A few days ago, as a friend and I were chatting about The Pauline Conversion, he said he’d bet any money I’d researched it to within an inch of its life. He believed I wouldn’t be happy until I’d followed every research thread to its end, mapping every paving slab and shop front of Cambridge in February 1972 to be sure the story played out in a historically accurate setting.

I laughed. He’s not right, but neither can I say he’s entirely wrong. The truth is, research is the best way to keep my flat clean. When I need to research something – say, the miners’ strike in progress at the time of the story – I approach it as I would a rattlesnake doing its little dance. I run away screaming.

It’s not that I don’t want to know what happened, or don’t want to do the research. I enjoy knowing. I enjoy learning more about things that interest me. The problem is, I’m a details person and I can be a bit of an obsessive. Research has an inertia for me: hard to start, hard to stop.

And I know I’m like this: hence the procrastination.

Research lays a number of traps for the unwary.

First: I researched it and you’re damn well going to know it. You know the sort of book. The author vomits every factoid in his research notebook into the manuscript to justify the time spent discovering it. The precise shade of blusher under a talcum face and wig. The twirl of brass on a pistol. The manufacturing history of a pair of contemporary bloomers. The author’s trying to paint a picture with too many megapixels. I don’t think I suffer from this: I try to make details serve the story as much as possible, or add a smattering of colour, or at least be an excuse for a joke.

Similarly, the past is a foreign country, endlessly fascinating, and it’s easy to dwell upon the differences – “only 7 1/2p for a bag of chips!” – and the multitude of hats, and forget that to the characters this is the now, the normal. Although a book might be written almost as the memoir of a participant, it’s best to avoid the literary equivalent of pointing and sniggering. There are always exceptions – for stories that flit back and forth in time, for example. But comparing then to now brings the certainty that, before long, now becomes then and the book exists in two time periods: when it was set and when it was written.

To put it another way: all books, even novels set in the present day, are historical. (Some day Disunited will be a curiosity in a world where out gay professional male footballers are normal and common. I hope it happens soon.)

Another risk is the possibility the research becomes more interesting than the book. I was worried I’d be knocking on the library doors at 7am desperate for my fix of 1972 local newspapers, fast becoming the world’s foremost authority on postwar Sidney Street roadworks. An honourable title, to be sure, but unlikely to power a page-turner.

And there’s the risk that sometimes becomes a gift: the wondrous, terrifying prospect of the author finding a piece of information that upends the entire story. Is it crisis or opportunity? That’s part of the fun of research. And by fun, I mean nightmare. The author never knows if, lurking over the page, is a photo that burns the plot to vapour. This worry can spur research far beyond what’s required to write the book.

Related to those last two: a discovery that real life events make a better story than the one the author had in mind. Truth really can be stranger than fiction – coincidences abound that would be laughed out of a novel. (Good fiction can get away with one coincidence at the start, to set the plot rolling, and that’s it.)

A pragmatic approach is best.

With The Pauline Conversion the vast majority of readers, most likely, didn’t experience February 1972 in Cambridge, or perhaps anywhere south of an ovary. But I had to assume a reasonable level of knowledge of that time, gleaned from popular culture and so on. Nationally and globally, I wanted to be as accurate as I could. I’d never have dropped a smartphone into 1972, but not everything is so clear-cut. TV remote controls? In the US, yes. They were rare to non-existent in the UK. Our house didn’t have a TV with a remote control until the late 1970s, and we were probably early adopters. So, no remote controls.

But did it matter that characters in a scene didn’t mention the rather large Tesco along the street at that time, even though it would’ve altered their behaviour? Absolutely not, since the store had no relevance to the story. (It would’ve mattered had characters later visited it for some reason: consistency is vital. As it happens I learned about this Tesco shortly before publication.) Local contemporary norms can be subverted, I suppose I’m saying.

Partially this is to allow the story to happen. St Paul’s is a fictional college superimposed upon a real geographical area, after all. And partially it’s because local knowledge can be tremendously hard to get right: the information just isn’t readily available at that level, if it exists at all. There are always people who know more than the author about the topic – local geography, terminology, technology, etc. That’s a fact of life. But a writer can’t read everything, can’t interview everyone, otherwise we’re back to mapping paving slabs. Immersing is great: drowning is bad.

Sometimes I’ll get things wrong by accident, and sometimes it’s better to fudge things to simplify matters. My goal is “factually correct” (aside from the obvious fiction of the story), but I’ll accept “not obviously wrong to the layman”.

Perfect is the enemy of good. Serving the story is the key. The author needs just enough research to get from beginning to end. Breadth, not depth.

For the mildly obsessive like me, that can feel like cheating. Like setting foot in a foreign country for the first time claiming to know the language, when you’ve no more than skimmed the “Useful phrases” section in the Lonely Planet guide and can order up to, but no more than, six beers. You want to know only enough to get by, to bluff your way through, because beyond a certain level of knowledge the time spent in research feels wasted.

I say “feels” because I do need to know more than appears in the finished book. A tourist in a foreign land who takes the effort to learn just that little bit more of the language gains a buffer of confidence (“seven beers” is so much more impressive than “six beers and one beer”, if an additional drinker appears unexpectedly). A similar buffer in my research lets me add nuance, a little depth, as long as I don’t start vomiting out my notebook.

There is a balance, somewhere, which I hope to find as I grope around in the dark.

My friend knows me well, of course: it’s true I learned a lot more about Cambridge in February 1972 than I actually needed to know. To return to his original point, I don’t think I researched The Pauline Conversion to within an inch of its life: but maybe to within a yard. About the length of a paving slab.