How the Steam Algorithm Works

This is exactly how Steam’s algorithm works with timing, trending, and visibility windows for:

Coming Soon Launch
Full Release
Early Access
Launch

I’ll explain what matters, when it matters, and how long you have before Steam buries your game.

1. Coming Soon

How long you need to trend
Best timing:
2–6 months before launch.

Why? That’s the period when Steam’s algorithm actually cares about wishlist velocity.

What Steam tracks during “Coming Soon”:
Wishlist velocity (wishlists per day)
Page activity (people scrolling screenshots, watching trailer)
CTR Demo installs (if you have a demo)
Follower increases
Dev posts
Traffic source quality

How long you can trend:

You typically have 48–72 hours in “New & Trending (Coming Soon)” when the page first goes live if your CTR is good.

If CTR is weak → you get almost zero time.

Ideal schedule:

Page goes live → send traffic for 2–3 days to push into “Trending Coming Soon”.
Keep building steady wishlists for months, not days.

Push harder again 1 month before launch.

2. Launch Window (Full Release or EA Release) This is the holy moment for the algorithm.

Steam gives you:

A short testing window

A boost if you perform well

A penalty if you underperform

How long do you appear in “New & Trending” at launch?

This depends entirely on performance:

Strong launch → 48–72 hours Very strong → up to 5–7 days Weak launch → 6–12 hours… sometimes even less

Steam literally tests you hour by hour.

What matters most in the first 48 hours:

Wishlist → purchases conversion CTR from discovery traffic

Review velocity (how many reviews per hour)
Demo conversions (if demo exists)
Playtime retention
Refund rate
Price perception (is your price right for this genre)

Golden rule:

Your entire launch success is determined by the first 48 hours.

3. Early Access Launch algorithm rules

Early Access launch is treated exactly like a full launch, but with some differences:

Steam evaluates:

Wishlist conversion
Review velocity
First-week playtime
Retention Refunds
Page CTR
Trailer playthrough rate

Time in New & Trending (EA):

Same as full launch:

Strong EA launch → ~48–72 hours Average → 24 hours Weak → <12 hours

Good EA timing:

Tuesday–Thursday
Avoid holidays
Avoid big releases
Avoid Steam Next Fest overlap unless you have a great demo

4. How long do you have to “scale up” before Steam buries the page?

You MUST hit these milestones fast:

Coming Soon:
Steam expects 200–1000 wishlists per week to consider you healthy. If you go <50 per week → your visibility drops sharply.

Launch Window:

Within the first 48 hours, you need: ~3–8% wishlist conversion

At least 10+ reviews within 24 hours

Solid CTR (2%+) Low refund rate (<10%)

Good retention If you fail these → you drop off trending instantly.

5. Steam Algorithm Key Variables (simple list)

Steam prioritizes:

CTR Conversion (wishlist → buy)
Review velocity
Playtime retention
Refund rate
Genre match to audience

Page activity (scroll depth, trailer watched)

Wishlist velocity (not total count)

Total wishlists matter only for your launch spike, not for trending visibility.

6. How to hit Trending during Coming Soon

Steam needs:

First 3 days

Big traffic push High CTR (>2%)

Wishlist velocity (100+ per day)

Months after:

Consistent wishlist flow

Community posts

Trailer updates

Devlogs Demo (if possible)

7. How long should a game stay in “Coming Soon” before launch? 2–6 months is optimal.

1 year+ is bad unless you are a top-tier studio.

Too long → the algorithm “gets bored” and your velocity dies.

notes:

Coming Soon: Trend for 2–3 days first

Build wishlists for months

Push again 1 month before launch Launch (EA or Full):

First 48 hours decide everything

Good numbers → you stay in Trending

2–4 days Bad numbers → you disappear in hours

Steam algorithm tracks:

CTR Wishlist conversion
Reviews/hour
Refund rate
Player retention
Genre alignment

If you fail to scale fast, Steam hides your page.

What is CTR in the Steam algorithm?

CTR = Click-Through Rate.

On Steam, CTR is the percentage of people who see your game and click on it to open the store page.

Formula:

CTR = (Clicks on your game page / Impressions of your capsule) × 100

Why does CTR matter on Steam?

CTR is one of the strongest signals Steam uses to decide:

how much to promote your game

whether to show it on the home page

whether to push it in the Discovery Queue

whether it appears in “More Like This”

how much exposure you get during sales and events

High CTR = Steam pushes your game more.

Low CTR = Steam hides your game.

What influences CTR?

Capsule art (thumbnail).

The most important element. Also there are MULTIPLE other factors

If it's unclear, messy, or the logo is unreadable → CTR drops.

Game title visibility.

If the name is hard to read or doesn’t signal genre → fewer clicks.

Short Description.

If it's vague or generic , players skip it.

Trailer thumbnail.

This image also strongly affects click rate.

Tags

If Steam shows you to the wrong audience → low CTR.

Audience match

Steam tests your page on small sample groups.

Low CTR in early test = reduced exposure.

What is a “good” CTR on Steam?

Ranges vary by genre, but roughly:

<0.5% = very poor

0.5% – 1.0% = poor

1.0% – 2.0% = average

2.0% – 4.0% = good

4%+ = excellent (viral potential)

CTR = how many people click your game when they see it.

High CTR = more exposure.

Low CTR = Steam suppresses your visibility.

“Improve store conversion” means turning more visitors into wishlisters, followers, demo players, or buyers.

On Steam, conversion = what percentage of people who visit your page actually take action.

✅ Simple definition

Store conversion =

(People who interact meaningfully with your store page , people who visit your store page) × 100

Meaningful interactions include:

Wishlist

Follow

Play demo

Buy (after launch)

Add to cart

Improving store conversion = making more of the visitors take one of these actions.

Steam’s algorithm loves two things:

CTR (people clicking your page)

Conversion (people taking action once they’re on your page)

High conversion tells Steam:

“Players who land here want this game.”

Result:

You get more organic traffic

You appear more in Discovery Queue

You show up more in Similar Games

You get more front-page impressions

You go viral much easier

Low conversion tells Steam:

“People don’t want this.”

Steam then shows your page less → fewer impressions → fewer clicks → a dead spiral.

🎯 What improves store conversion (practical examples)

These are the biggest levers:

1. A strong, clean short description

Clear, niche, and focused.

If it’s vague → conversion dies.

2. Clear “Wishlist Now” CTA

People convert when they’re told what to do.

3. High-quality screenshots

Steam users scroll → screenshots sell the game.

If the first three images are weak → wishlist rate tanks.

4. A trailer that shows gameplay in the first 5 seconds

If there is slow intro → many bounce.

5. Accurate tags

If people enter expecting X but they get Y → instant bounce → conversion drops.

6. UI language support

If a big region sees “English Only”, they lose interest.

7. Unique hook visible at the top

Players must immediately understand your “special thing”.

📈 What a “good” store conversion looks like

Numbers vary by genre, but roughly:

Wishlist conversion:

3–5% = okay

5–10% = strong

10%+ = excellent

Demo conversion:

1–3% = okay

3–6% = good

6%+ = strong

After release:

Purchase conversion:

1–2% = okay

2–4% = strong

4%+ = excellent

Improving store conversion =

making more visitors say “YES, I want this game”.

It’s one of the most important parts of the Steam algorithm as important as CTR.

Improve this:

new translations needs to be added after you are changing the description, don't forget to localize, the title in every translated page for the images too.

Steam DOES allow localized images — but only for capsule art

Steam automatically supports localization for:

✔ Library Capsule

✔ Header Capsule (616x353)

✔ Small Capsule

✔ Main Capsule (portrait)

You can upload a localized version for every language through Steamworks.

Path:

Steamworks → Your Game → Store Presence → Edit Store Page → Graphical Assets → Add Localization

You will see a list like:

English

French

German

Japanese

Chinese

…etc.

You can upload different capsule art for each language.

BUT:

Steam does NOT allow language-specific screenshots or trailers.

Those are global.

2. For screenshots, feature images, infographics, icons, etc. you must upload ONLY ONE VERSION

Steam does not let you upload localized screenshots.

If you need text in screenshots (bad practice), you have two options:

OPTION A

Remove text entirely

→ Best practice

→ Screenshots should NOT contain language text anyway

→ Use in-engine UI instead of baked text

OPTION B — Use in-game language settings

This is how Nintendo, PlayStation, Ubisoft, etc. do it.

Steps:

Change the game language to each target language

Capture in-game screenshots with localized UI

Upload those screenshots

Arrange them so they make sense together

Steam cannot assign screenshots per language, but players will understand when UI changes between images.

3. If you need localized infographics or feature banners for your page, use Steam’s “Store Page Artwork” section

There is a place called:

“Large Capsule” / “Community Header” / “Upload Additional Artwork”

These ALSO cannot be localized.

Steam forces a single version.

Solution:

Make them text-light

Or use universal icons instead of actual text

Examples that work worldwide:

:x: “Story Mode

Cute Adventure Platformer”

✔ Icon: :books: :video_game: :feet:

:x: “Collect 200+ items”

✔ Icon: :gift: :heavy_plus_sign:200

4. For the About section (text block), Steam auto-detects languages

This is where you should focus all localization efforts.

Every language gets:

Title

Short Description

Long Description

Updates / announcements

System Requirements

And THIS is where conversions skyrocket.

Steam auto-switches language based on user region.

5. Workflow used by publishers to localize image assets

Here’s the method:

Design the English capsule first.

Export layered PSD/AI file.

Duplicate it for each language.

Replace only the text elements (never reposition art unless required).

Generate final PNG/WebP per language.

Upload each to Steamworks → Graphical Assets localization.

Run A/B tests (different headlines for CN, JP, KR, DE).

6. SPECIAL NOTE:

simplified Chinese, Japanese, Korean capsules

These markets convert 50–200% higher when the capsule is localized.

Examples:

Chinese (CN):

shorter text, bold color palette

Japanese (JP):

cleaner composition, small text

Korean (KR): gamer humor + dynamic fonts

If you localize only one capsule language, pick:

:flag_cn: Chinese Simplified

:flag_jp: Japanese

:flag_kr: Korean

7. How to test localized images

Before uploading:

Post them in Discord and ask players

Use X polls

Even YouTube Community tab

Compare CTR after upload (Steamworks → Traffic → Clickthrough Rate)

Localized capsules often boost CTR by:

+20% to +80% in non-English regions
Steam page “first 5 seconds” test

Most devs think Steam optimization = tags + description.

Not really. It is important, but for this i'm writing an article how to optimize more from steamworks.

The ONLY thing that matters is what players see in 5 seconds:

Capsule

First 3 screenshots

Trailer thumbnail

Reviews at a glance

“About this game” first line

If these 5 elements don't instantly trigger curiosity → they close the page.

Tip to devs:

Ask 10 random people: “Do you understand what this game is in 5 seconds?”

If they hesitate means that your page is broken.

Some of the best #indiegames on Steam never had a chance.

Not because of gameplay.

Because their first 5 seconds on the store page :

• Wrong screenshot first

• Trailer hides gameplay

• Capsule tells you nothing

• Boring About section intro

I’m reviewing Steam pages free when i got some time .

If you need more insights or growing with me, let me add your game in my portfolio , complete the form on my website and let's talk.

If you want eyes on yours, here’s where:

🔗 form