How to Build a Demo That Sells Your Game

TL;DR: Your demo is a sales pitch, not a chunk of your game. Keep it 15-30 minutes, end on a cliffhanger when players are most engaged, and release it weeks or months before Next Fest (not on Day 1). Most Next Fest wishlists come from people who never touch the demo, so your store page does the real converting.

Key Takeaways

68-88% of Next Fest wishlists come from players who never downloaded the demo, so your capsule art and store page matter more than the demo itself
Target 15-30 minutes of content: long enough to hook players, short enough for creators to finish in one session
End on a cliffhanger at peak engagement, not on a neat stopping point, and immediately show your wishlist CTA
Release your demo weeks or months before Next Fest to build momentum; entering with 0 wishlists means earning minimal during the fest
Use a separate Steam page for the demo to collect reviews and feedback

A demo is a sales pitch disguised as a free game. The moment you stop thinking of it as "a chunk of the full game" and start thinking of it as "a carefully constructed argument for why someone should wishlist," everything about how you build it changes.

This distinction matters most during Steam Next Fest, where over 2,000 games compete for attention in a single week. Your demo isn't just a thing players try. It's the thing that determines whether your game gets noticed at all. Without a demo, you don't appear in Next Fest. Without a good demo, you appear and then vanish into the algorithm's basement by Day 3.

But here's what breaks most developers' brains: the majority of your Next Fest wishlists will come from people who never play your demo. Zukowski's February 2025 survey showed that 68 to 88 percent of wishlists came from players who saw your store page and hit the button without ever downloading. The demo unlocks your visibility. Your capsule art and store page do the converting. Your demo exists to appear in the festival, to give content creators something to record, and to hook the subset of players who actually try it.

That subset still matters enormously. They're your most engaged potential buyers. They're the ones who leave reviews, join your Discord, and tell friends. So the demo needs to be excellent, even knowing most people will never touch it.

The Sweet Spot: 15 to 30 Minutes

Demo length is one of those topics where everyone has an opinion and nobody has conclusive data. But the consensus among developers who've done well at Next Fest lands in a consistent range: 15 to 30 minutes of content.

Short enough that a content creator can play through it in a single recording session. Long enough that the player actually understands your core loop and gets invested. What creators consistently report is that demos under 10 minutes feel like they barely started, while demos over 45 minutes rarely get finished on stream—the 15-30 minute sweet spot exists for a reason. Think of it as the pilot episode of a TV show. You've got one shot to prove the concept is worth their time.

Under 10 minutes and players feel like they barely started. Over 45 minutes and two problems emerge. First, content creators don't have time to finish it alongside the other 20 demos they're trying that week. Second (and this is the more dangerous one), players might feel satisfied. They played your game. They got the experience. The itch has been scratched. Why would they buy the full version?

Valve's own documentation is deliberately noncommittal on length, saying only that you should "balance giving the customer enough content to get them excited, without giving away so much that they feel like they've experienced everything." Which is about as helpful as a strategy guide that says "try not to lose." But the 15 to 30 minute window threads that needle for most genres.

For some genres, this range shifts. A narrative adventure might work with 20 minutes of a single chapter. A strategy or management game might need 40 to 60 minutes because the core loop takes longer to reveal itself. Adjust for your genre, but err on the side of leaving the player wanting more.

The Cliffhanger Technique

End your demo at the moment the player is most engaged. Not when they've finished something. Not when there's a natural stopping point. Cut it right when they're thinking "wait, what happens next?"

Dramatic cliff edge with path ending abruptly

This is counterintuitive. Developers instinctively want to wrap things up neatly, give the player a complete experience, and close with a bow. That instinct will cost you wishlists. Developers who've tested different demo endings report that cliffhanger endings consistently outperform "complete experience" endings in wishlist conversion, sometimes by 2-3x.

The best demos end on a cliffhanger. A door opens to reveal something the player can't explore yet. A new mechanic is introduced that they don't get to use. A story beat raises a question that won't be answered until the full game. The credits roll right after the first boss, and the player is left staring at the screen like the game just whispered "Continue? 9... 8... 7..."

Your end screen should appear at this exact moment. Thank the player. Ask them to wishlist. Point them to your feedback channels and Discord. The emotional momentum of "I want more" is what converts a demo player into a wishlister. Don't let it dissipate by giving them a cool-down period.

Is This Seat Taken nailed this during the February 2025 Next Fest with a simple, warm end screen that appeared right as the demo's story hit its most compelling hook. Nothing aggressive. Just present, at exactly the right moment.

What to Include (and What to Cut)

Your demo is not a vertical slice of your game. It's a curated experience that showcases your strongest content.

Include:

Your core gameplay loop, fully functional and polished
Your best visual set piece or moment (the thing that looks amazing in a trailer)
A clear sense of progression or story momentum
Your most unique mechanic, the thing that makes your game different from the 2,243 other games in Next Fest

Cut:

Tutorial sections that run longer than 3 minutes (teach through play, not text walls)
Systems that aren't ready yet (a half-finished crafting system is worse than no crafting system)
Content that spoils major story beats
Anything buggy. Players won't file a report. They'll close the demo and move on.

Some developers build a demo-specific experience that doesn't exist in the full game. A custom starting scenario, a tutorial designed specifically for the demo context, or a curated set of levels that show off variety. This takes more work but can produce a tighter experience than simply slicing the first 30 minutes of your game.

Technical Setup on Steam

Your demo is a separate App ID associated with your base game. This is a technical requirement, not a suggestion. The demo has its own depot, its own build, and its own release checklist.

Steamworks dashboard for demo configuration

To create your demo App ID in Steamworks:

Go to your base game's App Landing Page
Click "All associated packages, DLC, demos and tools"
Click "Add Demo"

You'll need to upload a separate build, configure depots, and go through Valve's build review. The review checks that the game runs and matches what you've described. It's not QA. Finding your own bugs is still your job.

One critical detail: after releasing the demo, you need to manually re-publish your base game's store page for the "Download Demo" button to appear. Developers miss this constantly.

Separate Store Page vs. Same Page

Since late 2024, Valve offers two options. You can have your demo live on a separate store page with its own reviews, or simply add a download button to your base game's page.

Zukowski's data from February 2025 shows a roughly 50/50 split in adoption with no meaningful difference in median wishlist performance between the two options.

But the separate page has one significant advantage: reviews. Players can leave reviews on your demo's separate page, and those reviews are some of the most honest feedback you'll ever receive. If the reviews go south, you can pull the separate page and the reviews vanish. Low risk, high information value.

Our recommendation: use the separate store page. Especially if you're participating in Next Fest to test market fit. You want that data. If you need to optimize your store page itself, our Steam page optimization guide covers the details.

Shared Cloud Saves

If your full game is already structured enough to support it, share cloud storage between the demo and the full game. This lets players pick up where they left off when they buy the complete version. It's a small touch that signals professionalism and removes one friction point from the purchase decision.

When to Release Your Demo

This is where most developers make the biggest mistake, and it's the hill that Chris Zukowski has been willing to die on for years: do not launch your demo on Day 1 of Next Fest.

The data is unambiguous. Games that entered the February 2025 Next Fest with fewer than 1,000 wishlists earned a median of 462 wishlists during the event. Not one of them reached Diamond tier. Games entering with 10,000 to 99,999 wishlists earned a median of 6,360.

Next Fest rewards momentum you've already built. It does not create momentum from nothing.

Release your demo weeks or months before the festival. Use smaller events and festivals to beta-test it. Get feedback, fix bugs, polish the experience. Release updates. Build wishlists organically. Then enter Next Fest as the grand finale of your demo marketing campaign, not the opening act.

Here's a timeline that works within the broader indie game marketing timeline:

3 to 6 months before Next Fest: Release your demo publicly
2 to 4 months before: Participate in smaller festivals (check howtomarketagame.com/festivals for options)
1 to 2 months before: Iterate on the demo based on player feedback
4 weeks before Next Fest: Submit your polished demo build for Valve's review
11 days before: Demo is live for Press Preview

If you can only manage a shorter window, release at minimum two weeks before the fest. This gives you time to catch critical bugs, generate some organic wishlists, and let early players spread the word.

The Press Preview Window

Valve shares the list of participating games with select press outlets 11 days before the fest begins. If your demo is live at that point, journalists can play it and start drafting coverage. If it isn't, you've missed this window entirely.

For the Press Preview to work, your demo needs to be publicly playable. There's no special press-only access. Whatever the public sees is what press sees. So releasing your demo before Press Preview also means real players will start finding it through Steam's normal discovery channels, which starts generating wishlists before the fest even begins.

The Cairn Approach: Demo Live for Months

The Cairn team kept their demo available for an extended period before their Next Fest. This allowed them to:

Collect months of player feedback and iterate on the experience
Build a community of players who'd already tried (and liked) the game
Enter Next Fest with significant existing wishlists and social proof
Have content creators make videos about the demo at their own pace, not squeezed into a single week

The tradeoff is that some players who try the demo months early might lose interest by the time you launch. But the data suggests this risk is outweighed by the benefits of entering Next Fest with momentum. You arrive at the fest's starting line already warmed up, while competitors are still stretching.

After Next Fest: Keep or Kill the Demo?

There's no universal right answer. Both approaches have defenders, and the decision depends on your specific situation.

Keep the demo if:

Launch is more than two months away
The demo is solid and generates steady wishlists passively
Content creators continue to find and cover it
You're planning to iterate on it further

Pull the demo if:

Launch is weeks away and you're concerned about players feeling "done"
The demo contains content that's changed significantly in the full game
Reviews on the separate demo page are negative and you need to regroup

Whatever you decide, communicate it clearly. Add a note on your demo's main menu about whether (and when) the demo will be deactivated. Many Next Fest players download demos during the event but don't play them until weeks later. Nothing frustrates a potential customer more than downloading something they can't launch.

If you keep the demo up, consider adding new content or updating it periodically. Shotgun Cop Man added extra content during the February 2025 Next Fest itself, which drove sustained engagement throughout the week. Post-fest, occasional demo updates give you reasons to re-engage your audience and trigger fresh content creator coverage.

Adding Analytics and Feedback Loops

Your demo is a testing lab. Treat it like one.

Crash reporting. Even basic logging to a local file helps. You need to know when your demo crashes and, ideally, where. During Next Fest, you can push hotfixes without re-review, but you can only fix what you can identify.

Player drop-off data. If you can track where players stop playing (which level, which section, which mechanic), you've got a roadmap for what needs improvement. Simple analytics like session length and completion rate tell you whether players are finishing the demo or bouncing early.

Feedback links. Put them on the main menu and the pause menu. Link to your Steam discussion forum, a Google Form, or your Discord. Valve's own tips page specifically recommends this. Make giving feedback require zero effort.

A sub-forum for demo feedback. Create one in your base game's Steam discussion forums. This gives players a dedicated space to report issues and discuss the demo, and it gives you a centralized place to monitor reactions.

The Demo Checklist

Before submitting your demo build:

Your demo is the most important piece of marketing you'll build before launch. More important than your trailer. More important than your social media presence. More important than your press kit (though you need that too). It's the thing that gets you into Next Fest, gets you in front of creators, and converts curious players into committed wishlisters.

Build it like it matters, because it does.

Free Tool: Launch Checklist — Track your demo and launch preparation with our interactive checklist. Runs in your browser, no signup required.

Frequently Asked Questions

Should I keep my demo live after Next Fest?

Keep it if launch is more than two months away and the demo generates steady wishlists. Pull it if launch is imminent and you're worried about players feeling "done," or if the demo contains outdated content. Whatever you decide, communicate clearly on the demo's main menu.

What if my demo gets negative reviews?

That's actually valuable feedback. Use the separate store page option so demo reviews don't affect your main game's rating. If reviews highlight real problems, fix them before launch. I've seen developers transform their games based on demo feedback.

How long before Next Fest should I release my demo?

Ideally 3-6 months before to participate in smaller festivals, collect feedback, and iterate. At minimum, release two weeks before to catch critical bugs and generate some organic wishlists before the fest begins.

Do I need separate analytics for my demo?

Basic logging helps identify crashes and drop-off points. Even simple data like session length and completion rate tells you whether players are finishing. Create a Steam discussion sub-forum specifically for demo feedback.

This article is part of our series on steam next fest. Start with the complete guide:

The Steam Next Fest guide

Also in this series:

Technical details verified against Valve's Steamworks demo documentation as of February 2026. Wishlist benchmarks from Chris Zukowski's surveys at howtomarketagame.com. For the full Next Fest timeline, see our Steam Next Fest guide.

How to Build a Demo That Sells Your Game