Why Isn't My Roblox Game Getting Recommended? The Official Signals Behind "Recommended for You"

You published your Roblox experience, told your friends, maybe even ran a few ads — and the impressions from Home are still close to zero. Before you conclude the algorithm hates you, it helps to know what Roblox's own.

Why Isn't My Roblox Game Getting Recommended?

You published your Roblox experience, told your friends, maybe even ran a few ads — and the impressions from Home are still close to zero. Before you conclude the algorithm hates you, it helps to know what Roblox's own documentation says the "Recommended for You" system actually measures. Most stalled games are stalled for one of three reasons: weak engagement signals, metadata that suppresses discovery, or a configuration setting that quietly shrank the eligible audience before the algorithm ever ran.

This guide walks through each, based on Roblox's official discovery and experience-configuration documentation.

First, understand the two-stage system

According to Roblox's discovery documentation, Home recommendations work in two stages: a retrieval stage that selects a subset of candidate experiences each user might enjoy, and a ranking stage that orders those candidates for each individual user in a personalized way.

That distinction matters for troubleshooting, because "my game isn't recommended" is really two separate problems:

1. Not being retrieved at all — your game never enters the candidate pool. This is usually a metadata, eligibility, or configuration problem.
2. Being retrieved but ranked low — your game enters the pool but loses to other candidates. This is an engagement-signal problem.

Fix them in that order. No amount of engagement optimization helps a game that isn't eligible to be surfaced in the first place.

The engagement signals that drive ranking

Roblox's documentation lists the most important signals the ranking stage uses:

• Play-through rate — of the users who saw your tile, how many actually joined and played.
• First play bounce rate — how many players leave almost immediately after joining.
• Play days per user — how many distinct days a player comes back.
• Playtime per user — how long players play, capped at 60 minutes per day in the measurement.

There's a crucial detail here: per the documentation, each signal is based on users who organically joined from the Recommended for You sort on Home. Roblox doesn't count the engagement, monetization, or retention of users first acquired from ads, curation, friends, search, social media, or any other source. Practically, that means you can't advertise your way into recommendations — a paid-traffic spike that bounces quickly tells the system nothing, because the organic Home cohort is what's being graded.

If your game is getting some Home impressions but not growing, these four numbers are where to look. A high bounce rate in particular points to a first-minute problem: slow loading, a confusing spawn area, or a store page that promised something the game doesn't deliver.

Metadata that actively hurts you

The discovery documentation is explicit that some metadata choices are penalized rather than just unhelpful:

• Monetary bait in titles — titles that lead with monetary implications, such as giveaway-style hooks, receive less exposure.
• Metadata that doesn't match actual gameplay — experiences whose metadata and content are highly mismatched are not recommended.

The fix is boring but effective: write an accurate description of what players do, and use original imagery and naming from your actual game rather than borrowed or misleading thumbnails. Mismatched metadata also feeds back into the engagement signals above — players lured by an inaccurate tile bounce fast, and bounce rate is one of the numbers being measured.

Write your description for semantic search

Roblox search supports semantic search for all of the platform's officially supported languages. That means search matches intent, not just exact keywords — players can find games through natural-language queries such as "food games" or "avatar editors".

The practical takeaway: describe your game in plain, natural language that covers what players do, what genre it fits, and what makes it distinct. Keyword-stuffing is not just unnecessary under semantic search — it works against the accurate-metadata guidance above.

Configuration settings that shrink your reach before the algorithm runs

This is the category most creators forget to check, and it lives in the Creator Dashboard rather than in your game design. Per Roblox's experience-configuration documentation, you can set restrictions so that only players over a certain age or from specific regions can play and discover your game:

• Age restrictions. Setting a minimum age determines the youngest players who can access your game, and your answers to the Maturity & Compliance Questionnaire govern which age groups see it on the Home and Discover pages. A restrictive setting shrinks the pool of users you can be recommended to.
• Geographic region access. Region settings control which countries and regions can access your experience. If a region isn't enabled, users there are out of your audience entirely.

Audit these before touching anything else. They gate the retrieval stage: an ineligible audience member can never generate an impression, no matter how good your engagement metrics are among the players you do have.

How to monitor progress

Once your configuration is clean and your metadata is honest, use Roblox's analytics dashboards to track the engagement signals over time rather than refreshing your visit count. Watch play-through rate and bounce rate first — they respond fastest to store-page and first-minute changes — then play days per user and playtime as measures of whether the game itself retains people.

Also remember that Home recommendations are one surface among several. Roblox's documentation describes additional discovery surfaces, including Search, the Discover page charts, Standout Games, and notifications. A game that isn't winning on Home yet can still build an audience through search (which is why the semantic-description work matters) and through re-engaging existing players.

A troubleshooting checklist

1. Check age settings and region access in the Creator Dashboard — make sure you haven't excluded most of your potential audience.
2. Rewrite your title and description to accurately describe gameplay in natural language; remove anything resembling monetary bait.
3. Replace any borrowed or misleading imagery with original screenshots and art.
4. Fix the first minute of your game to cut bounce rate: fast load, clear objective, immediate interaction.
5. Track play-through rate, bounce rate, play days per user, and playtime in analytics, and judge changes by the organic Home cohort — not ad traffic.

The recommendation system isn't a lottery. It's a measurement pipeline, and every input it measures is something you can inspect and improve.