Friendship Simulator: a Procedure for Friends and Contacts

The same way equipment and gear help you get things done in the dungeon, friends and contacts help you get things done in the city.

What this is for

No rules are needed to legitimize PCs' relationships with NPCs, but procedural scaffolding cuts down on ad-hoc decisions and provides rationale for how both the players and GM may approach this dimension of the game. Like how I use reactions tables and x:6 chances in circumstances I want to avoid GM fiat, a procedure for camaraderie takes the fiat out of PC-NPC relationships.  

Not every friendship needs to follow this procedure. If the circumstances are such that it's obvious an NPC is friends with a PC, let it happen. 

The Friendship Simulator is a way to provide texture time spent in town between adventures. Downtime is inevitable in OSR games, but it's largely unexplained what you're supposed to do  other than research spells, heal, and wait for your stronghold to be built (and train I guess if you're playing AD&D). The common fix is writing a menu of downtime activities, but this procedure is an experiment in another approach: instead of giving players more activities, give them more tools. With an abundance of hammers they'll see nails everywhere. 

Antoine Marchalot

Meeting Contacts, Making Friends

Friends start as contacts. Contacts are acquired the same way hirelings are: hanging out at bars, wandering aimlessly, and greasing palms. Other activities, like carousing and insinuating yourself in local drama, may also net you contacts. 

A contact can be tapped for one favor or bit of information. Make a reaction roll to determine how they follow through. 

2-: Refuse to help and they drop contact 

3-5: Refuse to help; they're busy

6-8: They help but ask for time (1 week) or something in return (1d6x50 sp or its equivalent)

9-11: They help

12+: They help and a friendship is born 

Unless a contact becomes a friend, the relationship ends after they help you—whatever they owe you has been repaid. 

Instead of tapping a contact for a favor, you can pursue a friendship. This takes a week of downtime getting to know them. At the end of the week there's a 50% chance the friendship is reciprocated. This can be attempted up to three times, after which the contact ignores you because you're not getting the message. 

Because adventurers lead fringe lifestyles, they can only accommodate a number of friends equal to the amount of retainers they can recruit. There is no limit to how many contacts they may have. 

Contacts and friends can be organized in the following categories: 

1. Rich
2. Tough
3. Streetwise
4. Connected
5. Famous
6. Eccentric
7. Weird
8. True

If a player is looking for a specific sort of individual you can let them choose, but it's more fun to roll. NPCs may fit in two categories, but never more.

Rich

The extreme distance rich friends have from the tragedies of this fallen world make them enthusiastic, generous, and uninhibited. Occasionally they fall victim to bouts of melancholy or get embroiled in petty disputes, but by and large they are easygoing and carefree. Rich friends value old-fashioned courtesy and you swallowing your pride when they show you off at dinner parties.  

Lean on them to spot you money, get you into a fancy party, or lend you some credibility when dealing with the aristocracy. 

Tough

Tough friends are bruisers, risk-takers, and braggards. They are earnest, brave, and impulsive; they drink hard and love harder. Tough friends value mettle above all and have little respect for bellyaching and bullshitters. 

Lean on them for protection, intimidation, and shit-kicking. 

Streetwise

Streetwise friends are cooler than you. They know all the gossip. They'll tell you who's legit and who to avoid, where to buy and where to sell, and if there's ever a time to lay low they're the first to know. Streetwise friends are witty and have effortlessly good taste; they are inevitably prey to some vicious vice that lends them a sense of mystery and danger. They value the free flow of information (there's no fun in keeping secrets) and someone that can hang. They have a casual distaste for squares and a very serious hatred of narcs. 

Lean on them for rumors, sourcing fences and contraband, and underworld dealings. 

Connected

Connected friends have some ambiguous but important-sounding position in the local government, a guild, or some other enterprise. They make themselves indispensable by doing the work of three people, which they always complain about but actually love. Despite being chronically overworked, they always remember your birthday. They value low-drama collaborators and not having their time wasted. 

Lean on them for official permissions, insider intel, and cutting red tape.

Famous

Famous friends are outgoing and glamorous and always failing to keep a low-profile. They're some variety of performer or artist or folk hero known far and wide throughout the land. They are distinctly well-mannered and gregarious if a bit off in their own world. They're always the unwilling participant in some drama or other yet it's no secret they thrive off the attention. Famous friends like having their ego stoked but also need you to keep it real with them. They get turned off by sycophancy. Famous friends always have a nemesis, a rival or notable detractor, that they will need you to be fully against. Associating with their nemesis will undoubtably ruin your relationship if word gets back to the famous friend. 

Lean on them for influencing trends, impressing people, and drawing attention. 

Eccentric

Eccentric friends are kooky and curious and excited about everything. Any time you go over to their place it's packed to the ceiling with evidence of their newest obsession. You're always finding them absorbed in some questionable experiment or pinned beneath a collapsed apparatus. They are relentlessly optimistic, nonjudgmental, and eager to share their interests with others. They are used to being written off, but still value someone who takes them seriously (despite their offbeat approach to fashion and personal grooming). They can't be bothered by naysayers and pragmatists. 

Lean on them for odd jobs, knowledge on niche subject matter, and repairs. They might throw in a few "improvements" for free. 

Weird

Weird friends are pariahs, hermits, and entities who don't fit in normal society—the kind that get the torch-and-pitchfork treatment if they visit town. Unnerving to most, but not to you. Weird friends are in-tune with many worlds beyond our own and spend their time communing with beings unseen by normal men. You are likely their only "mundane" friend. They value people who know what trees call one another and can hear the songs whispered by the stars. They are curious about but generally don't care to be involved in the struggles of civilized life. 

Lean on them for credibility among unseen powers, mystic insights, and freaking people out. 

True

True friends are unremarkable save for their bone-deep admiration for you. They are provincial, stoic, kind, averse to change, and above all dependable. No songs will be sung of them but they act with virtue nonetheless, as if they knew no other way. They have perhaps a superficial vice that only lends to their chumminess. True friends expect nothing from you but your camaraderie; abuse and neglect only makes them wish you'll find your way back to who you were, you they know you to truly be. 

Lean on them for anything. They don't stand out in any way but will follow you without question. 

Loyalty

Each friend has a loyalty score, similar to hirelings. When you ask for them to stick their neck out for you, make a loyalty check on a 2d6. If it rolls below, they follow through and their loyalty reduces by 1. If it rolls above, the PC can either back off, in which case their loyalty is reduced by 1 and the friend doesn't follow through, or the PC can press them, which will get them to do the favor but their loyalty is reduced by the difference. True friends only reduce loyalty on a roll of double sixes. A friend whose loyalty drops to 1 won't do any favors, but the relationship is still salvageable. A friend with a loyalty less than 1 is no friend at all. 

A week of downtime spent fostering your relationship with a friend adds 1 to their loyalty score, up to a maximum of 10. Loyalty may only be raised above that by doing your friend a major solid or by going on a daring misadventure. 

Misadventures take 1d3 weeks and will leave you 1d10x100 sp poorer. There's a 1:6 chance your friend gets killed, injured, kidnapped, or imprisoned, which has circumstantial consequences but results in the relationship being put on indefinite hold. Otherwise, their loyalty increases up to baseline (determined by Charisma) +1d4. You can only go on one misadventure with a friend—they know better after the first time. 

Other things friends are good for

They'll pitch in for your funeral if you die. Each friend shells out 1d4x50 sp, save for rich friends who ostentatiously drop 1d4x500 sp and send flowers to each surviving PC. 

A friendship can become a romance. A friend who becomes a lover is True in addition to their other type for the sake of determining loyalty. Their heart breaks every time you leave for adventure. 

Using This 

This is essentially just a way to set up players getting to go "I know a guy..." You can probably do that with an abstracted just-in-time system or quantum NPC but that's no replacement for actually knowing a guy.

How friends and contacts are useful in city adventures is self-evident, but in circumstances where the town is just a place to crash between adventures then friends hold less immediate value. That being said, consider: Dealing with NPCs Should Be Expensive and Irritating. 

This should be extended to dealing with anything in town. Everything costs money, takes time, and has a chance of failing. Having the right friends reduces the friction. Couple that with upkeep costs and faction developments and you have sources of time and money pressure to keep players on their toes. Towns should be boring, sure, that's why you go on adventures, but that doesn't mean they should be easy. 

Encounter Motives Over Encounter Activities

About month ago I made a Bluesky thread (follow me) some people found useful, and I wanted to share it here. This isn't me cannibalizing my thoughts across platforms. This is an adaptation.

A lot of people conceptualize random encounters as the party encountering the monsters, but they make more sense the other way: the monsters encounter the party. Encounter activity tables/"what are these monsters doing" lists can be helpful tools in certain situations, but they represent a way of viewing random encounters that works against how wandering monsters function best in the game. 

Arnold K's seminal "WTF are these goblins doing?" table is the clearest and probably most popular example of what I'm getting at. 

This table works well for a number of reasons. 

1. It conveys a large amount of information about goblins and their lifestyle, habits, and behavior in a condensed form. Even without making explicit use of the table, it gives a GM a strong enough impression of what goblins are like that they can easily make them come to life at the table.
2. It allows repeat encounters to be distinct from one another. 
3. It gives goblins depth. If you run into a goblin in the middle of something it makes them seem like a living creature instead of a knife tied to a bag of hit points. 

Nowadays its even common to bake activities in to the encounter table itself. "2d6 goblins playing kiss the blowfish" is more exciting than "2d6 goblins." And you get two of the three benefits above without having to rely on a whole other table. 

But, despite these benefits, monster activities such as these have a number of downsides. 

First, they imply the monsters are stationary and not wandering. 

Monsters hang out in a hallways, sure, but stationary activities are generally the purview of room encounters—what you key when you're prepping a dungeon. Random encounters can happen anywhere, even when the players themselves are stationary. Think about what happens if a random encounter is rolled while the PCs are in an empty room. The monsters are the ones on the move, so they're not going to be engaged in activities like "playing knucklebones" or "repairing a trap." This is why, traditionally, random encounters are rolled on a "wandering monsters" table.

Additionally, such encounter activities imply the PCs are walking in on a group of monsters in the middle of something. 

While surprise is a possibility, it's not guaranteed for every encounter. Some games even deny the possibility that monsters can be surprised by a party carrying blazing torches through the inky darkness of the underworld. 

Above is from OSE, below is OD&D Vol. 3: THE UNDERWORLD & WILDERNESS ADVENTURES

Since in most circumstances monsters are aware of the party's presence, it's reasonable to assume they'd have stopped what their doing by the time the party sees them. 

Most important, encounter activities don't tell you WHY monsters are interested in the PCs. If encountered monsters are preoccupied with something else, careful players have little reason to engage with them, and it's harder to contrive a reason for why monsters should care. 

When that happens, encounters run the risk of becoming just the DM describing monsters engaged in an activity and then the players going "ok cool, let's move on."

Saying "just ignore nonsensical results!" doesn't work here. If a tool assumes we will be regularly ignoring what the dice tell us to do (at the table, specifically; I contend prep is different), we could be using a better tool. 

Are these kinds of encounter activities always bad? Absolutely not. But there's something else we can use to add color to random encounters.

Wandering monsters already have an activity baked in (wandering, of course), but we can add more dimension by determining their MOTIVE. The expanded reaction table that's been floating around for a while is a good example of what I mean. A monster's needs inform the encounter.

Please tell me if you know who made this

Determining wandering monsters' motive doesn't just tell you why they're wandering, it—crucially—tells you what they want from the PCs, and how they'll go about getting it. That's a lot of bang for your buck! 

Ditto above

When designing encounters, think about activities that can answer these two questions:

1. Why are these monsters wandering?

2. What would these monsters want with the PCs? 

These questions don't need to be elaborate answers. You don't have to be particularly clever or creative when coming up with monster motives. In fact, simpler answers may be better because the players catch on faster. 

"Looking for food," "guarding territory," "hunting for treasure" are all suitable motives. Acclaimed Canadian Levi Kornelsen points out: "In human complexes where work is done, a *huge* amount of the time, what wandering humans are doing in the halls is moving things around. Food, waste, furniture (it me!), bodies, materials, messages. Any hive of activity is constantly shuffling stuff around." 

Not only do motives give immediate explanation for what monsters are doing and why, they tell you what kinds of decisions they would make when encountering the party. 

The 10 types of special rooms

"A Special is anything not exactly a trap, but placed for special reasons." Moldvay Basic pg. B52

Traps, monsters, secret doors, and treasure all cohere into continuous flows of challenge/reward and mystery/discovery. Special rooms, on the other hand, largely stand apart from these conventional elements while also replicating them in miniature: drinking from the magic fountain might raise an ability score but it might also make you vomit bees. The one-way elevator to the sixth floor might be a death sentence for low-level parties but incredibly valuable for experienced adventurers. 

The purpose of this exercise is to categorize common forms of special rooms found in old school-style dungeons. 

While the point of special rooms is to have bizarre, unique dungeon set pieces that defy expectation, having a list of common forms might help DMs come up with new ideas, or new ways to approach conventional ones. 

A trio of caveats: 

1. This is not intended to categorize the specific contents of special rooms (fountains, statues, etc.), but rather different meta-categories of special room.
2. Special rooms may have two or more categories layered atop one another; i.e. the reward of a puzzle room may be transit to another part of a dungeon.
3. These categories are not exclusive to special rooms—monster lairs, traps, and other kinds of rooms can also incorporate elements from this list. 

Anyway there are 10 types of special rooms

1. Random Effect
2. Experiment
3. Alter Environment
4. Exchange
5. Side Quest
6. Puzzle 
7. Minigame
8. Transit
9. Trick
10. Curiosity

Random Effect 

• The room with the magic pools from In Search of the Unknown. 
• The room with the magic stone, also from In Search of the Unknown.
• A white tree bearing opalescent berries; eating one imparts a random boon or bane
• A strange monolith covered in glowing runes of various sizes and colors. Each one does something different when touched. 

Random effect rooms offer players an opportunity to interact with something and roll on a table to see what happens. Alternate version is when there are multiple things to interact with and each one has its own effect. In both cases, the effects are unknown to the players and the general assumption is they are figuratively rolling the dice to see what happens. This is the OSR DM’s fast ball. The most fundamental special room, and the perfect crystal distillation of the OSR “play to find out” ethos. All you need is a random list of oddball effects and distinctive set dressing to frame it and you’re all but guaranteed a memorable table experience. 

Not every kind of dungeon needs a room like this, but every dungeon has space for one. However, much like the fastball, it can get stale if you use it too frequently. I like having these on an early dungeon level, since low-level characters are easier for players to gamble with.

Experiment

• A solar laser trained on a stone plinth; pulling a chain activates the laser, cutting anything on the plinth in half. 
• A machine that spits out two shoddy copies of whatever is thrown in its hopper. 
• Jeff Rients's classic tiny door. 
• Anyone who passes under the gaze of the Mercury Idol turns to mist for the next 1d6 turns (as Gaseous Form). Nonliving material like equipment and armor are unaffected. 
• A pedestal polymorphs anything placed upon it into a beautiful vase. If removed, the vase reverts to its original form after a day or if it shatters. The vase currently atop the pedestal is a polymorphed green slime. 

Experiment rooms offer a feature of open-ended utility to mess around with. You are probably familiar with them from the Dungeon Checklist. 

Experiments simultaneously resist understanding yet invite interaction, putting them in the sweet spot of good dungeon set pieces. The act of engaging with one is itself rewarding, because once the players learn the rules of an experiment they get to exploit it. The risk is lost goods, unwanted mutilation, and the like, while the reward is a new way for the players to use the dungeon to their advantage. 

Experiments operate on two axes of challenge: the challenge of figuring out how to properly interact and the challenge of figuring out how to exploit. Most good experiments only have challenge along one of these axes—in other words, something that is difficult to figure out how to operate should have a pretty good payoff (the splicing vats from Gardens of Ynn are a good example: there’s a complicated multi-step process you essentially have to trial-and-error, but getting them to work can lead to some pretty major benefits with no downside), whereas something that has an open-ended but not immediately beneficial utility should be rather straightforward to operate. 

Alter Environment

• A valve that raises or lowers the water level throughout the dungeon. 
• Flipping the switch to the blue side opens all the blue doors and locks all the red, flipping it to the red side does the reverse. 
• Removing the skull from the center of the pentagram banishes the warding seals and spirit guardians but awakens the Hungering Darkness. 

Alter environment rooms offer means to change how the dungeon is navigated. This can mean any alteration to layout, doors, hazards, and so on. 

It is widely understood that good dungeon design involves both rooms where interacting with something in one influences something in another and that player actions can have an influence on the shape of the dungeon. Alter Environment rooms accomplish both of these imperatives.

The more area of the dungeon affected by an alter environment room, the more it becomes a central gimmick. If it's just a one-off thing where the lever in room x raises the gate in room y you're not communicating much of a theme, but if the players must repeatedly jump through the magic mirror to solve puzzles in the mirror-dimension version of the dungeon then it becomes a mirror-dimension dungeon in addition to whatever else you have going on. Good to keep that in mind. 

Exchange

• Pour a potion in the Purple Witch's cauldron and she'll ladle you a vialful of her brew, which has the effect of a random other potion.
• A giant copper-shelled snail eats copper pieces and excretes gemstones of comparable value.

Exchange rooms offer an opportunity for players to change their equipment, gear, or treasure in unique ways. 

Exchange rooms are particularly effective if you frequently use randomly generated treasure hordes, as they give the players an opportunity to offset the quirks of the dice in the event they end up with six scrolls of protection from lycanthropes or whatever. They also expand the toolset available to the players in dungeons, allowing for a selectively reliable way to trade unwanted inventory for something more useful providing new access to goods or services.

Side Quest

• A statue comes to life when its missing hand, found elsewhere in the dungeon, is restored. 
• A vault's crystal door only opens once all the crystal braziers scattered throughout the dungeon level are lit. 

Side Quests offer a new goal for players to pursue in addition to their conventional dungeoneering activities.

Side quests work because they provide a little spice to dungeon exploration, giving players a new avenue for progression. They also bring that much-coveted interconnectivity because, as you can guess, the object of the side quest must involve at least one other room.

Puzzle

• Giant statue of a long-dead emperor looking menacing; on the statue's plinth an inscription reads "Salvation through submission." Doors lock, poison gas fills the room, in a low hidden niche in the plinth—clearly visible to anyone kneeling before the statue—is a small switch that deactivates the locks and gas.
• A room full of mirrors; one mirror reflects a door on the opposite wall that's not actually there. The mirror that reflects the door has a secret passage behind it. 
• Anything involving a riddle.

Puzzles offer an in-game scenario that directly challenges the players’ ingenuity (or ingénue-ity, but I have less experience with that). 

It should go without saying that everything in a dungeon can be a puzzle, like combat, traps, navigational hazards and so on, so special rooms are no exception. 

Puzzles can be contained within a single room, but they don’t have to be—a common tip for designing puzzles is to seed clues in other parts of the dungeon to promote interconnectivity. 

“But if a puzzle has clues seeded around the dungeon, what distinguishes it from a side quest?” the hypothetical gun aimed at my head says. “What’s the difference between moving room to room to collect macguffins vs. moving room to room to collect clues?”

 

The main difference is: Puzzles are about thinking and Side Quests are about doing. Imagine a treasure vault guarded by a demon who will only let the party through if they utter a specific passphrase. In a scenario where each word in the passphrase is found within a different room, and the sequence is rather easy to intuit, this challenge is a side quest—it’s pretty easy to figure out what the problem is and the challenge becomes finding each word. But in a scenario where the passphrase is the solution to a riddle, or encoded in a mural elsewhere on the level, you’ve got yourself a puzzle. Admittedly there can be a lot of overlap between the two, but my rule of thumb is: the more demanding a side-quest element is, the less demanding the figuring-out parts should be. And vice versa. 

Minigame

• A pack of goblins invite you to race alchemically powered junk-karts. 
• The mind-cube imprisons you in a mental maze, resolved by solving an actual pen-and-paper maze. 

Minigames offer an in-game scenario resolved through novel game systems. 

Minigames make for good diversions and give players a break from the main game. I ran a session recently where the party got access to an exclusive gambling hall. We spent most of the evening playing through the dice games in Appendix F of the AD&D DMG. It was a lot of fun, but not the kind of thing anyone wants to do every session. 

Transit

• Room drops off to a massive void; lighting the candle on a nearby plinth attracts giant floating jellyfish that bring anyone who grabs ahold a harmless tentacle to a deeper dungeon level. 
• Two identical circular chambers, each with a lever on the floor and a single door, north end for one and south for the other; stepping in one chamber and pulling the lever appears to rotate the room, door and all, 180°, transferring everyone to the other chamber elsewhere in the dungeon.

Transit rooms offer novel means to navigate elsewhere in the dungeon beyond your typical hallway or staircase.

Transit rooms allow for alternate routes for navigating a dungeon, can help parties save time and cut down on backtracking on repeat dungeon ventures, and can serve as navigational challenges. 

Trick

• Illusory stairs conceal a spiked pit. 
• A statue of a stern-looking angel. Anyone that approaches within a few inches must save vs. paralysis or be grabbed and restrained by the statue, which proceeds to release a blaring alarm.
• A room lined with racks and racks of ornate swords; removing one causes the swords to animate and attack.

Trick rooms offer a nonstandard challenge or hazard. 

Tricks feel very old-school in the sense that many of the ones you see in classic modules feel deliberately designed to mess with the players in a kind of funhousy meta-gaming-from-the-DM-side way. That flavor of adversarial DMing has fallen by the wayside these days, as well as funhouse dungeons in general, but a thoughtful trick here or there can create a memorable moment and convey the more malevolent aspects of the mythic underworld. A hallmark of a certain kind of old-school mindset is that having horrible things happen to your character can be as fun and engaging as having them succeed. 

Curiosity

• Room echoes with the voices of everyone ever slain in the dungeon. 
• An idiosyncratically cozy and well-appointed study, totally unlike the rest of the dungeon, complete with wizened sage reading & smoking a pipe in front of a warm hearth, somehow unmindful of the fact he's in the middle of a giant dungeon. 

Curiosity rooms offer a unique and distinctive location that lends to the dungeon’s mystery/majesty/dread.

Dungeons should have a healthy amount of idiosyncratic areas interspersed among standard rooms and corridors filled with traps, monsters, and treasure. Sometimes, these idiosyncratic locations may not have a clear utility, challenge, or problem to solve. Curiosity rooms are less focused on the objects within than the locale itself; while they may be focused around a distinctive set piece, like a unique NPC, the primary effect of a curiosity room is atmospheric.   

Art by Pedro Friedeberg

Notes on OSR-style social challenges

We know that rolling ability tests or skill checks for social situations is an ill-considered way to resolve social situations because it stops players from thinking—it turns a strategic challenge into a dice roll.

That being said, having no procedure for dealing with social challenges could lead to a lot of circuitous and time-consuming back-and-forthing between the players and the DM until it is decided that the situation is resolved either favorably or unfavorably for the players. Relying on DM fiat is fine in a lot of circumstances but too much can lead to decisions that feel arbitrary—the DM either lets the players do what they want, in which case they can feel like they are getting away with something, or not, in which case they can feel like the DM is being hard and punitive. 

The reaction roll does a good job of providing a simple mechanic for social challenges. The offer refused+lowered standing/offer refused/unsure/offer accepted/offer accepted+added benefit dynamic complements the 2d6 probability spread well. 

But what about the actual challenge? 

The obvious answer: PCs should to be able to do something gain additional bonuses to the roll. What Arnold K. brought up some months back in his post on parleying can be applied to all forms of social interaction where the desires of the PCs and the NPCs are different. Adding a goal, value, or motivation to an otherwise one-note NPC not only makes the world seem more vivid etc. but more importantly gives the character something the PCs can latch on to and use in negotiation. 

I'd only want to do this if the players are going to want something from an NPC that they are not necessarily willing to offer. You don't want to add texture to every NPC for the same reason you wouldn't want every supporting character in a book or movie to have a full backstory and character arc—it adds a lot of extraneous detail that pulls focus away the important parts.

The game turns into a playacting slogathon if every pre-dungeon supply run involves stilted RP with a shopkeeper. Feeling like you need to fully roleplay every conversation the PCs have is a mistake a lot of newer DMs run into, but really you only need to zoom in on the things that actually provide engaging gameplay. 

If, for example, the PCs have a fake gem or conspicuous stolen art piece they are trying to dupe someone into buying, considering the identity of the NPC and what they value suddenly becomes more important. Are they religious, and thus more willing to trust people of similar faith? Do they like to drink perhaps a bit too much, and thus  inclined to be genial with someone who takes them out for a beer? Instead of just having a flat chance of the object getting identified, there is now a way for the players to strategize a method of maximizing their chances for the shopkeeper in question to agree to buy the thing. 

Of course, these features would need to be telegraphed. The shopkeeper has to be seen displaying a holy symbol or acting a bit buzzed before the players can know how to game them. 

This is the kernel of a social challenge: the PCs want something from an NPC; the NPC in the way of what they want has a discernible characteristic, and it's up to the players to identify and exploit that characteristic to gain the NPC's favor. Bonuses to the reaction roll could range from +1 to +3 depending on how much the players invest into their efforts.

Brainstorming a couple NPC values + ways to telegraph them:

1. Romance. Flirts unabashedly with the prettiest member of the party. 

2. Kindness. Mopey and dejected. Perks up at the slightest compliment.

3. Booze. Heavy eyelids, slurred speech, periodically pulls from a flask.

4. Faith. Casually recites lines of scripture. Wears a holy symbol and/or has one prominently displayed somewhere. 

5. Flattery. Loud and boastful. Demands to be called "sir" or "madam" or by some other official-sounding title. 

6. Status. Snooty social climber. Gaudy clothes, always gossiping, directs attention to the most important-seeming person in the room.  

7. Company. Quiet and sullen at first, but will launch into an animated conversation at first opportunity.  

8. Niche interest. Conspicuously displays their fixation (garish hats, porcelain dolls, painted tortoises, etc.).  

9. Process. Surrounded by piles of forms and paperwork; constantly refers to various rules and regulations; fervent adherence to the strictures of exhaustive bureaucracy. 

10. Relief. Currently burdened by some misfortune like a stolen heirloom, sullied reputation, blood feud, weird curse, etc. made obvious to the PCs.

Obviously not an exhaustive list. The point is to give something the players can see that makes them think "hmm maybe that's something we can exploit." 

These sorts of desires/values can stay general or become really specific as the situation demands. More challenging social challenges would necessitate more specific values. Perhaps the only thing the high slayer of the headsman's guild values more than duty is his beloved golden python he keeps as a pet, hidden somewhere in his watchtower lair. Good luck figuring that one out.  

This post from Was It Likely? provides another great option (by way of this post from To Distant Lands)  for creating NPCs with motives and desires that have fuck-all to do with the PCs. Allow them to permeate and enrich your mental ecosystem. 

What are social challenges for?

Like traps, puzzles, and monsters, social challenges should act as an obstacle between the PCs and their goal (which is in most cases treasure, but y'know not always). Social challenges fit nicely in areas where traps and monsters aren't appropriate, like in settlements, or when you want to add more variety to an otherwise challenge-rich environment.

While combat and puzzles can contribute to the adventuresome spirit of the game, social challenges are in many cases better suited for practicality. When the players want something that shouldn't be too easy to accomplish, popping in an NPC that tells them "no" can often be all you need in the way of an obstacle.

Of course, the PCs could always just choose not to engage and just be like "Hey, you should let us pass we're friendly and you can trust us," in which case just a flat reaction roll would work. But an additional dimension to an NPC opens up new avenues for the players, similar to how more detailed and interactive elements in a combat encounter give the PCs more things to play around with. Not to mention they could always resort to violence or magic or whatever. 

I find that players are less like to take the mercenary/murderhobo approach to dealing with NPCs when they are given another option to deal with them. Like Arnold's example in the post linked above, it's no wonder players would default to killing every guard that gets in their way if every one is perfectly disciplined and unable to be bribed, intimidated, or otherwise convinced. But give a PC a reason to get shitfaced with one and they'll take it, 9 times out of 10. 

Art by Max Ernst, the eminently D&Dable surrealist.

Conflict-forward factions

Most advice on how to design factions can usually can be summarized as “[group] wants [goal], but [complication]; fill in the gaps.” This is a solid and fast way of getting ideas down for groups but I find that it, at least when making factions for a full campaign as opposed to a dungeon or adventure locale, it's largely left up to the DM to come up with compelling interplay— the most important part and, as is often the case, the hardest to do on your own. What this method lacks is a way of organizing the factions within a larger framework or scenario for the players will contend with. 

What I suggest instead is a way to think about faction creation that starts with the overarching circumstances the factions exist within and working outward from there in order to build out how the figurative position each faction occupies in those circumstances and thus in turn the potential energy of how they might play off one another (once the players get involved). 

清已

So where to begin? Conflict. Conflict is always compelling because it gives rise to change, which gives rise to opportunity. It creates an instability or tension from which could arise any number of new or unpredictable scenarios. Because of that, compelling factions ought to be in involved in conflict. Instead of coming up with the factions first and figuring out why they hate each other, it's more productive to start with the conflict itself. Working from a blank slate, that means identifying a scenario characterized by instability (the root of conflict) and designating factions based on different positions a group can take relative to the resulting tension. 

There's no hard-and-fast rule for this but generally instability appears in the interim between when one paradigm declines and another is ascendant, or when something big and consequential happens unexpectedly. 

A great example of a sandbox adventure that takes full advantage of instability is Better Than Any Man, probably the best thing to come from LotFP. You have the foreground circumstances of the magic strangeness going on in town (instability) coupled with the emergence of a mysterious cult (more instability), as a war in the backdrop (but wait!), with the looming threat of the Swedes (there's more!). Interweaving layers of conflict form a beautiful tapestry of mayhem. I also did something similar in The Cerulean Valley, my never-before-plugged adventure in noisms' sensational zine In the Hall of the Third Blue Wizard vol. 1. Check it out. 

Here's what it looks like building this from the ground up. The situation: 

A few years back, a cruel warlord invaded a thriving city-state, slew the aging king, and installed himself as the city's Overlord. 

Given that this is a city locale, we'll take the bog-standard D&D city fixtures to become our factions: nobles, wizard's guild, church, and thieve's guild. We're also going to have the Overlord and his agents as their own faction, since they make up one pole of the conflict. 

Once the factions are decided, it's time to identify different ways a group can perceive the circumstance they're in, based on what hardships and what opportunities this tension offers a given faction. Or even more simply, just thinking of what different perspectives would be appropriate in this scenario. In its most basic form there are the two sides to the conflict, the old thing at the new thing, and a handful of groups that seek to exploit the conflict for their own gain—one who wants to swoop in when both are weak, one who wants to side with the whoever has the most to offer them, one who wants to prolong the conflict and maximize chaos, and so on. 

First is the Overlord and his agents. Perhaps they have insinuated themselves into the cities municipal and trade bureaucracies and now control the city’s commerce and the rule of law. They're the ones in charge, but the transition has been rough and their hold over the city is tenuous at best. 

Opposed to them are the noble families, who have been allowed to maintain a diminished stature in exchange for their deference to the Overlord. Some are content and try to make the most of the situation, but by and large the nobility are against the new regime and look for any opportunity to undermine the Overlord and return things to semblance of the way they were before. 

A powerful group indifferent to either side is the magicians’ Collegium, a loose conglomerate of magic-users sanctioned by the city-state. The Collegium was formed generations ago to allow magic-users to collaborate and work as they please in exchange for agreeing to keep civic disturbances within a manageable threshold and come to the city-state’s aid in times of need. The agreement has remained unchanged in the new regime. The Collegium typically wants everyone to fuck off and leave them to their business, yet given that the institution possesses massive power that can only marginally be controlled by politics, they have much to offer both the Overlord and the royalists. While they're tentatively willing to support whoever has the most to offer them, they're more interested in suppressing both sides and sweeping as much influence as they can. 

The faction who wants to side with the group with the most influence is the Church. Already a fossilized institution before the new regime, their relevance has only further declined since the Overlord has little interest in preserving the traditions and observances that were at least nominally upheld under the king. Still, the Church has the power of divinity on their side. Within the vast catacombs beneath the great cathedral are secrets of immense consequence, forgotten by all save for their holy custodians, granting them a not-insignificant amount of leverage. The Church hopes to reclaim its influence by siding with whoever has the upper hand, both in the hopes of bringing an end to the conflict and also to curry as much favor as possible with whoever comes out on top. 

And the group that thrives in the turmoil and wants to prolong it for as long as possible is the thieves' guild, or maybe the local serpent cult or whatever you should get the idea by now. 

So here we have an idea of not just what each faction is and what they want, but also how each faction’s respective want ties into a larger overarching theme. Potential conflicts and factional moves can emerge from this organically, while allowing it all to stay focused around a larger tension. Of course, you can have a bunch of factions with disparate goals (ie the magic-users want magic artifacts, the thieves' guild wants money, the church wants adherents, etc.) but organizing the factions from the center outward like this provides you with an organizing framework to more easily build adventures.