There are lots of moving parts that have to come together in order to write a good mystery.
You need a sleuth, a victim, a murderer, and enough suspects to keep the reader from guessing whodunit too early in the book. Each of your suspects and, of course, the killer, must have a motive to kill the victim.
You need an interesting setting. Most people don’t consciously think about this, but could Robert B. Parker’s Spenser live anywhere other than Boston? Setting often becomes a character in itself. Certainly how your protagonist sees their setting tells you a lot about them. When the setting is different for every book in a series, like Nevada Barr’s Anna Pigeon mysteries, it’s even more important to focus on it. What is it about this particular place that makes it essential to this story?
You need a convincing murder weapon, one that was available to the murderer and that he or she would logically use.
You need clues, carefully planted, so that when the reader comes to the climax, they are not only surprised by who the killer is, but also have the reaction, “I should have guessed!” In other words, then ending has to be logical and proceed from the information earlier in the story.
If you’re writing part of a series, you have to remember to bring in whatever continuing subplots you have so that readers who are waiting for the answers to questions such as “Will they or won’t they?” or “Does this couple succeed in adopting a child?” or “Whatever happened to X’s sister?” are answered or at least addressed. And you have to bring in just enough of them to satisfy the reader without overwhelming the main plot.
You also have to have what I call the writer’s hook. Just as there’s a hook for the reader (usually stated in the book description) that makes them want to read the book, the writer needs something to make her want to write the book.
Since I do write in series, I already have my sleuth and a cast of characters from which to choose for the victim, killer, and suspects. But I’m not fond of killing my continuing characters, so I usually come up with a new character for the victim. Unfortunately, sometimes that means killing off a great character. I really loved Fox Fordyce in “Royal Purple Murder” and would have liked to keep her around. But then I would have had to write a different book. There are days when I regret that I didn’t do that.
Making one of the continuing characters the killer has the same problem. If I send them off to prison, they won’t be available for future books. So, again, I usually come up with a new character who won’t be missed by either me or my readers. It’s easier to consider a continuing character as a suspect, but you can’t always have them not be the killer, or regular readers will eliminate them right off the bat, spoiling the surprise.
Right now, I’m at the stage of developing the characters for a new book. My brain has had some ideas for new ones, but I’m not sure whether they’re victim, killer, or suspect. There’s a lot of exploration at this stage. I spent four hours yesterday researching something for one of those new characters. Fascinating stuff, but probably very little of it will make it into the book.
This morning I pulled some books on Arizona off my bookshelf for another character. I’m not sure whether they’ll help me or not. But I’ll spend a lot of time looking through them to see if a bit of information sparks a new aspect of this character.
I spend a lot of time on developing characters since it’s the characters who define the story, Most of the time when I’m writing, I sit back and watch the scene play out in my head. The characters do things that are natural for them to do, so it’s important for me to know them well. That doesn’t mean one hasn’t popped up in the middle of a story and demanded to control what happens next. This is both unsettling and delightful. Usually, these unexpected intrusions have a mind of their own, like Athena leaping from the head of Zeus full grown, and I don’t have to work at character development for them very much.
So far, in the African Violet Club Mysteries, the setting has been the village of Rainbow Ranch, Arizona. But having too many murders occur in a small town, often referred to as Cabot Cove Syndrome, becomes unbelievable over time. So should I have Lilliana go somewhere else once in a while? I’ve been thinking about that and even have a few places in mind.
As an example of a weapon, the softball bat in "True Blue Murder" fit perfectly with Lilliana Wentworth, my senior sleuth, as the killer. She played softball regularly and the bat belonged to her. It was left in a storage room, so it was also available to other suspects who had access to the building.
Right now, I’ve got a good murder weapon for the new book, but I’m not sure it’s capable of being wielded by all of the suspects. A puzzlement.
Since I don’t even know who is killed, much less who killed him or her, the clues are on the back burner.
I’m juggling subplots in the background, weighing what could be the next development with each of them while not focusing on them. Again, this has to wait until I know more about the people in the book.
As far as the writer’s hook, that was easy for this book. In “Double Pink Murder,” I needed a disruption at City Hall when Lilliana went to visit the police chief. Out of a newspaper article I’d read not too long ago, I came up with the idea of a developer who wanted to make an old ghost town into a tourist attraction. Now, I love ghost towns. The whole idea fascinates me. There are many of them in Arizona where mining operations petered out, and the population abandoned them. I also love Old Tucson Studios, a western town that was used for the sets of a whole bunch of western movies. They do tours and reenactments and have a lot of memorabilia from those westerns. So rubbing those two ideas together got me really excited. I’m getting all sorts of jumping off points from that for the plot.
It’s early days for this story, but I love the discovery stage of writing a new novel. I thought you might find it interesting.