Demo Tables: Teaching a Stranger in Minutes

GrantPicture a stranger standing at a demo. Three minutes of patience tops, then they're gone.

MilesOuch. Which stage, though? A screen or a table?

GrantBoth. That's the beauty of it. Okay, so get this. Steam Next Fest runs October nineteenth to twenty-sixth.

MilesAnd SPIEL Essen opens October twenty-second.

GrantOverlap. Somebody out there's gonna alt-tab out of a demo to book a train to Essen.

MilesWelcome to the Design Deck. I'm Miles, and yes, that alt-tabbing person is me.

GrantAnd I'm Grant. Plot twist, I've already seen your wish list.

MilesHush. So here's my question: Can a free demo teach a game as well as a human sitting across from you?

GrantSystems answer, yes. A good demo is a tutorial that never gets tired, never skips a rule, never has a bad day.

MilesMm, no. A human watches your face. They see you squint and slow down. A built can't do that.

GrantOh, you're gonna regret that.

MilesWe'll see who regrets what.

GrantFine. But before we settle it, we should look at how crowded both stages are.

MilesCrowded is putting it mildly.

GrantBecause it's the same problem on both stages. A stranger shows up with no context and very little time.

MilesAnd whoever's running the demo, screen or human, has to turn that confusion into something good before the patience runs out.

GrantOkay, so get this. Steam has a rule that sounds boring until you realize what it does to the shelf. Steamworks documentation for the October edition says a game needs a publicly playable demo by the time the festival begins.

MilesSo no vaporware with a nice trailer.

GrantRight. And the store page has to be public. Plus, a prologue or chapter one of content that customers already get doesn't qualify.

MilesOh, that's interesting. So it's not a cheap sampler of something already out. These are upcoming games shown cold.

GrantShown cold. Now, how many? A site called Steam Page Analyzer captured one hundred and thirty-seven unreleased store pages with a playable demo on fourteen September.

MilesOnly a hundred and thirty-seven? That feels tiny.

GrantHold on. That's a snapshot from weeks before the fest. It is not a count of Next Fest demos. Think of it as a photo outside the door.

MilesFair. The real crowd shows up later.

GrantExactly. Okay, now cross the Atlantic because Essen does the same thing with cardboard.

MilesAnd Essen is a different animal. I've walked those halls with a notebook and a blister. The organizers' press release on last year's show gave the baseline: nine hundred and forty-eight exhibitors from fifty nations.

GrantNine hundred and forty-eight booths, each with somebody who wants you to sit down.

MilesAnd over seventeen hundred new releases. Those are last year's numbers, mind you. The same release says this year's show runs October twenty-second through twenty-fifth, but it gave no twenty-twenty-six totals.

GrantSo I can't promise this year is bigger.

MilesYou can't, but I wouldn't bet on smaller.

GrantNobody ever walks into Essen hoping for fewer boxes.

MilesNever. So what does a normal person do with all that?

GrantThere's a developer marketing guide from Gamosy, and it says players try five to ten demos and give each one ten to fifteen minutes.

MilesSource?

GrantNope. The guide doesn't give one. So treat it as an industry rule of thumb, not a measured fact.

MilesStill, it tracks with how I behave. Five to ten tries, and most of them die fast.

GrantDo the math. Ten demos at fifteen minutes each is two and a half hours, and your whole evening is gone for hundreds of games.

MilesPlot twist, you're the bottleneck.

GrantThe player is always the bottleneck. A designer can make a thousand games, but a stranger has one evening.

MilesWhich is brutal if your first ten minutes are a rule book.

GrantAnd that's the designer's real competition. Not just the other game on the shelf, but the stranger's evening and the next demo in the queue.

MilesWhich means every minute spent explaining is a minute not spent playing, and the stranger can feel the difference.

GrantSo attention is scarce. Let's see how a Steam demo spends it. Okay, so get this. One developer marketing guide, a Steam Next Fest survival guide from Gamosy, has a very short list of demands for the opening. Skip the logos, skip the text crawl.

MilesOof, my favorite part of every game.

GrantI know. You love a scrolling wall of lore. Gamosy says to drop the player into something interesting within about two minutes.

MilesTwo minutes.

GrantAnd teach through play instead of pop-ups. No little window saying "Press X to jump" while you stand there.

MilesFine. That one I'll give you. Nobody reads the pop-up anyway.

GrantThen there's a second guide from Immutable on how long a demo should be. It says the defining interaction should land in the first one to five minutes.

MilesWhy that window?

GrantBecause festival players sample several demos and leave quickly. And to be fair, both of these are secondary advice. Practitioners, not Valve data.

MilesRight. So it's smart people's opinions. Good opinions, though.

GrantAnd notice what they have in common. Both are about minutes, not hours. The clock starts the second the demo launches.

MilesWhich is a scary clock when the player can just close the window.

GrantHere's why I love it. Think of Portal. First room, you've got a portal gun and a wall. No lecture. The level itself teaches you the verb.

MilesHold on. Portal is a great example, but it proves my side too.

GrantOh, here we go.

MilesPortal teaches the rules and it's funny. It has a voice talking to you. You can build a tight tutorial that's perfectly efficient and still leave a player feeling like they filled out a form.

GrantOkay, but a form is fast. In a demo, fast wins.

MilesDoes it, though? Picture a platformer. It drips out jump, then double jump, then dash, each one in its own tidy little room. Every rule lands exactly on schedule.

GrantAnd?

MilesAnd you feel nothing. The controls were taught, the game wasn't.

GrantOh, so the problem is a tutorial that's too good at its job.

MilesThe problem is a tutorial with no personality. Give me one weird moment in minute one, a character who says something odd, a sound that makes me grin.

GrantWait for it, because I can answer that. The weird moment can be a system too. Make the first jump matter. Put a gap, put a coin, put a consequence right there.

MilesMm, so the choice is the charm.

GrantExactly. If I'm deciding something in minute one, I'm already invested.

MilesGive me a concrete minute one then. What does that first choice actually look like on screen?

GrantSay it's a deck builder. Before any explanation, the game hands you two cards and a single enemy, and you pick which card to play. No menu, no glossary. You see what happens, and now you know what a card does.

MilesOkay, that's a real choice, and it's teaching. Cards, enemy, consequence, all in one move.

GrantAnd notice nothing in there needed a pop-up. The choice was the lesson.

MilesBut I'd still lose you if the whole thing felt like a spreadsheet wearing a costume.

GrantFair. Tight and warm. I'll take both, but I'm sequencing the tight part first.

MilesAnd I'm sneaking the warm part into the very first screen. We'll see who's right.

GrantCold or warm, though, at least the screen doesn't get tired or lose its patience with you. It just waits.

MilesWhich is exactly what a person across the table can't do. Switching gears, no screen, no pause button, just someone with a rule book and a stranger sitting down.

GrantOh, that's a whole different kind of pressure. Okay, so a booth explainer. You think the job is reading the rule book out loud with better eye contact.

MilesThat's the amateur version. A demoer with SN and UK Games Expo behind them wrote a first-person account, and they say they rarely teach all the rules.

GrantRarely? Not even close to all?

MilesNope. They give the setting, the actions you can take, and the choices in front of you. Then they explain only what you need to start.

GrantSo it's the anti-rule book. I love it.

MilesThe goal is that you feel you can begin immediately. Everything else shows up when it matters.

GrantHold on, though. Doesn't that leave holes? Someone discovers on turn four that they could have done something on turn one.

MilesSure, sometimes, but that same demoer says convention players tend to go with the flow because their time is limited. They'd rather stumble forward than sit through a lecture.

GrantHuh.

MilesHome groups are the opposite. They often want the full rules explanation before a single card moves.

GrantWhich makes sense. At home, the table is the whole evening. At a booth, the table is a line of people staring at your back.

MilesStaring politely, but staring.

GrantWalk me through it. Say I sit down at a booth for a card game I've never heard of. What's the first minute sound like?

MilesGoing by that approach, something like, "Here's the setting. Here's what you can do on your turn. Here are the choices in front of you. Go ahead and pick one." Notice what's missing. No design history, no scoring appendix, no list of exceptions.

GrantOkay, so is there a how-to for the hands, not just the mouth?

MilesThere is. Pine Island Games, a small publisher, put out convention demo tips. Let people handle components or roll dice during the teach.

GrantOh, that's smart. Your hands are busy, your brain stops drifting.

MilesAnd they say skip the detailed turn structure explanations. Keep the teach fun.

GrantTurn structure is the first thing every explainer wants to recite and the first thing everybody forgets.

MilesThey also say to sell softly. Set expectations, offer a few ways to engage, and don't pressure anybody into buying.

GrantWait for it, because this is where I get suspicious. Doesn't soft selling just lose sales?

MilesMaybe on the spot, but a stranger who felt cornered doesn't tell anyone about the game. A stranger who got to roll the dice might.

GrantFair. Now, here's what's bugging me. Listen to what both of these people are doing.

MilesCutting.

GrantCutting. The Steam demo cuts the logos in the text crawl. The booth explainer cuts the rules. Different stages, same trick. Get the stranger making a real choice fast.

MilesAnd the human has a huge edge. They watch your face. If you look lost, they simplify. If you look bored, they skip ahead.

GrantWhereas the demo has to guess. It's teaching a stranger it can't see.

MilesWhich is why I say the booth is the harder job. You're adapting live to a dozen different people all day with a sore throat.

GrantNo, no, the demo is harder. A human gets to fix a bad guess in five seconds. The demo's guess ships to everybody forever.

MilesHmm. Okay, that's a decent point, but the demo never loses its voice at hour six.

GrantTrue. Machines don't need tea.

MilesThough I'll admit the demo's version of a sore throat is a build that stumbles right when a stranger is paying attention.

GrantAnd nobody's standing there to apologize for it. That's the cost of teaching without a face to read.

MilesEither way, we've agreed on what the first minutes are for. The cutting is where people actually fight.

GrantSo what actually goes on the chopping block and what survives? Okay, so the cut list. First fight, how long should the demo even be? Immutable's practitioner guide says about thirty minutes of focused play.

MilesThirty? Gamosy's survival guide says fifteen to twenty-five.

GrantSo even the advice can't agree. Wait for it. Neither one is Valve data. It's two guides with opinions.

MilesWhich is fine. Nobody's giving you a law, you're getting a range.

GrantRight. And the part I like is what Immutable says about structure. The common shape is a horizontal slice.

MilesExplain that for the people who skipped game dev school.

GrantYou show the major mechanics, all of them, but you limit the content. A little of everything, not a lot of one thing.

MilesHuh, so the cut is content, never the mechanics.

GrantYes. You cut levels, not verbs. Cut the sixth biome, keep the grappling hook.

MilesMm. But I'd cut something else first, the tutorial. Not all of it, just the part that stops the game to talk.

GrantDude, I'm with you. Gamosy says teach through play instead of pop-ups.

MilesSure, but you were the one who wanted every system explained cleanly last time.

GrantExplained by the game, not by a banner. A door that only opens one way teaches faster than a paragraph.

MilesAnd what about content you love? Cutting that sixth biome hurts if it's where the best art lives.

GrantIt does, and that's the point. Whatever you cut should hurt a little, or you're not cutting enough. The demo needs to be a clear sample, not a greatest hits reel.

MilesFair. Now the first turn hook. Do you cut that to save time?

GrantNever. Never ever. The hook is the one thing you protect because it's the first real choice.

MilesProtect it how, though? Give me something a designer can actually check.

GrantOne question and a stopwatch. How long until a stranger decides something with a visible result? If the answer is after a cutscene and two menus, those are your cuts.

MilesAnd the booth has the same problem, just with a human holding the knife. Think of a heavy Euro, something like Brass.

GrantOh, you're gonna cut the turn structure.

MilesGone. I won't walk you through every phase of a round. I'll put cards in your hand and say, "Pick one."

GrantAnd the edge case rules?

MilesThose die first. What happens when the deck runs out, how ties resolve. Nobody at a convention table needs that in minute four.

GrantHold on, though. If the explainer skips it and you buy the box, you hit that tie at home and feel cheated.

MilesMaybe, or you open the rule book, find the answer, and feel clever. I'll take that trade.

GrantI'd still keep one sentence of warning. Something like, "There's a wrinkle later. Trust the book."

MilesA footnote with a smile. Okay, I can live with that.

GrantSo whatever survives all this cutting is what sits on the table or on the screen for a stranger to judge.

MilesRight, and that's the test of a good cut. If the stranger can still see the whole shape of the game from what's left, you cut well.

GrantAnd if they can't, you cut the wrong thing.

MilesAnd that's the interesting part. A demo that small can't hide anything. What's left says a lot about the design underneath.

GrantOkay, so get this. A demo is basically a lie detector for the design. If the core loop is muddy, 15 minutes will not hide it.

MilesHmm, loop meaning what, for anybody new?

GrantThe thing you do over and over. Draw, play, score, or jump, grab, land. If I can't tell what the loop is after a short session, either the game is hiding it or it doesn't have one yet.

MilesHmm, I'd say the sharper test is where the fun sits. Does it show up before the rules click or only after you've memorized them?

GrantOoh, say more.

MilesSome games are fun while you're still confused. You grab a card, something happens, you laugh. Others are homework first and fun later. A demo shows you which kind you're holding.

GrantGive me one from each venue. On a screen, what does fun before rules look like?

MilesSay you press a button and the whole scene reacts. Things pop and move, and you grin before you even know what the button was for. At a booth, it's someone handing you a die and saying, "Roll it," and the table reacts before you've learned why.

GrantWhich sounds like a clean test. Too clean, maybe.

MilesHere we go.

GrantHear me out. If a game can't be shown in minutes, that's a design signal. Red flag. Somebody should be able to make it legible fast.

MilesNo, no. Plenty of great games pay off late. Think of a slow engine builder where the first hour feels like laying bricks. The wall only looks like a wall at the end.

GrantFair, but then the demo owes me something. A taste of the wall at least.

MilesDoes it, though? That's the part I don't buy. A demo that can't show a late game payoff isn't failing, it's just short.

GrantOkay, so how do I tell short from broken?

MilesYou can't fully. That's the honest answer. A bad demo might mean a muddy design, or it might mean a demo built in a hurry by three people on no sleep.

GrantOh, the sleep-deprived factor. Very scientific.

MilesVery real, though. So I'd read a demo as a clue, not a verdict.

GrantI'll give you that, but a clue that points the same direction twice starts to feel like a verdict. If the first choice is dull and the explanation drags, I'm not waiting for the payoff.

MilesSure, your patience is a design constraint too.

GrantPlot twist, I'm the market.

MilesAnd for the record, none of this says anything about any specific unreleased game. We're talking about what the tool can and can't tell you.

GrantRight. It can tell you whether the fun shows up early. It can't tell you whether the ending earns it.

MilesWhich is a decent thing to know before you stand in a queue.

GrantAnd a decent thing to know as a designer. Ask which kinda game yours is, fun first or homework first, before you build the demo.

MilesAnd if it's homework first, be honest about it and give the stranger a reason to stay.

GrantSpeaking of the queue, time to wrap with what you can take to the show floor and onto the wishlist.

MilesShort list then. I've got a few things I'd scribble in the notebook.

GrantOkay, so the checklist. Three items, no new arguments, I promise. One, name the first meaningful choice a stranger makes.

MilesTwo, cut whatever doesn't feed that choice. Menus, side rules, the third faction nobody meets until later.

GrantThree, and this is my favorite, get the thing into their hands early. Cards, dice, a jump button, anything that isn't a paragraph.

MilesHandled before explained. Yes.

GrantWhich is the same rule on both stages. Make the stranger do something real and let the rule arrive exactly when it's needed.

MilesAnd if you only remember one item, remember the first one. Everything else is just protecting it.

GrantAnd we still disagree about the machine.

MilesWe do. Grant thinks a demo that teaches itself is the purer test. I'd still take a person across the table who can see a face go blank and change course.

GrantA face going blank. Miles, that's a bug report with eyebrows.

MilesFine, but the eyebrows are right more often than your analytics.

GrantWait for it. Listeners get the tiebreaker. Tell us who's winning at thedesigndeck.com or tag us on social.

MilesNew episodes every Thursday, so you've got a few days to build your case.

GrantAnd hey, next time you're standing in a crowded hall or staring at a wishlist, remember that stranger from the start.

MilesThree minutes of patience, one good choice. Make it a good one and they might even sit down for the second turn.