• 75

The art of alpha-reader feedback

Alpha-reading… wait a moment. Alpha? How's that different from beta-reading?

For those not familiar with the difference, alpha-reading is that first small group of readers you give your work to who understand that it's not polished. You may have passages that say [insert jargon here]. You may have other bits you've intentionally skipped, or even sections you intend to remove. With beta-readers, you're providing a first draft of a finished-ish product. (Which doesn't mean it won't get rehashed beyond all recognition, but that's what the alpha-readers are supposed to help with.)


First, credit where credit's due. What follows is a modified version of Mary Robinette Kowal's approach to obtaining meaningful feedback.

One important point of alpha-reader feedback is that you are looking for symptoms, not diagnoses or prescriptions. The logic here is simple. You are writing your story. You understand it on a level no one else ever will. If something isn't working, the alpha-reader's job is to tell you that it isn't working. Not to tell you how to do it right. If they tell you how to change it (the prescription), it becomes their work, not yours.

So, to be clear about what you're asking your alpha-readers for:


Symptoms are the details the alpha review is intended to identify. What the reader feels, whether positive or negative. It is based on grading the material using the matrix below. Some symptoms will require an explanation of the reader's response – a clarification of the marking.


The diagnosis is the why behind the symptom. For the most part, the reader won’t need to supply this information, but you may ask for it subsequently. A diagnosis from someone who understands writing and story structure should carry far more weight than a diagnosis from someone who doesn't.


The prescription is the solution, how to fix the problem. It’s the author’s job, not the alpha-reader’s. Except in the rarest of cases, and then from professionals, alpha-reader prescriptions should be ignored.

And now, the matrix.

A – Awesome
An A is for anything that is just perfect prose.
Beautiful wording that needs to be kept, whatever happens to the story around it.
C – Curious
A C is for anything that pulls the reader into the story.
Events or descriptions that make them wonder what happened, or why it happened. It includes anything the reader feels includes a promise of a resolution.
B – Boring
A B is for anything that feels dull and uninspired.
Wording that drags on for no apparent reason, convoluted or unnecessary descriptions, etc.
D – Don’t Care/Believe
A D is for anything that relaxes the story’s grip on the reader.
Events or descriptions that distance them from the characters, that they don’t care about or don’t believe.
Ds will often benefit from an explenation, or at least a subcategorisation.

E – Continuity Error
An E is for a continuity error.
These errors add confusion to the story, uncertainty over details you believed has been defined.
E’s should include a brief explanation of the discrepancy.

This markup can be applied at any scale. Sometimes, it will apply to a few words. Other times, to a scene or more.

Obviously, Bs and Ds are the "bad" ones here, but they're also the ones you really want to have pointed out to you as early as possible. So you can understand them and fix them.

So, with that all said… Helpful? Useful? Let me know if it works for you, or you think the model could be enhanced.

Replies (9)
  • An interesting idea, Karen.

    I think it depends very much on the alpha reader's knowledge of craft. As I apply it, I'm finding that I give comments to explain why something feels right or wrong - often far longer than the thing itself. (I've just given a 31-word reason why the word "suddenly" throws my attention off in the first line of a manuscript.)

    Colour coding might work in addition to the matrix, to show relative importance. But it won't call out the types of dissonance the matrix aims to capture. At least not without a lot more colours, and then we're back to explaining...

    • I had not come across this approach before. Has it been working well, in practice, for you?

      • That's a question better asked of the people I've used it on. (So far, they've told me it was useful.)

        Mary Robinette says that her version is very useful in her writing.

        I've had it applied to a small piece of my writing, but want to see the effects on a full book…

        • How would 'Show Don't Tell' fit into this, if it does?   I hope that's not a dumb question!

          • There's no such thing as a dumb question.

            The answer is fairly simple. The issue with telling is that it distances the reader from the story. As such, it increased the boredom factor. A bit of telling is fine. Too much earns a Bored.

            • 9 friends