The 5 Most Common Writing Mistakes That Break Reader Immersion

Today marks the publication of the 50th review in my ImmerseOrDie indie book review series. For those who don’t regularly follow it, the premise is simple: every morning I step onto my treadmill, open a new indie ebook, and begin my daily walk, reading the book for as long as I can maintain my immersion. When that immersion has broken three times, I stop, and write up a short report of what caused my attention to wander. This article today is a reflection on the first 50 such reviews, and a synthesis of A) whether or not I’ve been consistent in my evaluations, and B) trends I’m seeing in the causes of those immersion breaks.


To begin the analysis, I started with a simple graph. How many of the first 50 reports lasted less than 5 minutes, how many ran between 5 and 10 minutes, how many from 10 to 15, and so on. For most measurements of human behavior, this kind of analysis will produce a bell-shaped curve, with a “hump” near the most common value. But to my surprise, when I produced the graph, it showed a distinct secondary hump at the right-hand end.



At first, I wondered if this might be evidence that books were falling into two primary categories: the so called “weak ” ones, and the “strong” ones. But in thinking about it further, I don’t think that’s what’s happening. If I had not imposed a 40-minute limit on how long I am willing to read, and had instead simply timed how long it took for me to reach 3 WTFs, however long that took, I think we would be looking at at fairly classic skewed bell curve, with the tallest bars clustered around 8 minutes, and then the rest tapering off toward the right-hand tail.

But my policy of stopping at 40-minutes effectively pushes all those right-hand bars together into one big clump at 40 minutes. The slight rise in frequency counts in the 30- and 35-minute buckets give the impression that the right-hand end of this graph might actually be a second high-frequency cluster, but I’m not yet convinced, since there are only five books that fall into that region. For now, I’m going to proceed on the assumption that this is an “artificial” secondary cluster, and that this is actually a classic skewed bell, however, I will probably repeat this analysis at the 100-review point, and at that point, we’ll have a better idea which way things are trending.

One thing worth noting in this graph is that, of the books I found to be problematic, most of them seem to make their flaws evident fairly quickly, with about 2/3 of them having been identified in 12 minutes or less. Writers should pay close attention to this. We all know that the beginning of a book is crucial in grabbing the reader’s attention, but how do we define “beginning?” Well, in the case of myself as reader, the definition appears to be “12 minutes,” or appx. 4000 words. You have that long to prove to me that I am in good hands, and that the story is going to take me somewhere I want to go. Use that time wisely.

And if you ask me, that is an entirely sobering number.


One thing that has been a concern for me from the outset is whether my tastes or my process would shift over time. Would I gradually become more jaded? Would I go on long “failure” jags, getting increasingly more desperate to see a survivor? Would I get frustrated and increasingly irritable? To examine this, I’ve created another graph, showing the survival scores over time, with the first report on the left, and today’s report on the right.


To my surprise, however, there has only been one change in the distribution of scores – and even that does not appear to have had any substantive impact on the results. Of the 9 books that have survived for the full 40 minutes so far, half of them occurred in the first month and half in the second. Similarly, the 6 contenders (the books that came close to surviving) were also split evenly, with 3 in the first month and 3 in the second. So in that sense, there has been no appreciable change in my standards.

That trend of consistent classification holds true for the more problematic books as well, but with one small change. We still see the same number of books being classified into this group, but it appears that I’ve been spending less time making that determination. Their survival times have gone down, with the average score in the first month being 13.1 minutes, compared to only 9.8 minutes in the more recent month.

At first glance, this suggests that my patience has ebbed for the books with the most serious immersion issues, and that, on average, it’s taking me about 3 minutes less to pull the trigger. If true, this would be something of a concern for me, and I’ll definitely be watching this stat in the future. But while there may be some truth to that, I think there might be a slightly different (and less worrisome) explanation.

When I began, I would often feel guilty about pulling the cord too quickly, and would continue reading beyond my third immersion break, hoping that maybe I’d overreacted and that things would smooth out. But I think that what’s really happening is that, over time, I’ve simply become more willing to trust my instincts. If that’s true, then what we’re seeing here is not so much that I’m now putting less effort into the problem patients than I used to, but just that I’m spending less time in denial after the patient has flatlined. As I said, I’m not certain of this analysis, but that’s my gut conclusion, and I will be monitoring this trend over the coming weeks.

But enough about the time stats. According to my recent straw poll, most of the readers of the ImmerseOrDie Report are authors looking for tips on how to keep readers engaged, so let’s turn our attention now to what you can learn from the actual WTFs themselves.

The Taxonomy of WTFs

Over the 51 reports written to date, I have cited 131 immersion breaks, which I’ve now classified into 27 different kinds of errors, which I call “WTF moments.” (This classification process is an ongoing thing, so the categories may change slightly in the future, as more WTFs occur and more obvious clusters emerge.) The following table shows the current list, along with a brief explanation of what the error means, sorted from most common to least common.

17weak mechanics Simple editorial issues such as spelling, missing words, grammar, etc.
15implausible character choice When a character does something contrary to his/her established traits or in violation of basic human nature.
14echoing When words or sentence structures repeat frequently, in a way that calls attention to the pattern.
10illogical world features Aspects of the world building that do not bear scrutiny.
9conspicuous exposition Presentation of backstory in inappropriate places, or in dense passages, or for insufficient story reasons.
6weak language style Poor execution of linguistic styles, such as bad accents, incorrect historical language, etc.
6tell mode Overlong passages of telling instead of showing.
5weak dialogue Words put into characters' mouths that are boring, or inconsistent with established character, or unrealistic human speech.
5conspicuous coincidence Important plot points resolved through unlikely or convenient concidences.
4word misuse Employing a word that does not mean what the author thinks it means.
4past perfect Missing or incorrectly applied use of the past perfect tense when needed.
3weak pacing Stories that go too quickly, or too slowly, or with a fixed pace that does not change, to the point that it attracts the reader's attention.
3insufficient exposition Not enough information given to the reader for him/her to follow the story. (More extreme than simply creating mystery or intrigue.)
3inconsistent time flow Events related (unintentionally) out of time order, or with confused tenses, or with effects happening before causes.
3inconsistent tenses Switching between past, present, and future tenses without apparent reason.
2conspicuous borrowingAny story element that seems too recognizable as the intellectual property of another author that is not being used in a satirical or referential way.
2weak proprioception Insufficient description of the scene and characters' relative positions within it. Often results in 'talking floating heads' syndrome.
2weak logic Explanations that do not bear scrutiny.
2ungrounded pronouns Use of pronouns for which the referrent is unclear.
2pointless scene A scene (or especially a prologue) that adds nothing of substance to the story.
2morning ritual A cliched story beginning in which the protagonist walks through the usual events of their morning. Waking up, brushing teeth, etc.
2weak noun coinage Author has created names for people, places, or things that do not seem appropriate to the story or the world.
2show vs tell mismatch When the author tells us something is happening, but then shows it in a way that does not agree.
2inconsistent characters Characters who act or speak in contradiction to either their established personalities, or the situation at hand.
1whiny narrator Self explanatory.
1present tense I simply can't immerse into present-tense stories. It feels silly to me.
1missing explanation Something important that happens in the story that is either not explained, or is insufficiently explained.

To me, it comes as no surprise that the most common citation was for weak mechanics. This is the #1 issue for which the indie publishing movement gets criticized, and unfortunately, the ImmerseOrDie reports confirm that these criticisms are warranted. Authors, please ensure that you are putting sufficient energy into removing this single most damning form of errors from your work. It’s not just about protecting our collective reputations, but (if my evidence here is at all reliable) it actually means that more people will be able to enjoy your stories and actually finish them.

What is a little more surprising, however, is that, of the 28 problem-types I’ve cataloged, just 5 of them account for fully half of the WTFs logged to date. Those top five gaffes are:

  1. weak mechanics (spelling, grammar, etc.)
  2. implausible character behaviors
  3. echoing words, sentence styles, and images
  4. illogical world building
  5. conspicuous exposition (info dumping).

If you are an author who is looking for the highest-impact place to sharpen your immersion blades, examining your work for the above five problems would be an excellent place to start. And if you can’t see it for yourself, ask your trusted beta readers specifically whether they see any of these problems in your writing. (The response you get from readers will be much more pointed and critical if you ask them to look for specific issues.)

The Shocking Discovery

But, in looking at those five high-impact issues, I noticed something odd: There is no obvious pattern. No master classification that seems consistent across all five. Some are problems of a simple editorial nature, while others are fundamental to the conception of the story itself.

This got me thinking about the fact that these issues can be organized at an even more fundamental level. I often think of the process of fiction writing as being arranged into 3 fundamentally distinct skill sets: story building, story telling, and text editing. And it takes mastery of all three of these areas to produce an engaging story that fans will love. So with this in mind, I’ve classified those 28 error codes even further, into those three fundamental categories:

  1. Story Building Problems: These are weaknesses in the story design itself. Examples include tired old cliche plots, illogical economic systems, illogical or impossible physics, inconsistent or unbelievable characters, etc.
  2. Story Telling Problems: Here we find the problems related to how the conceived story is translated and organized into text. This accounts for things like bad pacing, clichéd scenes, bad dialogue, show vs. tell, and so on.
  3. Editorial Problems: These are the problems that could have been avoided with better copy editing. Spelling, verb tenses, missing words, words used incorrectly, etc.

Now here’s where my expectations got entirely kicked in the face. When we draw a graph of the WTF frequencies, grouped by those fundamental categories, I was absolutely shocked to see the following results. Yes, mechanical editing is the single most common WTF type that I’ve charged. But as a group, editing flaws are the least frequent, accounting for only 25% of all the WTFs I’ve assigned. Instead, problems with the way the story is being told are far and away the largest culprit, accounting for almost half (44%) of all the WTFs. And problems with the planning or design of the story account for an additional third (31%). So, combined, 75% of the problems I encountered were for issues that have nothing to do with copy editing.

Want an even more in-depth resource with over 50 different immersion busters, including explanations and examples for each? Check out the next report in this series.


Yes, it’s true that a good copy editor will actually help you with some of these other issues as well, but technically speaking, that isn’t their job, so it’s not a good idea to rely on them for that. To deal with issues of characterization, plot logic, world logic, etc., you need to bring an entirely different team to bear. A developmental editor can help you with some of it – pacing, event sequencing, character consistency, etc. – but who are you going to get to help you fix a broken world-design? There’s no easy answer to this, I’m afraid. At least, not one that I can think of. Some editors have a good feel for those kinds of things, but many do not. A better strategy, I think, for authors who work in speculative fiction worlds, is to assemble a kick-ass squad of alpha readers specifically aimed at big-picture analysis – folks who can take a very early draft of your work and who can look past all the mechanical issues that may still be there, and help you work out the world-level problems before you’ve invested in writing drafts 2 through 7.

But there is one other person who can (and should) be called upon to help polish these kinds of problems out of your work. And that person is sitting right there, right now, inside your very own pants, reading this article.

In my opinion, the authors who make these kinds of mistakes are not doing so because they are incompetent at these aspects of writing – they are doing so because they are completely oblivious to the fact that these issues are even a thing. They’re just inexperienced, and they have not yet become sensitive to this dimension of their own work.

So if you weren’t previously aware that readers pay attention to story logic, details of your invented worlds, realistic character behaviors, and so on, don’t allow yourself to continue being that kind of author. Become aware. Have a look at the above list of WTFs, or read through my previous reviews. How many of those WTFs being charged are ones that you weren’t even aware of as a potential problem? And if you do understand them all, do you know whether your work suffers from any of them? You simply cannot fix a problem that you do not realize is a problem, nor can you fix a problem before you know that your work suffers from it. So your number-one tool for improving audience engagement with your work is a healthy sense of self awareness and a willingness to do something about the problems you find. Ask some friends what they think. Conduct a poll with your alpha readers. Or hey, submit your work to ImmerseOrDie, and I’ll tell you what I think. (The rules and submission form are here.)

Well, that brings us to the end of my first substantive analysis. I hope you’ve found this useful, and that these ImmerseOrDie Reports are helping authors learn what sorts of things to watch out for in their work. And hopefully, when you finish your next book and send it in, yours will go straight to the hallowed halls of 40.

If you’ve found this article useful, or the series itself, please consider sharing a link with your reading and writing community. It really does help keep me motivated to continue the series.

Take the Pepsi Challenge: Curious about whether my own work would survive the treadmill? Why not download one of my free short stories and decide for yourself?

Why Present Tense Bugs Me In Fiction

About the author

Jefferson Smith is a Canadian fantasy author, as well as the founder, chief editor and resident proctologist of ImmerseOrDie. With a PhD in Computer Science and Creativity Systems compounded by a life spent exploring most art forms for fun and profit, he is uniquely unqualified in just about everything. That's why he writes.