Typos: How May Are Too Many

— A Guest Post by Chuck Litka

As a reader and/or a writer, how many typos are too many? One, you say? We’re talking about self-published books here. Still one? We are talking in this world, not heaven, right? Okay, having no typos is an ideal to aim for, but realistically, how many are acceptable? Still none? Be that way.

Putting an actual number to the acceptable number of typos (more than none) is hard. The real world measure is likely the ratio of typos to the quality of writing, i.e. is the writing and story compelling enough to overlook the errors, or not? Still, let’s see if we can try to put a number in that ratio.

I’m going to compare software code to writing. A line of software code does about the same thing as a word or punctuation mark in writing – it conveys a bit of information. And software has errors – bugs. A line of code usually has more letters – things that can go wrong – then a word, but we’ll make adjustments for that as we go.

It is estimated that a shipping software program might have an error rate of 3 to 5 errors per 100 lines of code. This is eventually reduced to somewhere around 1 error per a 100, to 1 error per 1,000 lines with bug fixes after shipping. For writing we’ll consider misspellings, wrong words, missing words, double words, missing or wrong punctuation, and missing or double spaces as typos. In writing 1 typo per 1,000 words, or about 1 every 3 to 4 printed pages is pretty high. If we cut that rate in half, to 1 typo in 2,000 words, we’d get 5 typos in 10,000 words or 50 typos in a 100K word novel, roughly 1 typo in every 5 to 6 pages, on average. I think that we’re now nearing the ballpark, if perhaps still a little high. However, if we cut that rate in half again, to 1 typo in 4,000 words we’d have only 25 typos in a 100K word novel, which I think is an acceptable error rate.

While 15 to 25 typos in a typical self-published novel might still sound like a lot, a significant number of them will never be seen by most readers, just as bugs in software do not affect most users. I send my manuscripts out to 5 or 6 beta readers after we do our in-house proofreading who report the typos they find back to me. Some find more than others, but time after time their lists of typos have remarkably little overlap between them. Different readers find, and miss, different typos. Hopefully between them they find most of my typos, but individually, most of them find less than half of the actual typos present. And these are readers looking for typos. This suggests to me that most readers will notice significantly less than the total number of typos in a story. There are, of course, natural born proofreaders who see every typo. And some of them get annoyed when they bump up against them – and mention it in a review. However, these readers are edge case readers, and I’m not sure whose problem it is.

In software, it is possible to achieve a zero rate of errors, and it’s necessary when human life is at stake – think of self-driving cars. To do so, however, is very expensive and time consuming. The same can be said for eliminating all typos in a manuscript. Traditional publishing has a series of procedures with multiple editors going over every aspect of the work to eliminate all sorts of problems. But this is an expensive and time consuming operation, and it is reflected in the cost of traditionally published books. Self-publishing authors do not have the resources to conduct multiple edits by different professionals. And even self-publishers who spend the money to do it “right” and hire professional proofreaders still aren’t guaranteed a zero error copy. I’ve seen a review that mentioned the “typical self-published” typos for a book that I know had been professional proofread. Still, there may be self-published books with zero typos. Indeed, I find them all the time, but then, I’m blind to typos.

Being blind to typos, I likely underestimate how annoying typos are to most readers. And as someone who has never mastered spelling the idiotic language of English, I’ve long since realized that for a person to know that a word is misspelled, they must know the correctly spelled word – which means that the misspelled word actually worked – it did the job of the correct word. The reader simply had to spend a fraction of a second to translate it. However, I understand that for some readers typos derailed and bump them out of the story. Ideally, all authors would like to prevent this from happening. Our name is on the cover of the book, and its imperfections are our imperfections. However, given how expensive it is to even get close to zero typos, I don’t think it’s realistic to expect a traditional publishing level of copy in a much less expensive self-published book. As I see it, when it comes to books, one can either be persnickety or cheap, but not both. Self-published books are cheap. Very persnickety readers might be more comfortable with the more expensive traditionally published books.

So to sum it all up, I feel that if a writer has a process in place that reduces the rate of wrong words and punctuation down to 1 in 4,000 words or fewer, they’re as close to perfection as any reader of self-published books can reasonably expect. And that’s a hill I’m prepared to die on. But what about you, dear reader? What do you think is an acceptable rate of typos, either as a writer and/or as a reader? Am I being too cavalier about typos? Am I being too dismissive of the adverse effect of typos on readers? What is your standard?