As much as I love lingering over collaborative setting creation (throwing Fan Mail at Questlandia and Microscope), I don't think I can afford any of that here.

With the goal of introducing complete newbies to the idea of RPGs in no more than 30 minutes, I think it's probably best to design a game with a clear and exciting situation baked right in. The players need to know where they are, what they're facing,  and what they can do about it immediately.

That's pretty easy to do with a physical conflict. Pick a familiar genre, threaten the players, go. Whether it's an Inn surrounded by zombies or a sky-ship being boarded by pirates, it works the same. The stakes are high and the bad guy is obvious.

I struggle more, always, thinking of similarly engaging non-combat scenarios.

What's your approach to creating a scenario for your game? What advice do you have for scenario design?
Or maybe you disagree with my entire premise and are taking your game in a very different direction?

I guess it depends on what part of the game you're trying to introduce to the player - combat, role playing, character creation or all of it?  It's the fun part of this project - thinking about the core of the experience.  

I'm worried less about the scenario.  The experienced GM can guide or decide.  The tools to quickly get to what's fun about playing is my focus.

So far my approach has been to have a specific scenario theme, then provide a list of questions to place that scenario into a setting. I have some pre-baked scenarios in there as well, but I may have to cut them for space or drop the questions and add "or create your own" as an option.

I feel like when a player understands the genre of story you're trying to use it's going to be an enormous boost in how approachable your game is. And I hope I don't need to say it, but genres where the solution is violence are going to leave a lot of people out in the cold. 

One way to solve the scenario problem in a very compact way is to move away from "traditional" fantasy (which needs to be explained, and defined, and might not be familiar to unPlayers) and instead embrace some kind of setting that can be instantly understood and accepted by pretty much anyone. This includes "structural setting"... as in... there is THIS SITUATION, what do you do about it? Go!

It's basically a different way to do the thing that Brand_R friend does when larping ;)

Nothing has been explained in the traditional sense, and players might be imagining wildly different things, but as play moves forward and details are added here and there, what's important will emerge and crystallise. What you need is an effective way to handle the game conversation. Here I prefer gmLess solutions because they are tighter, more reliable and more challenging to design, but a decent set of instructions for a gmLed approach would serve the same purpose. No need to chain the whole game experience around physical combat, which to me is possibly the most uninteresting element of any game :P

Most veteran players (and designers) seem to seriously underestimate the ability of people to imagine stuff, given the opportunity to do so. It might not go in the direction one expected, but it might also still work amazingly and be a lot of fun. There is quite a lot to say about game that don't expect players to follow a pre-determined imaginary. Leveraging this quality in people allows designers to spare a TON of text that would otherwise need to be devoted to lore, setting, scenario definition, etc.

My solution for this jam was to provide a near-normal setting : "It's the world we live in, but there are monsters now and they can get you." I put the pressure on players to exercise their agency by removing the option to refer problems to authority: "The police and National Guard are busy with big stuff." And then to  present a problem that they had to act to solve. "You've got a sick friend who needs medical attention." So it's absolutely a limited scenario, but there are a lot of details, left for the players to author in play - the nature and details of the monsters, the relationship to the sick friend, etc..

My goal was to leverage the familiarity of the real world and tighten the screws with familiar fraught situation stuff that they've seen in movies, on TV, or in books a million times - again, leveraging familiarity - to stay within the 10-30 minute time frame while still offering some narrative authority to the non-GM players.