I’ve made no secret of the fact that I sometimes go back to earlier, already-published works and rework them. Reworking the early Wren books to be more in line with the genre and reader expectation was what led to my sales breakthrough in 2021. I’ve talked about all the things I changed several times here, here, here and probably other places too.
In summary, I:
- hugely reduced violence and gore
- brought Wren’s emotions far more under his own control
- cut all secondary perspectives to focus more on Wren
- cut large chunks of world building, backstory chapters and Foundation info to focus on flow
- cut all heavy swearing, recreational drug use and alcoholism
- reduced or removed all political, cultural and religious didacticism
- added new, more exciting and immediate opening chapters
- made all threats personal
I did lots more things too, including new covers, titles and blurbs, not all of which I remember. All these changes transformed the uptake of book 1 as well as the readthrough to book 2. Pre-2021, readthrough to book 2 was around 10-20%. To latter books it was even lower, like a couple of percentage points.
Those got better after the changes, but they were still not high enough.
Writing is not a competition with other authors, but over the past couple of years it’s become very plain that my readthrough is a fair bit lower than similar authors to me, in my genre. Those authors report that they get at least 50% readthrough to book 2, as some drop-off is inevitable, but after that they maintain 90-100% of their readers through the series.
These numbers are easy to confirm in a general way. If I look at where Saint Justice is in the Amazon charts, I can compare with the first books of similar-selling authors around me. I then scroll down to where book 2, No Mercy, is in the charts. To find it, I have to go past numerous sequels for numerous other authors, all of whose book 1s are selling comparably to me. Then to find my book 3, 4 or later in the series?
They’re not on those charts, while other authors’ whole series’ are. For some reason, I’m bleeding readers. I could surely be retaining so many more, and funneling them to the latter books which I’m really proud of.
Here’s my rough readthrough rates for this year:
- 1-2: 37%
- 1-3: 30%
- 1-4: 16%
- 1-5: 14%
- 1-6: 14%
- 1-7: 18%
- 1-8: 16%
- 1-9: 19%
Laid out like this, the numbers are abysmal. They’re definitely better than my pre-2021 numbers, where I was hitting a few percent for latter books, and it’s good that after book 4 I mostly retain everyone, but it’s still terrible. 1-2 should be 50%+. Every book after that should be at the very least 40%+. To achieve those numbers, I’d need to at least double my readthrough.
So, how?
That’s the question. One I’ve spent the last few weeks/months attempting to answer, by making changes backwards across the series, culminating in a hefty rework of Wren 1, Saint Justice. I think I’m making these books better. I think this is an incredible learning experience for me as a storyteller, finally seeing all the mistakes I was making that I couldn’t see before.
Only time, the audience and my readthrough rates will be the judge.
So, here’s what I’ve done:
- Removed all cliffhanger endings. This was a big one, and took lots of time and thought, particularly for the transition from book 5-6, which is the biggest, most literal cliffhanger in the series, with Wren left in peril and book 6 picking up in the middle. That’s now gone. Book 5 ends. Book 6 starts a new, though obviously related, story. It’s the same for each book. Going through this process made me remember comments I’d received numerous times, that the Apex storyline was dragging on too long. I think that’s because I ended every book on an unfinished note, with reference to the Apex. It was basically the same cliffhanger again and again. ‘Don’t relax, Wren, the Apex is still out there!’ Now the Apex storyline is pushed further back. Wren has his life. He gets almost total closure at the end of each book. Now more than ever, the books all standalone, which very much brings them into line with every other thriller series I see out there. Nobody is doing long sequential series. I still have that, but it’s in the background more. This also puts paid to a feeling I always used to have when finishing a book, which meant I never felt Wren was really ‘ready’ for the next book – because he’d had no rest. No notes of happiness. Now he does. It’s even allowed me to pay off an early romance that was only ever hinted at, as well as give Wren some closer connection with his family earlier. Because of this, he’ll come to each new story fresh.
- Made things more realistic. Over the years, the biggest hammer of criticism readers have hit Wren with has been the realism issue. They say it’s not realistic, and I didn’t know what they meant. Gradually, though, by thinking about it, trying to engage with such readers as much as I can, hiring editors and beta readers again and again, I think I get it. Everything I put in my books is probably ‘possible’ to some degree, much like the stunts in Mission Impossible are technically possible. They are, however, sometimes wildly implausible. This is what the readers mean, I think, by ‘realistic’. They want Wren to follow more of the same rules you and I follow. So, I’ve been working on fixing this in the following ways:
- 1- Remove as much hacker magic as possible. I didn’t realize I was doing this at the time, but upon reflection I recognize it now – I often had Wren need to know something secret, ask his hackers to find it, and they did. Hey presto. Sometimes he’d just say ‘find me a location to strike’ and they would. No work involved, just an answer via hacker magic. Of course he also uses his hackers as seasoned battle operators (though they’ve only ever played computer games), has them hack things rapidly that would probably take weeks to do, has them hack and control fleets of self-driving cars within hours, and has them crunch magicky excel sheets of mega data to provide him with suspects. Some books are more guilty of this than others, in particular Make Them Pay (they came up with the compound to target wholly on their own, while Wren was unconscious) and False Flag (they make an algorithm to parse all traffic data in LA, combined with witness statements, to track an invisible vehicle for the rest of the book). In order to remove these magic tricks that advanced the plot, I had to replace them with something:
- 2- Add more realistic investigative threads. I did this pretty much any time I removed hacker magic, or Wren stumbling upon a clue by some accidental means. This had to happen pretty substantially in False Flag, which was largely driven by hacker magic. Wren had his hackers do that magic excel sheet thing to track Stephen Gruber then to track the Ghost, but even to me at the time that felt tenuous. I added in a bunch of chapters that saw Wren hitting the street and knocking on doors, ‘glamoring’ people or compelling them where necessary to get the intel he needed. It’s a great sequence now, much more believable I think, and relies on Wren using his skills and putting in the work rather than outsourcing it to magic. This also happens in Make Them Pay and Saint Justice too.
- 3- Take out/play down the wildest stunts. I love stunts and I love Wren going big, but the readers don’t. I guess Mission Impossible-level action doesn’t play so great in book form. If it can’t look amazing on screen, then maybe it just reads silly. Accordingly, I’ve reduced some of the wilder scenes. One that stood out to me in particular was in False Flag, where Wren jumped a Monster Truck over a dam with flags whipping in the wind, into a hail of bullets that largely get blocked by a huge swarm of drones – all put together within minutes. I guess it is kind of silly, though it’s also cool. Now it’s gone. Wren’s chase up the LA River still happens, but focuses on more realistic stunts. Another example is the fleet of self-driving cars in Make Them Pay that Hellion uses to storm a compound (she does the same thing with the Anaximanderian compound in a later book) – I kept the core of that, but reduced it right down. Maybe one or two cars.
- 4- Downgrade the Foundation. I love the Foundation, and I loved that Wren can just call on them. But people don’t get it. Some said, very specifically and helpfully, that Wren should get help from just one source, either the CIA or the Foundation. And as I re-read the books, I found myself agreeing. With the Foundation as it is, Wren is massively overpowered. It’s not realistic. Even to maintain the Foundation is pretty unrealistic. He claims to watch all 100 members all the time, to check if they’re keeping to their coin level, but when? He has a full-time undercover role with the CIA and a family. Coin meetings with that many people alone would take days, let alone the research to have something meaningful to say to them. So – I’ve downgraded the Foundation to make it more realistic. They are far less of a paramilitary force now. What they can offer is expertise and connections. I did this in Saint Justice – where Wren relies on Cheryl blackmailing a cop to get him CCTV footage he needs. There are now no Henry and Abdul and no Alli. Their contribution was overkill and unnecessary.
- Reduce stodge that slows down the story. I think there’s probably lots of this in my writing generally, especially when I’m spinning my wheels and not sure where the story should go. A story should be pretty fast, though. I don’t know how some authors get away with not fast stories. I can’t get away with digressing at all, it seems. Or maybe my digressions are the wrong kind of digressions? Who knows? Here are some ways:
- 1- Remove PVE and make it PVP. These are terms from MMORPG games – Massively Multiplayer Online Roleplaying Games like Warcraft. In Warcraft you can choose a PVE world, or Player Versus Environment, or PVP, Player Versus Player. PVE means you follow the game story, but it’s flat and not dynamic. You spend a lot of time ‘grinding’ out repetitive quests for XP and loot. PVP, however is dynamic, fast and pits you against a real live person. I think about this distinction when I end up writing scenes that don’t have any real sense of threat and provide no tonal change. A big example comes in Saint Justice, where Wren makes a huge effort (calling THREE Foundation members into service, risking their lives and liberty undergoing illegal actions) because he wants to steal all the Vikings bikes before he goes to hit the warehouse. I always loved this idea of Wren stealing the bikes, so I worked it in there, but it really makes no sense. At this point he’s seen the warehouse and knows what they’re doing. Going after the bikes is PVE – pointless grinding. To be properly focused and PVP, he should go straight for his objective, the warehouse. This happens also with the first investigative lead he undertook in Make Them Pay, where he went after Lars Mecklaren on his NameCheck campus. I had that in there because I was mad at social media, but it’s ultimately pretty pointless. Wren makes a huge effort to get to Lars, then basically tells him to stop doing it, then asks Hellion to find him a target and drops unconscious. It’s PVE. So, I altered all that to make Lars more PVP. Wren must have a goal and gain meaningful intel to progress his search, otherwise it’s pointless.
- 2- Reduce non-dialog sections. This is similar to the PVE issue above, but occurred to me later, as raised by Su. She has been reading the new rework of Saint Justice, and saying it’s cleaner and clearer than ever before. Before, she had found herself getting bogged down from the beginning, not knowing what was happening or what Wren was doing. That good reading experience continued until she reached the warehouse-Price PD-warehouse section of the book. In reworking this part, I’d already thought it was quite repetitive, and had reduced it. But still, it spans about 8 chapters early in the book, during which Wren is doing things he’s already done once before, where there is no meaningful direct dialog (other than a phone call – not bad, and text messages – not great), and also no real hints given or moving forward of that plot. In particular, there was a 4 or 5 chapter run where Wren speaks to no one, just searches the warehouse (again), thinks about his past, explains human trafficking, has a freakout and takes drugs, before finally getting the clue to Chicago. I can see more clearly that much of this needs to go. A good rule of thumb may be that there can’t be too many non-dialog chapters in my books – certainly not more than one in a row. It’s dull, foggy, wholly PVE, and lends itself to noodly navel-gazing. Wren’s gotta engage with people in the world, either standing in his way or giving him clues to accelerate him to someone who will stand in his way. That’s the spine of story. All these non-dialog scenes and PVE scenes are like fat laid on that skeleton. Trim them off, and suddenly the story can move. I feel it as I strip them. Add on to that, this morning I got another 1-star review complaining that Saint Justice is ‘difficult to follow’. Maybe this is why. It’s all gotta go.
- 3- Remove transit chapters. This is perhaps another PVE instance, where I would think that because the drive to somewhere is long, the story should fill some of that time a little bit. Probably it shouldn’t. Probably it should just whip to the next event. But instead, I’ve often filled these transit times with more noodly self-reflection, or taking a chance to check up on how various spinning plates are getting along, or indulge in a flashback, or grumble about Wren’s severe injuries. But all of those are pretty boring. Checking up on spinning plates becomes less and less necessary the more I trim out of the stories, because it becomes obvious not much has happened since the last event. The catching up is really just pushing ahead. What happens next? I remember upon writing these chapters, I’d kind of relax, because they’re easy. Writing a thousand words of reflection and emotional baggage is easy. It’s moving things forward that’s hard, and that’s what the readers want. I need to delete all these transit chapters
- Be less mysterious about Wren’s past. I had approached the series with the aim of gradually unspooling Wren’s past book by book, but I think now that that just pisses people off. It makes Wren unknowable and unrelatable – another common complaint. I’ve been reading in the genre more than ever, and most books tell you everything you need to know about the main character very early on. These are not psychological thrillers like Girl on the Train with an unreliable narrator. People want to know their hero and then see them go out with whatever emotional baggage they have, and still kick ass. So I give a little more on Wren’s background, a lot earlier. I don’t conceal it. And this, in particular in Saint Justice, gave me the new feeling that Wren was actually pretty broken and needed help. I’ve wanted that feeling for him before, but never actually felt it myself. I felt responsible for him, and wanted him to get happy. I hope readers feel that closeness.
- Wren is even more in control. I had done a wave of this before, but left in some instances where Wren has mini breakdowns as a major plot reversal. But this is kind of PVE thinking. It’s not very exciting for the main character to have a kind of panic attack and almost commit suicide in the middle of a hunt, or be so stressed out they won’t answer the phone even when the news might be essential, or take drugs and render yourself useless. Now those things don’t happen. Wren is way more unflappable. Emotions hit him and affect him but he doesn’t ever lose it. That’s the whole point of him – he’s in command of himself. No more drugs, no more freakouts.
- Make Wren more likable. This is similar to above, but I find it comes hard. People have said in reviews before that Wren is a hard man to like. They like Reacher, Puller, other heroes, but not Wren. I think this stems from the primary backstory I gave Wren, and how I wanted him to be long-term affected and damaged by both his childhood in the cult and his time in special ops. It makes him cold, afraid of human connection, unable to always be nice and nurturing, the way he should be. But it’s also an unappealing characteristic. As readers, we want a tough guy who can handle the horrors of the world and, to some extent, shield us from them. Like a father figure. But Wren wasn’t doing that. If anything, he was spraying pain out every day, in every way. In an early scene in Saint Justice, he buys kidnap gear from a kid in a roadside store, then confesses that he’s been attacked by bikers and going to find one of them. He even makes the kid complicit by asking him to get him a taxi to the biker’s house. It’s weird and kind of cruel, also selfish. Wren shouldn’t do things like that, so I switched it. He lies and says he had a hiking accident, no big deal. He smiles. That’s better. Now I need to apply that everywhere.
- Increase the role of women. This was something I’d had in the back of my mind for some time, particularly with a view to Saint Justice. Every book after that features Sally Rogers, who is a strong female character. Later on you’ve got Hellion and Clara Baxter too. But in book 1, two women who were featured, Cheryl and Sinclair, were presented primarily in terms of their physical attributes. Cheryl is a burlesque dancer and Sinclair tries to interrogate Wren using her sexuality. Not great to do this twice with the only two named female characters. So I reduced the description, gave Cheryl more to do (an essential role!), and flipped one major character from male to female – Dr. Grayson Ferat became Dr. Greylah Ferat. It’s an easy change, not impactful in any way, but just gives women more of an important/authoritative role.
- Reduced overall length. Generally, all the changes above meant cutting length. Often these were whole chapters, or repetitive sections. The books were all 80k+ words, now they’re mostly all around 70k. They should read way faster and cleaner for this.
- Add back in some alternate perspectives. I removed all these, but on reflection decided to add some back in. A book in this genre generally needs to open with a crime. If I follow that ‘rule’, while also following the rule about Wren as the only POV, then every book has to open with Wren in dire threat. But that’s exhausting and hard. Probably it’s unbelievable. Threat to Wren, after a few books, is not too scary. We know he’ll be OK. So threaten someone else. Maybe they’ll live. Maybe they’ll die. Either way, it’ll be a crime to kick off the story. For example, War of Choice opened with a flashforward of Wren in Ukraine battling Russians – because I had nothing else from his perspective. But it’s not good – it only works by making us curious about what he was doing there, which is not as strong an emotion as outrage at a crime. A way better opening is to have the POV of Wren’s hackers, and watch them get abducted. We like the hackers, but we don’t know if they’ll survive. So now the whole book is taut with the question of whether they’ll survive. Likewise, book 6 Enemy of the People used to open with Wren escaping Washington – a direct continuation of the cliffhanger in book 5. Now it doesn’t, now it opens with the threat specific to that book, as a woman burns herself alive.
- Remove/reduce colorful metaphors. I’ve always enjoyed language, and love to have fun, bright, colorful imagery. One time, Wren hit a compound so hard it crumbled like a wet Graham cracker. Another time a compound split open like an edamame bean. To my eye, those are fun metaphors. but to some readers they were ridiculous. They took them out of the moment. To some extent, it’s me, the author, being clever-clever with an offbeat comparison, when I should just say something straight and clear. The compound crumbled. The compound split. I changed these.
- Cool the overwriting. I’ve done this before with this series, grading the ‘intensity’ of my writing so it’s not always got to be absolutely amazing and every fight doesn’t have to be the hugest, coolest, most violent. I see this with a lot of ‘massive uppercuts’ and detailed descriptions of how much force a punch would deliver. I remember thinking these were very cool when I wrote them. I probably still write about ‘massive’ uppercuts now. But I imagine a lot of them is exhausting to read. It’s kind of telling not showing. I’m ‘telling’ the audience, wow, be impressed by this punch, rather than just letting them be impressed by the punch.
Those are some of the biggest alterations I’ve made. I’ve probably done lots of other things too, like feeding Wren more often and letting him rest more, fleshing out main bad guys like the Ghost and the Alpha more, reducing the epilogue-level narrative summaries of what happened next, and so on. Maybe I’ll write about those later. †
In the next day or two I’ll finish this rework of Saint Justice and all 9 books should be more in line with each other, and hopefully with reader expectation. Just one question remains.
Will the readthrough increase?
If it does, that’ll be hugely gratifying. I learned and I got better. It’ll also give me something to apply as I go forward and write later books – as well as more motivation to do so. If I’m carrying 50% of all readers through to books 10, 11, 12 etc, I’ll be even more keen to write them.
If it doesn’t, then I’ll keep looking for the reason. New reviews may offer some new solutions. But I won’t do that exclusively – it’ll take time to reflect on this and think about what I’m missing. I’ll get on with new books, never fear.
We shall see. Watch this space for a report back.