Backloggd Game Reviews & Database Scraper

Scrape Backloggd game reviews, ratings, metadata, and community statistics from title searches or public game URLs. Export clean datasets for market research, gaming analytics, recommendation systems, journalism, RAG, and player-feedback analysis.

Optional AI review intelligence summarizes recurring praise, complaints, bugs, performance issues, platform feedback, player sentiment, and spoiler-free consensus. No Backloggd account or login is required.

What this Backloggd scraper returns

• Ranked game matches from title searches
• Structured game metadata, platforms, genres, companies, release details, and image URLs
• Backloggd ratings, review counts, and community activity
• One clean Dataset row per public written review
• Reviewer username, rating, date, likes, platform, play status, and spoiler flag when public
• Deterministic review statistics without AI
• Optional AI-powered, spoiler-safe game review summaries
• Client-safe error and partial-result rows when one source request fails

Backloggd scraper use cases

• Compare player sentiment, ratings, and recurring complaints across games.
• Enrich game catalogs with platforms, genres, companies, release details, and cover URLs.
• Build recommendation, search, RAG, or gaming analytics datasets.
• Monitor public review themes for publishers, developers, journalists, and market researchers.
• Create spoiler-aware summaries of praise, bugs, performance issues, and platform feedback.

Quick start

The easiest input is a game-title search:

{
  "searchQueries": ["Elden Ring"],
  "maxReviewsPerGame": 100,
  "outputMode": "games_and_reviews"
}

You can also paste a direct Backloggd game URL:

{
  "gameUrls": ["https://backloggd.com/games/elden-ring/"],
  "maxReviewsPerGame": 500,
  "outputMode": "reviews_only"
}

Results are streamed to the run's default Apify Dataset. Each row contains itemType, which identifies whether it is a game, review, summary, search match, status, or error.

Input

Provide at least one game title, public game URL, slug, or numeric ID. You can combine input methods; duplicate games are scraped once.

Choose games

Input	Default	What to enter
searchQueries	[]	Game titles such as Elden Ring, Hades, or Baldur's Gate 3. Each query resolves up to five ranked matches.
gameUrls	[]	Public game URLs such as https://backloggd.com/games/elden-ring/.
gameIdsOrSlugs	[]	Known URL slugs such as elden-ring. Numeric IDs are accepted, but URLs and slugs are more reliable.

Title searches intentionally preserve several close matches instead of pretending that the first result is always correct. Search provenance and ranking are included in full output. Equivalent URLs, tracking parameters, and duplicate slugs are normalized within the run.

Choose the output

Input	Default	Meaning
maxReviewsPerGame	100	Maximum accepted public written reviews per resolved game, from 0 to 5,000. Use 0 for metadata-only runs.
outputMode	games_and_reviews	Selects which result row types are written. See the table below.
spoilerHandling	exclude_from_ai	Controls whether spoiler-marked reviews are kept, analyzed, or excluded.

Optional AI review summaries

Input	Default	Meaning
includeOpenRouterAnalysis	false	Generates qualitative, spoiler-safe insights in game summary rows.
maxReviewsToAnalyzePerGame	50	Maximum reviews sent to AI for each game. Raw scraping can continue beyond this number.

Clients do not enter an OpenRouter API key or choose a model. The Actor uses an encrypted operator-managed OPENROUTER_API_KEY environment secret and a built-in low-cost structured-output model. The key is never written to inputs, logs, or Dataset rows.

To receive an AI summary, enable includeOpenRouterAnalysis and select an output mode that includes summaries: summaries_only, reviews_and_summaries, or full.

If AI is disabled or unavailable, summary modes still return deterministic rating and review statistics. A missing or failed AI service does not discard successfully scraped reviews.

Output modes

Dataset content option	Rows normally returned
games_only	game
reviews_only	review
games_and_reviews	game, review
summaries_only	game_review_summary
reviews_and_summaries	review, game_review_summary
full	search_result for title searches, plus game, review, and game_review_summary

error and source_status rows can appear in every mode. They preserve clear diagnostics without stopping other valid games.

Apify Console provides focused column views for results, written reviews, game metadata, review summaries, and run messages. These are views of the same Dataset; use itemType when filtering JSON, CSV, Excel, or API exports.

Game metadata fields

A game row can include:

• gameId, gameSlug, gameUrl, canonicalUrl, and title
• alternativeTitles, description, releaseDate, releaseDates, and releaseStatus
• companies, developers, publishers, platforms, genres, and themes
• gameModes, franchises, series, and ageRatings
• coverImageUrl, backgroundImageUrl, and lightweight screenshot URLs
• averageRating, ratingCount, and reviewCount
• playedCount, playingCount, backlogCount, and wishlistCount
• completion, abandoned, and completionist counts when public
• communityStats, related games, editions, DLC, and public external links when available
• search queries that found the game and extractionMethod

Backloggd does not consistently label company roles. companies is the conservative complete list; developers and publishers are populated only when a role can be inferred reliably. Missing source fields remain null or empty rather than being guessed. Images are returned as URLs and are not downloaded.

Illustrative game row:

{
  "itemType": "game",
  "source": "backloggd",
  "gameId": "119133",
  "gameSlug": "elden-ring",
  "gameUrl": "https://backloggd.com/games/elden-ring/",
  "canonicalUrl": "https://backloggd.com/games/elden-ring/",
  "title": "Elden Ring",
  "releaseDate": "2022-02-25",
  "companies": ["FromSoftware", "Bandai Namco Entertainment"],
  "developers": ["FromSoftware"],
  "platforms": ["PC", "PlayStation 5", "Xbox Series X|S"],
  "genres": ["Adventure", "RPG"],
  "averageRating": 4.5,
  "ratingCount": 132000,
  "reviewCount": 21000,
  "playedCount": 179000,
  "backlogCount": 40000,
  "coverImageUrl": "https://images.igdb.com/example-cover.jpg",
  "searchQueries": ["Elden Ring"],
  "extractionMethod": "http",
  "scrapedAt": "2026-07-17T12:00:00.000Z"
}

The values are illustrative; live data depends on the public Backloggd page at run time.

Written review fields

Each accepted public written review becomes one review row. Available fields include:

• game ID, slug, title, and canonical game URL
• native reviewId when available, stable fallback identity, and reviewHash
• reviewRating, reviewTitle, reviewText, reviewDate, and reviewUpdatedAt
• public reviewer username, display name, profile URL, and avatar URL
• likes, comments, platform, play status, hours played, and completion status when visible
• recommendation, tags, isSpoiler, and spoiler warning text when visible
• extraction method and scrape timestamp

Illustrative review row:

{
  "itemType": "review",
  "source": "backloggd",
  "gameId": "119133",
  "gameSlug": "elden-ring",
  "gameTitle": "Elden Ring",
  "gameUrl": "https://backloggd.com/games/elden-ring/",
  "reviewId": "337912",
  "reviewUrl": "https://backloggd.com/u/example-user/review/337912/",
  "reviewHash": "9f86d081884c7d659a2feaa0c55ad015a3bf4f1b2b0b822cd15d6c15b0f00a08",
  "reviewRating": 4.5,
  "reviewText": "Example public review text.",
  "reviewDate": "2026-07-12T18:45:00.000Z",
  "reviewerUsername": "example-user",
  "reviewerProfileUrl": "https://backloggd.com/u/example-user/",
  "likes": 12,
  "commentsCount": 2,
  "platform": "PC",
  "playStatus": "Completed",
  "isSpoiler": false,
  "extractionMethod": "http",
  "scrapedAt": "2026-07-17T12:00:00.000Z"
}

The Actor paginates until it reaches the requested limit, the public feed ends, a page produces no new review identities, a cursor repeats, or retries are exhausted. Reviews are deduplicated within the current run. If a later page fails, already written rows remain available and a status row explains the partial result.

Review summaries and AI insights

Every game_review_summary contains useful deterministic metrics:

• reviews scraped and rating distribution
• average rating within the scraped sample
• rating-derived positive, neutral, negative, and unrated counts
• spoiler-review count
• most-liked review references
• platforms represented in the sample
• earliest and latest review dates

When optional AI analysis succeeds, the same row also includes recurring praise and complaints, performance and bug themes, platform signals, gameplay/story/difficulty/value consensus, player-fit guidance, developer opportunities, and a spoiler-free consensus.

Check analysisMethod before interpreting a summary:

• deterministic means statistics only, with sentiment buckets derived from ratings.
• openrouter means the qualitative fields were generated from sampled reviewer opinions.
• deterministic_fallback means AI was requested but unavailable or unsuccessful.

Illustrative AI summary row:

{
  "itemType": "game_review_summary",
  "source": "backloggd",
  "gameId": "119133",
  "gameSlug": "elden-ring",
  "gameTitle": "Elden Ring",
  "gameUrl": "https://backloggd.com/games/elden-ring/",
  "reviewsScraped": 100,
  "reviewsAnalyzed": 50,
  "averageScrapedRating": 4.22,
  "sentimentBreakdown": {
    "positive": 39,
    "neutral": 5,
    "negative": 3,
    "mixed": 3,
    "unknown": 0
  },
  "topPraise": [
    {
      "topic": "exploration",
      "count": 18,
      "summary": "Reviewers commonly praise exploration and environmental discovery."
    }
  ],
  "topComplaints": [
    {
      "topic": "PC performance",
      "count": 7,
      "severity": "high",
      "summary": "Some PC-focused reviews report stuttering or inconsistent performance."
    }
  ],
  "spoilerFreeConsensus": "Reviewers commonly praise exploration and combat variety, while some PC players report performance concerns.",
  "isAiGenerated": true,
  "analysisMethod": "openrouter",
  "generatedAt": "2026-07-17T12:05:00.000Z"
}

AI summaries describe sampled reviewer opinions, not objective facts. They can contain classification or model errors and should be reviewed before publication or high-stakes use.

Spoiler handling

Setting	Raw review rows	AI analysis
include	Kept with isSpoiler: true	May be analyzed; summary prose remains spoiler-safe by design
exclude_from_ai	Kept and clearly marked	Review text is not sent to AI or included in spoiler-free consensus
exclude_completely	Not written to the Dataset	Not sent to AI

Spoiler handling relies on Backloggd's public marker. Raw spoiler content remains visible whenever the selected policy keeps the review.

High-volume scraping and cost control

The Actor uses production-tested pacing and concurrency automatically. For inexpensive bulk extraction:

• use reviews_only when you do not need game or summary rows;
• leave AI disabled during the initial bulk collection;
• run a smaller AI summary job only for games that need qualitative analysis;
• keep maxReviewsToAnalyzePerGame lower than maxReviewsPerGame;
• start with a representative sample before running thousands of reviews;
• export JSONL for streaming pipelines.

The published default is 1,024 MB RAM with a 60-minute timeout. HTTP-only tests succeeded at 512 MB, including 5,000 reviews on the current build, while the 1 GB default keeps headroom for source variation and browser fallback. A prior conservative cloud run completed 50 games and 50,000 unique reviews.

Actual duration and cost depend on game count, available reviews, source latency, retries, Dataset writes, and AI usage.

Pay-per-event pricing

The Actor charges for completed, durable output:

• review_extracted for each unique review successfully written to the Dataset;
• ai_review_analyzed for each review that receives a valid accepted AI classification;
• Apify's platform-managed start event when applicable.

Failed requests, retries, duplicate reviews, errors, search matches, metadata rows, summaries, and deterministic fallbacks are not charged as custom extraction events. The Actor respects the run's Apify spending limit. OpenRouter usage is funded by the Actor operator's encrypted environment key and is controlled with the AI review limit.

Run through the Apify API

Install the Python client with pip install apify-client:

from apify_client import ApifyClient

client = ApifyClient("YOUR_APIFY_API_TOKEN")

run = client.actor("thescrapelab/backloggd-game-reviews-scraper").call(
    run_input={
        "searchQueries": ["Elden Ring", "Hades"],
        "maxReviewsPerGame": 100,
        "outputMode": "games_and_reviews",
        "includeOpenRouterAnalysis": False,
    }
)

if run is None:
    raise RuntimeError("Actor run failed")

dataset = client.dataset(run["defaultDatasetId"])
for item in dataset.iterate_items():
    print(item["itemType"], item.get("gameTitle") or item.get("title"))

The Dataset can also be exported as JSON, JSONL, CSV, Excel, XML, or HTML from Apify Console or the Dataset API.

Troubleshooting

No games were returned

• Enter at least one title, public game URL, slug, or ID.
• Use a canonical game URL or slug when a title is ambiguous.
• Prefer URLs or slugs over numeric IDs.
• Check error rows for INVALID_INPUT, UNSUPPORTED_URL, or GAME_NOT_FOUND.

No reviews were returned

• The game may have ratings but no public written reviews.
• Confirm maxReviewsPerGame is greater than zero.
• Select an output mode that includes reviews.
• Check source_status rows for no_reviews or partial_reviews.
• With exclude_completely, spoiler-marked reviews are not counted as emitted rows.

AI summary is missing

• Enable includeOpenRouterAnalysis.
• Select summaries_only, reviews_and_summaries, or full.
• The Actor operator must have a valid funded OPENROUTER_API_KEY environment secret.
• Check for an OPENROUTER_FAILED row. Deterministic summaries remain available when possible.

A run is slow or partial

• Retry with fewer games and reviews to isolate the affected source page.
• Review source_status and error rows for rate-limit or pagination details.
• Browser fallback is limited to eligible public game-page failures; it does not bypass CAPTCHAs or access controls.

Limitations

• Only public Backloggd data is extracted.
• The Actor is not affiliated with, endorsed by, or operated by Backloggd.
• No Backloggd login is used, and private profiles, private lists, login-required content, and access-controlled routes are not accessed.
• Review availability depends on the game. Displayed rating or review counts can differ from currently retrievable written review cards.
• Some fields are not public for every game or review and remain null, empty, or omitted.
• Backloggd community counters may be abbreviated or approximate.
• Company roles are not consistently labeled; prefer companies when exact role attribution matters.
• Reviews are collected in the public recent-first order. Review sorting is not exposed as a client input.
• Pagination can shift while new reviews are posted. Deduplication is current-run only, not a frozen source snapshot.
• Numeric game ID resolution is best effort; direct URLs and slugs are more reliable.
• Individual review pages are not visited solely to enrich optional fields.
• Backloggd can change its HTML, public feeds, route policy, or blocking behavior.
• AI analysis is optional and AI summaries may contain model errors.
• Spoiler controls follow the selected input and Backloggd's public spoiler markers.
• Current-run results are stored in the default Apify Dataset. The Actor does not maintain review history between runs.

Compliance and stateless behavior

The Actor uses public game pages, public search results, and public review-feed content without authentication. It does not bypass CAPTCHAs, paywalls, authentication, or access controls; access private APIs; or scrape private user collections.

Each run starts from its supplied input and uses only current-run default Apify storage. The Actor does not create a named persistent store, compare with earlier runs, retain review history, or monitor new reviews.

You are responsible for using exported data in accordance with Backloggd's current terms, robots policy, applicable law, and your organization's privacy and retention requirements. Public usernames and profile URLs should still be handled responsibly.

Support

For a useful support report, include:

• the Apify run ID;
• a small sanitized input example;
• affected public game URLs or search queries;
• expected and actual row types or counts;
• relevant log messages and timestamps.

Never post an Apify token, OpenRouter key, proxy credential, cookie, or private configuration.