Entity Rankings
Access complete crowdsourced rankings and discover where millions of entities rank across entertainment, brands, sports, culture, and more. Built on Ranker's first-party audience voting data, Entity Rankings provides structured relationships between entities and the lists they're ranked on, supporting AI, content discovery, recommendation systems, and analytics.
Each record connects an entity to a specific Ranker list and its relative ranking position. The dataset includes complete ranked lists and reveals how entities are ranked across different topics, themes, and attributes, providing a multidimensional view of audience preferences. Supporting entity metadata enables identification, mapping, and integration with existing data systems.
Explore the Data
Ranker's Entity Rankings provides structured access to individual entity rankings across its catalog of editorially curated lists. Each ranking connects an entity to a list and its relative position, allowing businesses to analyze both complete ranked collections and the relationships between entities and the topics where they appear.
Complete List Rankings
Access ranked entities and their positions within individual lists, supporting the reconstruction and display of complete audience-ranked collections.
Entity-to-List Relationships
Explore which lists an entity is ranked on, connecting people, places, things, and other entities to the topics and questions where they appear.
Broad Topical Coverage
Access rankings spanning entertainment, brands, sports, culture, and other subjects, encompassing a diverse range of entities and topics.
Entity Metadata & Identifiers
Use persistent entity identifiers, names, and available descriptive metadata and external source IDs to support entity matching, enrichment, and integration.
Entity Ranking Example: Ted Lasso
Contextualized Human Preferences
Ranker has over 100 rankings on the television show Ted Lasso, including:
These rankings reveal more than popularity. They connect Ted Lasso to emotional tone, narrative themes, genre conventions, perceived writing quality, and platform preference—each within a distinct editorial context informed by audience voting.
→ Download Ted Lasso Rankings Sample
Why Ranker Data is Different
Ranker's Entity Rankings data captures how audiences rank and prioritize entities within specific topics. Its combination of explicit preference signals, comparative rankings, and curated topical context provides a distinctive source of human-preference relationships that complement other data sources.
Explicit Preference Data
Unlike passive behavioral signals or text-based associations, Ranker's rankings are informed by explicit upvotes and downvotes, capturing positive and negative audience preferences within the context of each list.
Comparative Rankings
Entity rankings reveal relative audience preferences within a defined topic, providing more than simple associations, mentions, or measures of popularity.
Contextualized Human Preferences
Rankings across editorially defined topics reveal multiple dimensions of audience preference, connecting entities to the qualities, themes, and attributes audiences associate with them—beyond overall popularity or semantic similarity.
A Complementary Data Source
Ranker's human-preference data complements behavioral, semantic, and other data sources, enriching AI, recommendation, search, and analytics applications with a distinct perspective on what audiences prefer.
Key Use Cases
| Use case | Benefit |
|---|---|
| Ranked Content & Editorial Applications | Build complete ranked lists, leaderboards, and interactive editorial experiences using structured crowdsourced rankings. |
| Search & Discovery | Help users discover entities through topical rankings and explore the lists on which an entity appears. |
| Recommendation Systems | Incorporate comparative audience-preference signals and entity-to-topic relationships into candidate discovery and ranking workflows. |
| AI & LLM Applications | Enrich AI systems with structured, human-ranked entity relationships and comparative preference signals that complement semantic and descriptive data. |
| Knowledge Graphs & Embeddings | Connect entities to ranked topics and collections, adding human-preference context to knowledge representations and embeddings. |
| Cultural & Market Research | Analyze how entities rank across different topics to explore audience preferences, topical associations, and patterns of interest. |
Data Dictionary
Below is the structural breakdown of the Entity Rankings dataset and its supporting list and entity metadata.
Entity Rankings
| Field Name | Type | Description |
|---|---|---|
list_id |
integer | Ranker's unique identifier for each list. This identifier is persistent. |
item_id |
integer | Ranker's unique identifier for each item (representing people, places, things, etc.). |
item_rank |
integer | The rank of the item appearing on a given list. |
item_votes |
integer | Total votes associated with the entity on the specific list |
item_up_votes |
integer | Total affirmative votes across on this entity the specific list |
item_down_votes |
integer | Total negative votes across on this entity the specific list |
List Metadata
| Field Name | Type | Description |
|---|---|---|
list_id |
integer | Ranker's unique identifier for each list. This identifier is persistent. |
list_name |
string | The name of the Ranker list. |
list_name_alt |
string | A shortened version of the list name |
list_primary_category |
string | The primary topical category of the list (e.g., Film, TV, Gaming, Brands). |
list_description |
string | A detailed description of the Ranker list, typically including the purpose and theme of the list. |
list_attributes |
JSON array | List attributes assigned by Ranker editors. |
list_url |
string | The direct URL linking to the Ranker list. |
list_tags |
JSON array | Tags associated with the list, providing additional context such as categorization and topic. |
list_collections |
JSON array | Collections that the list belongs to, if applicable. Collections are curated around central themes or topics, though not all lists are part of collections. |
list_tagged_items |
JSON array | Items associated with the list, typically representing people, places, or things. |
list_items |
integer | Number of ranked entities appearing on the list |
list_voters |
integer | Number of distinct users who voted on the list |
list_votes |
integer | Total voting events recorded on the list |
list_up_votes |
integer | Total affirmative votes recorded on the list |
list_down_votes |
integer | Total negative votes recorded on the list |
Entity Metadata
| Field Name | Type | Description |
|---|---|---|
item_id |
integer | Ranker's unique identifier for each item (representing people, places, things, etc.). |
item_name |
string | The primary name of an item, regardless of list appearance. |
item_primary_category |
string | The primary topical category of the item (e.g., Film, TV, Gaming, Brands). |
item_description |
string | A description of the item (in this case the title of the tv show or movie). This field is intended to assist with matching and mapping. This field is not available for all records. |
item_release_year |
integer | The year of release, generally the first air date for TV and year of theatrical release for movies. This field is intended to assist with matching and mapping. This field is not available for all records. |
item_class_properties |
JSON array | Properties associated with the item, such as cast. This field is intended to assist with matching and mapping. This field is not available for all records. |
item_source_ids |
JSON array | IDs associated with the item, such as IMDb, TMDB, Wiki, etc. This field is intended to assist with matching and mapping. This field is not available for all records. |
item_list_count |
integer | Total count of distinct canonical Ranker lists on which the entity is ranked. |
item_voters |
integer | Number of distinct users who voted on the entity |
item_votes |
integer | Total votes received by the entity across all lists |
item_up_votes |
integer | Total affirmative votes across all lists |
item_down_votes |
integer | Total negative votes across all lists |
Data Samples
Entity Rankings
→ Download Sample Data (CSV)
This data sample contains 3 lists. Contact us to request an expanded evaluation sample.
An annotated version of this data sample is available with entity names included for easier viewing: Download Sample Data – Annotated (Excel)
Sample Record (JSON)
This preview includes the top 3 ranked items from the list The 100+ Best Feel-Good TV Shows. Note: this preview is truncated to 3 (of over 100 items) for space considerations on this page.
List Metadata
→ Download Sample Data (CSV)
This data sample contains 200 lists. Contact us to request an expanded evaluation sample.
Sample Record (JSON)
This preview includes the list metadata for a single list: The 100+ Best Feel-Good TV Shows.
Entity Metadata
→ Download Sample Data (CSV)
This data sample contains 1000 entities. Contact us to request an expanded evaluation sample.
Sample Record (JSON)
This preview includes the metadata for a single entity, the TV series Ted Lasso.
Delivery & Formats
Delivery Methods
REST API, Amazon S3, Google Cloud Storage, Secure FTP
File Formats
CSV, TSV, JSON, JSONL, NDJSON, Parquet, Avro
Update Frequency
Daily, weekly, and monthly snapshots are available
Ranker data products support clients with a range of business, analytics, and AI workflows. Data can be integrated into existing data environments, analytical pipelines, and AI applications, and used to enrich proprietary models, recommendation systems, business intelligence, knowledge graphs, and other data-driven applications.
Contact Ranker to discuss delivery options, formats, update frequency, and licensing arrangements for this product.
Data Provenance & Privacy
Ranker’s data products are built on first-party audience preference signals collected directly through Ranker’s platforms. Ranker employs active bot mitigation, voting bias detection, and data quality controls to help maintain the integrity and reliability of its data.
Ranker is committed to user privacy and compliance with applicable data privacy laws and regulations.
Frequently Asked Questions
Common questions about Ranker’s B2B data licensing datasets, schema coverage, and delivery options.
[faq]
Frequently Asked Questions
Common questions about Ranker’s Item Rankings dataset, vote schema, and delivery options.
DATA ARCHITECTURE & COVERAGE
What is included in Ranker’s Item Rankings dataset?
Item Rankings maps exact ordinal rank positions (item_rank) and item-level vote tallies (item_votes, item_up_votes, item_down_votes) for every entity appearing across 233,000+ ranked lists, backed by 1.5B+ explicit human votes.
How are item rankings determined on Ranker?
Rankings reflect dynamic, crowdsourced consensus driven by explicit UP_VOTE and DOWN_VOTE judgments evaluated against each list’s specific list_criteria—capturing comparative preference rather than static editorial picks or uncalibrated star-rating averages.
How does Item Rankings join with list and entity metadata?
Each record pairs list_id (joining to Lists & Collections for list titles, descriptions, tags, and list_criteria) with item_id (joining to Entities for canonical names, ontology attributes, and IMDb, TMDB, and Wikipedia IDs).
AI, ML & PRODUCT USE CASES
How do ML teams use Item Rankings for RLHF, DPO, and Learning-to-Rank (LTR)?
Because Item Rankings captures both strict ordinal hierarchy (item_rank) and bidirectional sentiment (item_up_votes and item_down_votes) within a defined topical context (list_criteria), ML teams can generate high-density pairwise and listwise preference datasets for Direct Preference Optimization (DPO), reward model alignment, and Learning-to-Rank pipelines without costly manual annotation.
How does Item Rankings support model evaluation and product discovery?
AI teams use Ranker’s ordinal consensus as ground-truth benchmarks to evaluate model outputs in subjective cultural domains, while product and editorial teams use rank positions and vote totals to power popularity priors, ranked carousels, and consensus badges.
PROVENANCE, INTEROPERABILITY & DELIVERY
How does Ranker protect ranking integrity against bot noise and social bias?
All votes are 100% first-party and audited with automated anomaly detection to filter suspicious voting patterns. Because voting on Ranker is anonymous and topically framed, the dataset minimizes the performative social confirmation bias common in reviews and social media.
What file formats, delivery methods, and update cadences are supported?
Datasets are delivered in CSV, TSV, JSON, JSONL, NDJSON, Parquet, or Avro via automated cloud storage drops (AWS S3, Google Cloud Storage) as well as low-latency APIs. Delivery cadences support historical snapshots alongside daily, weekly, or monthly updates.