Some thoughts here, far from being a theory. I may also just repeat a few notions you guys have already thrown in, sry bout that.
1. "The minigame sucks" problem: the issue I've encountered most often as a player is that some DMs do not try to make the wilderness *adventure* interesting, but *travel* itself. Instead of creating adventures, they create procedures that would, through simulation, "generate" adventure. It sounds like a great plan, but most of the time they've failed, as the result boiled down to all the hardships of travel (losing food, water, getting tired & fcked by all kinds natural disasters and bad weather), lots of semirealistic and tiresome details, and obligatory no-decisions (
yes, we are trying to find fresh water and food each day and
yes, we do stand guard during the night). Which is boring in my opinion. I mean, resource management is important, and it adds to the fun, but it is not the fun itself.
At best they use random encounters, which, when done creatively, can really make memorable adventures. But that might be another topic.
2. Lack of real decisions: another issue is that on large-scale hexmaps usually there are not enough decision points to make travelling really exciting in itself, and those we have get tired easily. Do I want to leave the safety of the road and cross that mountain or forest hex, just to reach the city/dungeon a few days sooner? It is a valid choice, it's just not really an exciting one. Now, if I have some more information about what dangers and rewards await me in those mountains or forests, the decision gets more interesting. Names can also help. "Green forest" means boredom. It is not appealing at all. "Forest of the Iron Wights" is something i might want to see just to find out whatever the f* is an "iron wight".
Also, if it is not a "hexcrawl for hexcrawling" campaign, the party is usually travelling somewhere. They have a target. If they really want to get there, they won't stray. So, if you want to make a "wilderness adventure" out of it, your options as a DM are mostly limited to roadblocks and train stops. The party has to stumble upon something really interesting/motivating to abandon their goal. I think it is easiest achieved by getting them really curious about something, or hinting at great rewards at the cost of great risks. (I might be wrong of course.)
I mean:
- A caravan massacred by something that left goblin footprints are... most probably goblins -> whatever guys, i've had enough of them gobbos, let's go on
- A caravan attacked by intelligent velociraptors that used sleep gas grenades, left everyone alive and only abducted
one of the passangers... you know you'd be curious wtf is going on.
Disclaimer: pure hexcrawl campaigns never really worked for me. I had to motivate players to travel. (Just as pure dungeon crawls never worked for me. My players like it, but I have to motivate them to explore those dungeons.) So there may be something in hexcrawl procedures I don't get.
3. "Exploration is not fun. Discovery is fun." <-- exactly this. Just as Yora said. Most of the time what makes wilderness adventures exciting is what the party discovers. It is also the easiest to plan ahead. It should be surprising or at least non-trivial. It should be possible (and rewarding in some way) to interact with. Best if it is something open-ended that supports multiple approaches. Discoveries with the sole purpose of creating atmosphere, foreshadowing or hinting at something are of course okay, just not as exciting as actual interaction.
"
Hexcrawls seem to be mostly empty maps with perhaps every tenth hex you move to having something in it." <-- and also this. Maybe it's just that I've read and ran the wrong hexcrawls, but most of the time what the party discovers is a standalone encounter or mini-location (maybe, but not necessarily because distances between locales are too big to be linked somehow). Which is shallow, even if the discoveries are interesting.
Examples:
- Hex 1234: a pair of griffons with golden feathers guarding a golden egg. 50% that one of them is away.
- Hex 2345: a tribe of 137 goblins worshipping Growlog petrified "statue" of a laughing hill giant (Tasha's hideous laughter + flesh to stone hit the dude the same time). The goblins all laugh like maniacs in a fight.
- Hex 3456: tower of the wizard Clorox, alchemist and archmaster of transmutation magic. Needs troll poop in great quantities for his latest experiments.
What I miss from these kind of adventure locales/encounters is a short description or hint of how they relate to each other. Like:
- The golden feathered griffons are Clorox's creations. He loves them and would be furious to see them slain.
- Clorox petrified the giant. Thinking of the foolish goblins idolizing his victim humors him greatly.
- The goblin shaman stated that Growlog demands sacrifice - the golden egg of the griffons. The goblins have tried but could not get it. The party would earn their respect if they presented the egg to the tribe.
These kind of relations help the DM create complicatons and consequences. Depending on the approach of the party, it could lead to multiple outcomes. The party will enjoy that they have agency, reason to interact, leads to follow and consequences to their actions, and it will also be exciting for the DM to see where the adventure goes.
Now, if someone really wants to add complexity to their wilderness locales, they'll usually add
- a dungeon(ish locale), which will make that session more like a dungeon adventure
- a settlement or habitat, which will make the session more like a social adventure
Which is totally okay.
Required, even. It's just that they are not really "wilderness adventures" anymore. If you want to stay aboveground and outside city walls,
and create all kinds of intertwining relations between nearby places and creatures you'll probably get down to smaller scale (like a pointcrawl). And with that, the adventure will be structured much like a dungeon.
TL;DR version of what
I think is important when designing a wilderness adventure:
- Do not waste too much time on resource management, especially on routine tasks, it's boring.
- Give a reason for the players to explore. Not to go somewhere, but to search for something. Or create the need to visit multiple locations in whatever order they wish to go. Create relations between the places and creatures living in the area. Give the creatures some habits that the players can learn and use to their advantage. Create conflicts and factions. Make discoveries non-trivial and interesting to fiddle with. Basically, do what you would do with a dungeon.
- Give the players interesting decision points, and provide information to help them decide. Throw in gossip. Make some stuff visible or audible from a distance, like (just as Two Orcs proposed) landmarks. Something twinkling on the peak of that mountain, a really unusual tower in the distance, red smoke rising from the forest, the sound of drums or screams in the distance, or simply a signpost telling travellers no to go right (especially if you want them to go right

).
Like I said, I find the latter two a
bit harder on large scale hexcrawls, but it sounds nowhere near impossible.
...annnd that is my 2 cents.
