Could text analysis tool The Bookalyser be a copy-editor's new best friend?

Could text analysis tool The Bookalyser be a copy-editor's new best friend?

This brand new copy-editing tool can scan a whole book in 20 seconds to analyse style, patterns, consistency and errors.

The pitch 

The Bookalyser is a new automated copy editing and manuscript assessment tool. It analyses complete books at high speed to help authors and editors spot sloppy writing and style errors. A future iteration is also in the works to help agents and publishers tackle the slush pile.

Why is it innovative?

The team claims that The Bookalyser is the fastest tool for analysing an entire book in one go. It typically takes about 20 seconds for it to reveal insights into word usage, and identify errors which are very hard for humans to spot.

"Right now it offers a ‘big data’ approach to copy editing – computers can, for example, conduct frequency analyses on a text in a way that people can’t," creator Andrew Chapman reports.These can reveal patterns to an author’s style, and indeed bad habits such as repeating certain phrases or tending to start sentences with the same words. The big data approach can also compare the author’s usage against a background corpus to see what stands out."

He also believes it has unprecedented accuracy in identifying inconsistencies across a full-length work, such variations in the use of hyphens, mixing British and American English, variant spellings, inconsistent house style and so on.

"The Bookalyser brings style analysis and error spotting together in a new way, and opens up possibilities for more sophisticated analysis of narrative elements such as pace and point of view," he insists. "Of course, we would never claim that a tool such as this can replace human editors – but what it does is makes their work more efficient, which saves everyone time and money. It’s another tool in their professional armoury."

The team

The Bookalyser was conceived and coded by Andrew Chapman, an Oxfordshire-based freelance copy editor who co-created the popular book recommendation website back in 2005.

"I found myself adding features to some old code I had written and using it to run checks on books, and it has snowballed from there," he says. "I also like online tools which are really quick and easy to use."

He now has& editorial support from C M Taylor, a novelist and associate lecturer at the Oxford International Centre for Publishing Studies. Taylor has a particular interest in narrative - he has written structural reports for authors and publishers on well over 200 novels, and the writing process of his novel Staying On (Duckworth, October 2018) has been documented in a unique experiment with the British Library, whose digital curators installed ‘spyware’ on his computer to record every keystroke.

What's the gap in the market? 

There are various add-ons or macros for Microsoft Word which can help writers with issues such as consistency checking. There are programs, too, which aim to help them improve their style. But these tools tend to be slow and require users to split the text up and analyse a chapter or a chunk at a time. Chapman believes he's found the fix.

"I wanted to have something which is ‘blind’ to specific tools such as Word and works just as well with any other software people might use, such as Scrivener, InDesign, or indeed anything that can copy and paste. And the other key thing is that my colleagues and I are editors, working on text every day – so we know exactly what would be useful in our work."

Success so far?

Bookalyser is freshly launched, so there are no stats to share yet, although Chapman reports that feedback from beta testers has been very encouraging.

The team is now focusing on devising ways for the tool to help publishers and agents. Last year they joined forces to provide a full outsourced editorial service to publishers such as Unbound, in conjunction with a team of freelancers. "We’re encouraging them to use The Bookalyser to add an extra layer of polish to the services they provide," Chapman explains.

Biggest challenges?

The biggest challenge is around how to provide a wide range of analyses without becoming cripplingly slow. "I have seen other software completely give up when analysing a whole book in one go," Chapman says. "The code for The Bookalyser is very lean, and actually runs thousands of checks on a book. But every time I add something else, it comes at a temporal price! The challenge then is to balance speed and efficiency with usefulness."

Ultimate ambition?

The team's ultimate goal is to provide extra features which help agents and publishers to tackle the slush pile. This means a whole new level of analysis – looking at narrative elements, genre, potential audience and commercial viability. "We believe there are ways to use textual analysis and nature language processing to help with a lot of this, as has indeed been proven by Josie Archer and Matt Jockers in their book The Bestseller Code," Chapman says."So we’re working on our own take on this to create an automated manuscript assessment tool which will help to flag up books among the thousands out there that might profit from further investigation."

Advice to other publishing entrepreneurs?

"In short: try things out. Get something out there, and iterate as necessary, rather than feeling the need to do everything immediately. That way you can learn from mistakes and grow as you go!"