About Reelgood
Streaming Content Intelligence
Built on 10 Years of Data.
Reelgood is the source of truth for streaming title availability and metadata. We track 300+ services across 25+ countries, delivering ML-verified data that updates in real time. The companies powering the streaming experience rely on us to get it right.
David Sanderson, CEO: The Reelgood Story (3 min)
Our Story
We Didn't Plan to Be a Data Company. The Data Forced Our Hand.
Reelgood was founded in 2015 to solve a straightforward consumer problem: streaming fragmentation. Figuring out what to watch every night required flipping between numerous streaming apps, not to mention figuring out where any title was currently available.
But the app would only be as good as its underlying data. We tried using data from all the existing players, including Gracenote, the industry standard at the time. But the data quality killed us. The data did not match reality. Titles listed as available on Netflix had already left. Major new releases would not appear in any feed for weeks after launch. Duplicates were everywhere. Users would click play, find the title was not there, and leave. No feature we could build would outweigh a broken experience caused by bad data.
So we built our own dataset.
Over 8 years, we developed a proprietary ML matching system trained on feedback from more than 100 million consumer app users. It operates independently of third-party IDs, using metadata signals and confidence scoring across cast, crew, runtime, synopsis, and release data to automatically match and deduplicate content across providers and regions. The result: 55 million raw entities resolved into 4.2 million verified unique titles, each assigned a single canonical ID. Real-time with accuracy over 99%+.
The companies that eventually came to us were facing the same frustration. They had used Gracenote and the same set of legacy vendors and were frustrated with the gaps and data quality issues. They quickly realized that Reelgood had the most accurate streaming data available, and wanted to license the dataset we had built.
B2B data licensing is now Reelgood's focus. And our popular consumer app is how we continue to validate our data at real-world scale. Today, our purpose is clear: Reelgood is the streaming data layer the industry relies on.
The streaming industry runs on accurate data. We built the infrastructure and data layer to provide it. Because nobody else had.
David Sanderson, CEO
Key Milestones
2015
Founded as a consumer 'where-to-watch' app
2015–2023
$20M invested building proprietary ML data engine
100M+ Users
Consumer app users used to train and validate the ML model
Enterprise
First inbound from major consumer tech companies
Today
300+ services, 4.2M verified titles, 25+ countries, 7 languages
What We Do
The Data Layer for the Streaming Industry
Our data products are available via REST API or bulk S3 export, delivered however your team needs them.
Streaming Title Availability
Where every title is streaming, right now. Real-time data across 300+ services globally, updated throughout the day. Covers SVOD, AVOD, TVOD, and TVEverywhere, with pricing model, territory, and multiplatform deep links (web, iOS, Android, smart TV).
See use cases →TV and Movie Metadata
285K movies, 70K shows, 163K seasons, 3.8M episodes, and 1.2M talent records, each assigned a single canonical ID mapped to EIDR and hundreds of streaming service-specific identifiers. Fields include cast, crew, synopses, ratings, runtime, genres, ML-generated contextual tags, localized titles, and artwork in multiple aspect ratios. Available in English, Spanish, French, German, Italian, Portuguese, and Hindi. Data is refreshed and available within 5 minutes of ingestion.
See use cases →Historical Availability Data
7+ years of streaming availability records. See precisely how any title has moved across services, platforms, and territories, including exact window open and close dates. New countries and languages can be added within 2 months.
See use cases →Trusted by the companies that run on streaming data
The Technology
Deep Expertise in Entertainment and Data.
Legacy metadata vendors were built for the cable era: teams of people manually entering data, batching updates, and inevitably introducing the kinds of errors that erode user trust. We built Reelgood differently from the start, and after 8 years of refinement, the difference is measurable.
from 300+ services
confidence scoring
single canonical ID
| Legacy Vendors | Reelgood | |
|---|---|---|
| Data collection | Manual entry teams | ✓ ML-based automation |
| Update frequency | Delayed, batch updates | ✓ Real-time, throughout the day |
| Accuracy | Error-prone, duplicates common | ✓ 99%+ ML-verified, 99.9% uptime |
| Matching method | Requires ID overlap | ✓ Metadata signals + confidence scoring |
| Title identity | Fragmented, duplicates across feeds | ✓ Single canonical ID per title |
| ID compatibility | Proprietary formats only | ✓ Third-party IDs like EIDR + streaming service IDs |
| Migration | Manual, high-risk cutover | ✓ Parallel-run validation, ML-matched |
Our ML pipeline consolidates 55M raw data entities from 300+ services using metadata signals and confidence scoring across cast, crew, runtime, synopsis, and release data. No ID overlap required to get accurate results.
Every title gets a single canonical ID, mapped to EIDR and hundreds of streaming service-specific identifiers. Portable across systems, durable across catalog changes, and reliable as a long-term source of truth.
Our ML-based matching ingests your existing catalog without requiring pre-existing ID alignment. We run parallel with your incumbent for validation before cutover. Most teams complete validation in weeks, not months.
Want the full picture? Our Technology page covers the ML methodology, data pipeline architecture, delivery specs, and SLA details.
Explore Reelgood's Technology →The Team
Deep Expertise. Direct Access.
Reelgood is a lean, senior team with deep expertise in ML, streaming infrastructure, entertainment, and data. When you work with Reelgood, you work directly with the engineers and data scientists who built the systems, not a layer of junior support staff.
David Sanderson
CEO & Co-Founder
David built early-stage products at Facebook, where machine learning was first being applied to scale complex operations without proportional headcount growth. He brought that same philosophy to Reelgood: build infrastructure that gets more accurate over time, not just bigger.
- Led Reelgood for 10+ years, from consumer app to enterprise data standard
- Oversaw $20M invested in building the proprietary ML matching engine
- Guided data partnerships with Google, Apple, Starz, Tubi, and others across 25+ countries
Pablo Lucio Paredes
Data Engineering & Architecture
Damien Capocchi
Backend Engineering
Javier Morán
Data Engineering
Felipe Lemarie
Data Science
Daniela Velasco
Data Analysis
David Markowitz
Marketing & GTM
-
Reelgood's data is refreshed continuously. New and changed data is available within 5 minutes of ingestion across all 300+ services. Streaming availability changes (new titles, removals, pricing updates) are captured and propagated in real time throughout the day, not in nightly or weekly batch jobs.
You can receive updates via the partner API (immediate) or via twice-daily scheduled S3 exports (incremental). Both include exactly what changed.
-
Yes. Our ML matching model maps any catalog to Reelgood's metadata using title, year, cast, and other signals: no shared identifiers required. If you have existing IDs (internal or third-party), those can be used as well, but they are not a prerequisite.
Reelgood returns metadata matched to whatever identifiers you prefer, minimizing the friction of onboarding or switching providers.
-
Data is available via REST API (JSON, TLS 1.2+, 50 calls/sec standard) or bulk S3 export. Exports support JSON, XML, CSV, Parquet, and ORC. Image assets are delivered via Cloudflare's global CDN at 99.9%+ uptime.
We can also deliver data mapped directly into your target schema if you provide one. Webhook notifications for data updates are available on request.
-
English metadata is available across all records as a baseline. Localized metadata (titles, synopses, posters, age classifications, and release dates) is available in Spanish, French, German, Italian, Portuguese, and Hindi, corresponding to supported territories.
Current territory coverage includes the US, Canada, UK, Germany, Australia, Ireland, New Zealand, India, Spain, France, Italy, Mexico, Argentina, and Brazil. New countries can typically be added within one month.
-
Yes. Reelgood's image library covers the full range of asset types across multiple aspect ratios: posters (2:3, 1:1, 4:3), backdrops (16:9), scene stills, logos, and celebrity headshots. Assets are localized per market where variants exist.
Images are delivered via Cloudflare's CDN (HTTPS, no authentication required) at 99.9%+ uptime. Movie poster coverage is 100% across the catalog. Season and episode coverage varies by asset type.
-
Reelgood maintains a canonical show-season-episode hierarchy. Multiple provider references collapse to a single canonical record, then expand back out into every market- and platform-specific variant. Where providers disagree on a season cut, Reelgood resolves to one canonical structure while preserving every variant beneath it.
The result is a consistent, deduplicated representation of any series regardless of how it's packaged across services or regions.
-
Reelgood's ML matching ingests your existing catalog and maps it to our canonical ID framework without requiring pre-existing ID alignment. We run in parallel with your incumbent for validation and reconciliation before cutover, so your team can verify accuracy and flag edge cases before any live traffic depends on our data.
This parallel-run approach reduces migration risk without extending timelines. Most teams complete validation within a few weeks.
-
Both the partner API and the image CDN operate at 99.9%+ monthly uptime. Critical issues receive a response within 1 hour and resolution within 24 hours. Reelgood operates 24/7 monitoring with an on-call rotation and severity-based incident response.
You work directly with our technical team, not a support tier or account management layer. Specific SLA terms, including service credits, are available in our enterprise agreements.
How We Work
The Principles Behind the Data.
Four principles shape how we build, how we make decisions, and how we work with our customers.
Simplicity over complexity.
Streaming data is inherently messy: hundreds of services, millions of titles, constant changes. Our job is to absorb that complexity so yours disappears. If our data requires explanation, we haven't done our job.
Fix the root. Not the symptom.
Bad data has a source. We find it. Patching errors at the surface doesn't build trust. Rebuilding the pipeline does. We've spent 8 years doing this right, not fast.
Quality is everyone's job.
Every person on this team, in every role, is accountable for the accuracy and reliability of what we ship. Data quality isn't a feature. It's the product. It always has been.
Own it.
We're a lean team. That means every person moves fast, fills gaps, and takes responsibility without being asked. That's not a constraint. It's how we've built something larger organizations haven't.
Let's Talk Data.
Whether you're acquiring and licensing titles for your service, evaluating your metadata stack, or researching competitor catalogs, we're happy to show you what accurate streaming data looks like in practice.
© Copyright 2026. All rights reserved.
T&C | Privacy Policy
© Copyright 2026. All rights reserved.
T&C | Privacy Policy