I know what you’re thinking. Ideas for interpersonal drama, that’s what you get from advice columns.

Actually, not really. I am an avid consumer of various advice columns and I don’t think I’ve ever been inspired to springboard off the problems therein. I guess because advice column issues are rarely romantic, more the opposite, and they’re also already…spent. Once a story is told I rarely have much interest in it from the creative side. It’s why I don’t over-plan my books, I need it to be a surprise sometimes!

No, what we’re talking about is the craft of writing an agony aunt letter. It’s not easy to craft a believable narrative, including antagonists, twists, and backstory, within the limitations of that format. The reader needs to be hooked from NOT the first paragraph, but from the first line.

All of the stories are Absolutely True, of course. That’s the social compact that lets you enjoy a good problem page. Otherwise it’s just too much work and infighting. Plus, in a brief serious aside, there’s a chance that even the most bizarre letter is actually written by someone in genuine need. It costs nothing to take these letters at face value, and could hurt someone if you didn’t. So why not?

However, even accepting that every story is true, non-fiction needs storycraft too. And like short stories you can see the architecture of a good story much clearer in something as short and pithy as an agony aunt letter.

The hook is the first part of the letter. It’s arguably the most important, since if the hook fails there’s a good chance people won’t read on. In my experience the best hooks are actually simply constructed. The ones that try too hard are usually the ones that get your hackles up and stop you committing to the story.

“I murdered three people yesterday. know this sounds bad, but if you hear me out I think you’ll be on my side!”  (Will I though?)

“Mother ate a whole donkey. Should we give the saddle back?” (Clickbait!)

Readers need to get enough from the hook to be intrigued, but not so much information they decide where they stand on the letter before going any further. Never try and fool the reader into reading on. All you get from that is a pissed off reader. I still remember a book I got once that had the BEST prologue, just absolutely engrossing, and then it turned out to be about something else entirely. The amazing brilliant worldbuilding in those first three pages was just…background detail that never really came up again. The story itself was actually good, but I couldn’t get over my disappointment.

The next part is the world-building. The letter needs something to stand on, but not a mountain. Some people spend too much time filling in every background detail and introducing a huge cast of characters who aren’t actually important. There is a skill in paring the background down enough that it sets the scene, but doesn’t steal it. A good problem gives you just enough background detail that you can imagine the writer and a sketch of the background. Readers are here for what they were promised in the hook. Keep them waiting too long and they’ll wriggle away. You need to hit them with WHY you left you husband after only three days of marriage.

Which brings us to characterization. Even though you probably want to control what side of the problem people support, it has to be subtle. If you make one side TOO obviously right, or wrong, then the suspension of disbelief goes.

‘My husband lives in the attic and beds down in the insulation like a raccoon. He makes me serve his dinner in the bin. What should I do?” (It’s too obvious. I’m not torn about what to do or understanding about why you haven’t done it. You married a trash panda and you need to  face it.)

The character can’t be lacking any self-awareness about their plight. Especially, actually, if they are the rare antagonist letter writer. I, the reader, have to believe that this is a genuinely perplexing problem they can’t solve on their own. Not that anyone can see the right–indeed, only moral thing–to do.

Finally, we have the problem. It needs to be simple. The more baroque the problem gets, the more moving parts are required, the less faith I have in it. The more places you can stick the crowbar of disbelief and bring it all crashing to a stop.

There was a big ‘I lie about problems on reddit’ reveal a few weeks (? maybe, I have no sense of time anymore) where someone admitted they’d created a whole big lie about a family name for her child. She said she couldn’t believe people fell for it. Thing is, most didn’t. There was too much going on, she introduced wealthy grandparents and then weirdly evil in-laws out of nowhere (plus cribbed a plotline from a tv show. Such a mistake, there’s too many repeats for someone not to have seen that quite recently).

Decide what your plot is and keep it streamlined. If you find a flaw go back to correct it, don’t add more elements to explain it away. “So the Inland Revenue were on my back, this was when I was living in England with my Dad. He’s a diplomat. But his new wife hates me, and that’s why he couldn’t help.” If it’s a serial problem (they exist) you can be a bit looser, but not too much. Always remember it’s best to reweave the fabric of the story than patch the holes.

All of which holds true for writing short stories and novels too. True fact! Just like the punishment clowns and the Iranian Yoghurt man (perfect example, by the way. An elegant turn of the agony aunt letter pen).

Find me on twitter, facebook, and my substack newsletter!