Ten years ago, I won the Black Orchid Novella Award, sponsored by the Wolfe Pack, AKA the Rex Stout Appreciation Society. Stout, who passed away in 1975, was a master of the novella and often produced a combination of novellas and short stories to fill out a Nero Wolfe book. The form is rare now, partly because it's too long for most magazines and too short to publish as a stand-alone book. There are few markets for them. Black Cat Mystery Magazine will look at a 15K-word MS, but reluctantly. The few other markets I know skew very literary.
Alfred Hitchcock Mystery Magazine co-sponsors the Black Orchid Award (Nero Wolfe supposedly raised orchids, a trait he picked up from his creator) and publishes the winning entry every year. The contest rules define a novella as between 15 and 20 thousand words. Other sources give different counts, but the point is that it's enough longer than a short story to need more meat or the bones will show through.
Then I saw a post about the Black Orchid Novella Award. Could I expand that short story and introduce those many characters more gradually?
Over the next three days (That's not a typo), I added 9000 words. I added one short transition scene, but nothing felt like padding. I sent it out and guess what? I'd written a novella that needed four years for me to recognize it.
Several years later, I won the contest again with only the second novella I've ever written. That novella had the opposite problem, though. About two years after "Stranglehold," I wanted to use the same characters in a novel, but it wasn't going anywhere.
My novels usually have two or three subplots that are variations on the main theme, and here everything except one minor variation felt forced and artificial. I struggled off and on for several years, then decided to lean on that subplot and try to cut the mess down to another novella. "Look What They've Done to my Song, Ma" won in 2016.
With that wealth of experience, I think I know how a novella works now. That's probably the kiss of death, isn't it?
Don't think of a novella as either a short story or a novel. Treat it as a distinct little creature. My ideal short story uses four or five named characters and no more than the same number of scenes, preferably in few, maybe even ONE, location. Novels are at least fifty scenes with more people or places, and several subplots.
A novella has one subplot and more scenes, a few of which might even be backstory, and more characters than a short story. Without going back to actually count, I'm going to guess that both the novellas above have about a dozen scenes and about the same number of characters. I try to keep the cast as small as possible, but let myself write big and messy because it's easy to cut scenes later. It's also easy to spot characters who serve the same function and combine two or three of them...if you even need them at all.
My current WIP, an early plan for another novella, has one subplot and a cast of 12. I'll probably eliminate some of those characters, either by cutting them or killing them, but I don't know which yet because we're still in the first date stage. I never kill someone until the second date.
That's another difference. When I begin outlining a novel, I think I know the ending (Sometimes that changes) and my main worry is how the PI will figure it out. I discover that by writing the scenes, and I often go back to change or add something so it all works at the end.
When I write a short story, I usually know the conflict, gut the rest of the story grows and develops while I write and rewrite as I go along. More often than not, the "real" ending shows up on the third or fourth draft.
I knew the ending of "Stranglehold" because it was a finished short story. According to my spread sheet, it was only the seventh short story I submitted anywhere, and I first sent it out in January, 2005, only about 18 months after I returned to writing after a long hiatus. Four years later, I expanded it into the novella.
"Song" didn't exist except as several pages of incoherent notes and a partial outline that made no effing sense. When I finally figured out the main plot, the subplot grew out of the characters and I pounded out a first draft in a week or so. I had a general idea of the ending, but didn't know how Woody Guthrie would solve the mess until I actually wrote that scene for the first time. It was like driving down a dark road at night and seeing a hitch hiker appear in your headlights.
That seems to happen to me more and more now. My WIP doesn't even have headlights yet. I don't even see the double line down the middle of the pavement. I have a general idea and I think I know the characters, but I don't quite know where I'm going. It's more interesting than worrisome.
I now allow myself to write quickly and worry about nothing except getting words on paper. A few years ago, I couldn't have worked this way, but now I know that if I write absolute junk on Monday, by Tuesday, something better will show up. Maybe I'll figure it out during the night or on a cardio machine at the health club, but something better will appear.
The way to solve a writing problem is by writing. You can fix anything you can put on paper. You can't do anything until then. Well, maybe if you're Mozart...
I'm beginning to look at novellas and short stories more closely because I've written myself into a dead end in both my series. That perception may change, but my mind is beginning to work in smaller units now. I suspect that in the next year or so I will move to publishing more short stories in digital formats, and a novella or two would flesh out collections. Rex Stout did it, and maybe what's old is new again.