely identify where the author moved you.
Do everyone a favor, though, and don't heap praise on them if you don't really mean it just to cancel out whatever bad stuff you might want to say. Here's the thing. It sucks to have to be the one to tell someone that their book isn't ready or that it has a serious problem, but think about the people who have been terrible failures on talent shows who embarrassed themselves in front of a nation. Why didn't some kind soul in their family tell them they can't sing? Why didn't their boyfriend tell them their dance needs more practice? How cruel! It's kinda awful to find some tactful way to tell someone their work needs work--especially if the person believes their masterpiece is an exquisite work of art that needs no improvement--but if that person trusts you as a beta reader, you've got to tell the truth. I would hope that everyone I've betaed for accepts that I've been truthful, and that they can trust me if I say they're ready, too.
Sometimes authors will take ANY criticism poorly, I'm afraid, and there's no real way to avoid that. There are nice ways to say almost anything, but you will occasionally run into an author who only wants honesty if it's positive. Be ready for that, and if the author defends their choices aggressively instead of taking fair criticism, either gracefully back out of the project or be willing to only give praise for the rest of your involvement, trusting that they will learn the hard way from industry professionals if they won't listen to you.
If you don't like something in the book because of a personal pet peeve, say so. And say it's a personal pet peeve. If you don't like something in a book because it's offensive, or it gives incorrect information, or it's potentially too tropey, or you don't understand the characters' motivations, or you hate a character you're not supposed to hate, or you wanted to see fulfillment of a plot point and you didn't get to, or you're bored, or a character action comes out of left field and makes you feel like you misunderstood them or missed a detail . . . TELL THE AUTHOR. TELL THEM.
If you just say you didn't like a part, they won't know how or why to fix it, and they'll be disinclined to do so. But if you say WHY, they can learn something, and they have a better opportunity to determine whether your opinion is likely to be shared by others and therefore likely to need attention. But GIVE YOUR THINKING. Your reasoning can help another author SO MUCH. There's a huge difference between something being bad and you not liking the thing, and the author gets to decide how to interpret what you said. And if you, as the beta reader, KNOW it's just something you don't like, you can avoid saying anything at all unless you think it's really important for some reason. What might be better is figuring out what the author is trying to ACCOMPLISH, and measure whether they ACCOMPLISHED THAT, not about whether they satisfied you.
Keep in mind that if you're a more accomplished author than the person you're betaing for, your words will have more weight, so tread gently. And if you're betaing for someone more accomplished than you, you might feel like you don't have the right to criticize them--which is hogwash. So when you phrase your thoughts as your perspective and avoid making it prescriptive, you can assuage the awkwardness that comes with either side of this issue.
This rides a little on the heels of Number Three: if you give the author perspective on why something of theirs isn't working, they already have the tools on how to fix it. Their instincts, as the creator of the material, are going to be more authentic than yours. And especially if you are the author's mentor or a senior author, they may feel like your word is law. You mustn't abuse this power. Telling them how to fix their book in very specific ways robs them of the experience of developing their own solutions.
Here's an example. Let's say you have a character who's distant and hard to relate to. Easy way to make the character more relatable? GIVE THEM A COMPANION. A companion can be really helpful in having someone for the character to bounce thoughts off of so the reader can understand them better, and has the added bonus of maybe making the character seem more accessible through the act of showing they at least have one friend. BUT! What if the author conceived this character as having no friends for a reason? What's your REAL problem with the character? It's that you find them distant and difficult to read. What if you say that instead, and the author decides to deal with this by giving the character a journaling hobby so you can hear her thoughts, or by giving her a pet to soften her up? If you plant the idea in the author's mind that the character's best option for being understood is to give her a BFF, the author might not solve the issue in their own way, and you'll have just put your footprint on someone else's work, possibly to its detriment.
So make sure you tell authors HOW YOU FEEL, not WHAT THEY SHOULD DO ABOUT HOW YOU FEEL. "I got bored during the action sequence," not "put more lasers and near-death situations in the escape." "I didn't really understand how his craft works," not "how about if you make him teach a class on the craft so readers can learn it too." "I got confused meeting all these characters at once," not "delete some of these characters, there are too many."
And finally, Thought Number Five: Ask questions!
And I don't mean leading questions, like "Don't you think this would be better if Protagonist ended up with the OTHER love interest?" Ask them general questions, specific questions, questions about the world, questions about the characters--especially if you feel like maybe the author doesn't know the answers or hasn't thought about why they should have answers. Ask questions! How does that character feel about her sister? What would they wear to a dance? What would they have done if the trial had been decided the other way or if they'd never recovered the treasure or if they'd lost the contest?
You might consider looking up some questionnaires with prompts for the author to ask YOU as well--exchange a lengthy e-mail, set up a phone or Skype call, whatever works for you--and have a real conversation about the book where questions and answers are exchanged. You may actually want the answers to the questions you ask (and write them down as you read!), or you may just want the author to come up with answers as an exercise. And for the author, they may want to ask you stuff like "where did you put the book down, if you did? did you skip ahead anywhere?" and "was there a character you wished had more time onstage?" and "did you have any favorite lines or favorite scenes?" and "did you like how it ended or would you have wanted it to end differently?" You can ask the author why they chose to do certain things--things you like and things you don't. Ask!
As a beta reader, you're entrusted with one of the most important jobs in the industry--being a fresh perspective outside the author's head that can help bring a better version of an author's story out to the reading public. Good luck! . . . You'll need it!
[Goofy writing comics from webcomic So You Write.]