Doctrine might be a bit strong.
Still, I'm trying to keep these three words in my head whenever I'm planning or running a game. They're the three-beat tempo of a game, even if you aren't thinking about it. Slacking on any one of them can result in a worse game. Get them all spot on and things will feel just right.
They're everything that's great about tabletop RPGs. Everything that sets them apart from video games, board games, novels, can be found in this, the spine of gameplay.
It all comes back to Player Agency, but I find it useful to break things down into threes.
In RPGs, questions are gameplay.
I guess this is the hill I want to die on. I've written about it in relation to traps before, but it's applicable to the rest of the game too. *clears throat*
Players cannot make a proper choice unless they have enough information!
Knowledge and Perception Rolls are the worst offenders of not understanding the importance of Information. When I see them in use I just wonder what could be lost by just giving the players the information?
I want players to imagine the situation their characters are in and think of a clever solution. Asking for more information should be rewarded! If they ask smart questions I give them great answers.
Whatever you're planning, think in advance about how you're going to present it to the players, and how you're going to give them enough information to make a proper choice.
No easy decisions.
This one is the most difficult to just insert directly. For there to be a proper Choice, there have to be multiple actions the players can choose between, and deciding between them shouldn't be easy.
This is really the glue between Information and Impact. Get those two right and this will often fall into place, but you still need to make sure your world supports multiple solutions to problems.
Look at your prep and try to identify the decision points the players will come across. Left or right at this junction? How do we get past this broken guard robot? How can we trust these shady mercenaries? If any of them have one really obvious solution then you need to make the situation more interesting!
One always-hostile orc guarding a chest isn't a decision point. You kill the orc and take the Treasure. Give the Orc a death ray, make him sympathetic just trying to hold down a job, or give him an alarm he can pull to bring the whole army down on you. Now you've got a Choice.
Everything you do matters.
I used to call this section Consequences but it felt negative-leaning, and Impact fits so well after reading Arnold's post on the topic.
Essentially, when your players have made a choice, things should happen so that they look back and think "huh, we did that!".
Maybe they regret it, maybe they're proud of themselves, maybe they just wonder how things could have been different.
My vice is making players feel guilty for killing some innocent monster or screwing over an NPC, so I ham it up real good. They always remember that, because it's over-the-top emotional silliness, but they know that they did it. I've seen players feel more strongly about killing an NPC than losing their own character to death.
This is the payoff for everything before this point, and without it your game is going to feel flat.