AI Writing Assessment

One of the biggest challenges for new authors is getting feedback on their first manuscript. Not only is it a nerve-wracking leap to get the first new set of eyes on your work, but also, you don’t know if your family or friends are really objective in their feedback. And it’s crucial to get initial thoughts before your manuscript goes to an editor. Is it worth the larger investment in publishing? What kind of editing does it need? How can you improve it yourself? How do you know what issues you need an editor to address? It’s important for authors to get answers to these questions before committing to and paying for an editor.

At Izzard Ink Publishing, our manuscript assessment is our answer to this problem - and a key component of our solution is the AI writing assessment offered by Fixional, in addition to assessment by two seasoned editors. Crucially, it adds an objective layer of analysis to the feedback from our two editors and an additional source of feedback for authors at this precarious and crucial stage of the writing process.

What is AI writing assessment?

Now, if you’ve never heard of AI analysis in publishing, you’re probably wondering exactly what that means! Fixional and other companies offer AI writing assessments for manuscripts, to give an idea of overall quality, originality, and readability issues before the editing process. AI has already been evaluating writing for years - look at Grammarly, or plagiarism checkers employed for college classes. But Fixional and similar services take things several steps further, assessing not only originality and sentence-level errors but also overall craft, readability, and more. It can evaluate whether a manuscript accomplishes its goals, and even offers recommendations to improve writing.

Primarily, it’s set up to supplement conventional editing and get a first pass assessment of a manuscript, with an eye for clarity, grammar, narrative potential, and inconsistencies. Paired with our assessment by two editors, Fixional provides additional screening for quality.

And part of the appeal of AI analysis is that it quickly assesses an entire manuscript. In order to make sure authors can get professional feedback on their manuscript quickly and affordably, Izzard’s two editors assess up to 7000 words of a manuscript. This does give a solid sample of overall quality and it’s enough to determine what kind of editing is needed. Often, the thorniest writing issues will appear early on in a manuscript. But having AI that can almost instantaneously assess an entire book is invaluable. The analysis may pick up on issues those editors miss, and provides a look at global problems that affect the entire manuscript. Authors can also rest easy knowing the AI analysis is objective, and not colored by an editor’s personal preferences or scope of experience.

 

How it works

Fixional started several years ago as a publisher—they were inundated with writing, and in search of a solution to the gargantuan task of properly assessing so many submissions for quality in a timely manner. They looked to AI for an answer and ultimately decided to focus exclusively on developing the software, especially as users became more open to AI in new roles. After all, if we’re seriously considering the idea of having AI drive our cars, why not give it a chance to assess writing?

Fixional’s algorithm uses pattern recognition and machine learning to develop an “eye” for what to look for in a quality manuscript. Fixional has fed its algorithm a variety of writing to use as a reference point, and over time, the algorithm “learned” what to look for. This machine learning technique has been employed in a wide range of roles in recent years, with impressive results.

 

Can AI really assess a book’s potential?

While authors might assume that AI analysis would miss the bigger picture potential, focusing only on mechanical and grammatical errors, Fixional goes beyond simply finding mistakes to analyze quality—it actually searches for narrative potential beyond the simple number of grammatical errors and typos. In other words, if you’re a great storyteller with a tendency toward typos or even unclear sentences, Fixional will notice—a testament to how far AI has come in recent years. It’s not just looking for mistakes—it’s looking at whether the manuscript accomplishes its goals. It can help determine, if typos and grammatical errors were fixed, the underlying quality of each manuscript.

It also assesses tone and can even provide authors with specific suggestions and insights on how to improve their writing. Even if you’re about to have your book edited, it’s important to get your manuscript as refined as possible in order to get the most out of editing. If you make sure the majority of easy-to-fix mistakes are taken care of, your editor can focus their time and effort on fine-tuning—not just turning a flawed manuscript into something passable, but turning what you have into a real home run.

Many hybrid publishing companies will be all too happy to simply take your money and move forward with publishing whether or not the book has potential. It’s in every author’s interest to ensure their manuscript has potential before investing in costly steps like editing, cover design, and marketing. Even the best editing can only go so far. And with AI analysis, you’ll go in with a sense of whether your manuscript needs a deep (but pricier) developmental edit or a simpler copy edit—or whether or not you need to go back to the drawing board altogether. AI is a way to arm yourself with the knowledge you need to make the right publishing choices. It may not be able to replace feedback from humans, but it can supplement it in ways that most authors never would have thought possible.