Follow 100 historical novelists on Twitter. Set a timer. Wait for one of them to post along the lines of “I spent an entire morning in a research rabbit hole that ended up giving me one lousy sentence in the MS, and I just cut it.” Check the timer. Less than 12 hours? Thought so.

A friend who’s writing a historical was wailing to me the other day:

My characters have gone to a new location and now I have to stop everything and research it, and I don’t even know how long they’ll stay there until I write it, but I can’t write it till I do the research! How do I know what I need?!

As it happens, I have just hit The End on the first in my forthcoming Doomsday Books duo (working title, books to come 2022) and I was going through my piles of research books, checking all those post-its and scribbled notes, and remarking sourly just how many of them I didn’t use. It’s a thing.

Say you decide your MCs are going on a long journey. You have two basic options.

1. Research it. Spend days digging into means of transport, travel times, and the relevant ports or stations, squinting at bad PDFs of timetables, googling “how fast does a horse and carriage go”, and otherwise plotting the exact route.
2. Fudge it. “Two days later they were in Berlin.” [XX check time later]

(Note: XX is a great way to annotate your unchecked details for future search as you write, so you don’t break your flow. It also has the great advantage of leaving all the boring hours with an etymology dictionary and an atlas to Future You. The disadvantage is that eventually Future You becomes you.)

You may need option 1. My book Subtle Blood has a long sequence set on a 1920s steam yacht. I spent hours finding a diagram of an appropriate yacht, pictures, looking up accounts of trips, badgering my endlessly patient sailor brother-in-law to make it make sense to me (sorry again, JP), and generally making the steam yacht a real solid moving thing in my head. In the finished book there’s not very much of this on page at all: enough for the reader to understand the action, but far, far less detail (like, 90% less) than I ended up knowing.

This was not wasted research because it didn’t end up on the top of the page. I say top because it was there, just not visible. Research informs your writing, and it shows through the text like a backlight. Readers can tell when you have a good sense of what you’re talking about. They don’t want the full facts about a steam yacht or rail routes from London to Berlin in 1920 themselves, but they need to feel like the book’s not winging it.

Or you may need option 2. If you need your MCs to be in (rather than get to) Berlin, put them in Berlin. Paragraphs of detail about how they get there are not going to grab anyone at all: the only reason we care is if something interesting happens on the journey.

Research it or fudge it. So far, so obvious. However, because writing books is always more complicated than that, we need to throw something else into the mix. Follow some more authors on Twitter and set a timer for this one:

I dropped in a casual reference to an obscure fact for no reason in book 1 and it’s become the cornerstone of the whole trilogy! Thank you subconscious! #WritingGods #Blessed

Then unfollow that #Person immediately, but you get my gist.

Every author has a ‘random detail changes everything’ story. I’ve just had one myself. My hero in Doomsday Book 1 has just moved to Romney Marsh, an isolated and sparsely populated area. I had an entire job for him in the synopsis that I had to jettison because let’s not talk about my inability to stick to a synopsis, so I was casting around for something for him to do all day. I decided he was an amateur naturalist because I’d read something that reminded me that was something gentlemen did in 1810, and frankly, I’ve already done heroes who are artists and classicists and merchants and I couldn’t think of anything else.

The naturalist thing is now not just a detail. It’s a key element in developing his character and his relationships with two other people, it’s specifically plot-crucial in three separate ways, and it will be a nifty moment in book 2. I cannot overstate how much this decision unlocked for me.

I didn’t plan any of that ahead of time. I used it because it was there. It was there because I put it there. Why did I put it there? No idea.

And this is where starting with “Two days later they were in Berlin” falls down. Because maybe if you actually looked into getting there, it might turn out there was a night train perfect for sex, espionage, murder, or all of the above. Maybe there’s an amazing place they go through with an old town square or church or mountain range begging for an action sequence, or a secret meeting, or a bandit attack. Maybe there was absolutely no way to get to Berlin in two days, it’s a minimum of five, and now you’ve borked the timetable of another plot strand, you idiot. You’ll never know if you don’t look.

Which sounds great. But let us just refer back to the first tweet, and the hours of research that went absolutely nowhere or led to irrelevant detail that got binned in the second draft…

I was going through the MS the other day, and I came across a single line that needed filling in. Let’s say it was identifying a minor character.

John Bloggs was the Earl of Blankshire’s brother. XX CHECK LATER

When I wrote this I didn’t know or care how he fits into the Bloggs family. For book 1 it doesn’t matter.

However, book 2 is all about the sprawling, weird, Gothic, and possibly homicidal Bloggs family. I know this much, but have I done the family tree and synopsis? Have I hell. I am not ready to identify John, with his particular knowledge, presence at a certain crucial book 1 scene, age, and personal characteristics, as the old Earl’s brother and thus uncle to the new Earl, our hero. I might very well need him to have a very different position in the family–married in, say, or not in the line of inheritance at all.

And the stakes are high with linked books. If you’ve written a trilogy, you know the pain of that one damn line in the published book 1 that’s completely screwing the thing you now want to do in book 3. Because a detail might open the whole book up for you (my naturalist) or it might close it down (NOO I said he was the Earl’s brother, I’ve ruined everything!) This is why they say the devil is in the detail.

In this case, I can leave him as the brother, and have that as a fixed point, which might well act as a spur for me to develop the plot. I can sit down and work out the family tree and the synopsis of book 2 now (but see above for me and synopses). Or I can fudge it:

John Bloggs was one of the more eccentric members of the Earl’s highly eccentric family. (Two days later, he was in Berlin…)

Going for the fudge is the right thing in this instance, probably. I’ve chewed it over and I can’t see any book-enhancing reason to specify the relationship at this point. And the fudge will cover, I think, more or less any choice I make. I think. We’ll come back to this post when this bites me in the arse.

There’s an old saying that 90% of advertising spending is wasted, but nobody knows which 90%. You could say much the same of research. I wish I could tell you how to distinguish between the throwaway detail that will become the solution to all your plot woes, the throwaway detail that helps anchor the book in reality, and the throwaway detail you throw away. Sadly, I can’t. You’ll find out when you write it.