Do you like the way I said "what could still be improved", rather than "what went badly" as if doing creative things is an ongoing journey rather than a pass/fail test? I hope so because it's hard work making my brain do that, but I think the rewards are worth it.
I had to force myself to sit down and go through all the major pieces of prep I did and ask "would the session have gone as well without it" to get my list of "what went well" because my first instinct was to assume that all the things the players did awesomely would have happened anyway, even though when I looked closer, a lot of them wouldn't.
Invent not "Who Are They" But "What are They Trying to Do"
In roleplaying -- or linear narrative fiction like books, tv, computer games -- any interaction is more interesting if the participants are actually interacting. If the PCs have a clear goal for the scene which could succeed or fail. And the NPCs are not just passive, but are pushing in a direction -- be it "make the guests happy" or "make friends" or "do a street performance (without being interrupted)" or "don't let anyone past the bridge".
But I just verbalised that dichotomy now, despite reading a lot of similar advice (goals, "what's my motivation for this scene", every scene should be about resolving a conflict of potentially thwarted success, etc)
So in the past a lot of my worldbuilding was too static -- a status quo of "this person/animal/society usually does this" instead of "is currently trying to do that", like a painting instead of a "in media res". I will try to do the opposite!
And further, not just minor conflict, but I always empathise too much with NPCs, I need NPCs who want something unreasonable and aren't willing to compromise, and even NPCs who are just antagonistic, so there is significant conflict for the players to overcome!
FWIW, I think some mediums make the opposite mistake, e.g. art, fictional encyclopaedias, exploration games are much better suited to showing a snapshot than an unfolding narrative. The same applies to larger bodies of work: The Robin Hood or Arthurian legends, or the stories about a pantheon of gods usually paint a picture of what the characters are like much more than presenting a beginning-to-end story. I think Magic:The Gathering is much better suited to showing what a world IS like than by showing some great transition, and I wish they would take an approach more like greek myths and less like the MCU.
Have clear goal
Related to the above, I tried to have a clear goal "a demon has escaped from the spirit forest, find it and fix it". I even explicitly asked people to think about a few ways that scenario could end. But I think it wasn't immediate enough, and didn't build on people's existing awareness enough, so it felt very abstract and not like a clear goal.
I could have fast-forwarded to start the party in the ruin of a farmstead, with an immediate "help, help, stop it, it went that way" or similar.
Or I could have done a cut-away scene to show the spirit demon causing mayhem even if they didn't know that in character yet, so they had a strong motivation to stop it.
I tried to establish the important points by having the party encounter a lesser dangerous spirit immediately, to establish rules of "how to deal with dangerous spirits" and "what damage they could do" and that helped, but I don't think it did enough.
More minor scene-to-scene goals (e.g. convince X to let you take her boat, scale the cliffs at Y) would also give more stakes and opportunity for establishing trade-offs -- sometimes you fail that thing you wanted, without failing the whole mission, and that makes the whole thing more interesting.
Have meaningful action resolution
I hadn't realised I'd done this, but half the session was "find out about the mission, get in a boat, go there", which was great for getting people used to the setting and mechanics, but didn't have a lot of "Can I do X?" "OK, well, roll, and we'll see", simply because I tried to seed in obstacles to the campaign, but I didn't think of every interaction as one that might go either way, even when choosing an NPC's attitude differently might have turned the conversation from "she gives you a quest" to "you try to convince her you're up to it" or turned "you get in a boat" to "oh no, a character acted out one of their flaws and now the situation is harder, can you fix it?"
Timing
I knew with six people, two of whom are 9 and 12, it would be hard to fit things in, and I pared the plot down a lot to a simple "establish premise, dangerous encounter [with fish], some more role-playing to establish characters, climax confrontation", but even so, people were losing concentration after a couple of hours. So we had a good session, hopefully memorable (especially the bees and the fighting the fish), and people got used to the characters and mechanics, but I feel like I could have done better to make two hours thrilling from the start.
Cookie economy
Because there just wasn't enough difficult resolution, people had few opportunities to spend cookies, so they loved earning them, but they didn't matter often enough so there was no real chance of running out, or a sense of how close they were to losing a conflict over something. Partly, I need more opportunities to make actions that matter, maybe I need to reduce the number of cookies.
Minor bits of prep
There were lots of minor things that would have helped. I planned to use physical counters for cookies but that was a bit risky with Ms Under One's inquisitive hands around, but I think they were much less resonant when they weren't being added and spent all the time. I wished I'd had a chance to prep my helpers a bit more specifically with like, this bit could be written small, this bit could be big so everyone can see, that are automatic if you're used to running games, but you don't necessarily know if you don't.
There were a few practical inconveniences like, how do I print out a few copies of the quick start rules and make it obvious at a glance which bits of paper are duplicates and which people should try to look at both of.
