ThinkAnalytics Alternative: One Discovery API You Can Test Before a Sales Call
ThinkAnalytics sells one platform across discovery, metadata and advertising. TasteRay does one thing: it ranks titles against what a viewer says and explains why. Your scope decides this.
TasteRay is an alternative to ThinkAnalytics for the viewer-facing discovery job alone: a public API that ranks titles against what someone says, with a written reason, and can be tested without a sales call. ThinkAnalytics suits broadcasters and operators who want one platform across recommendations, metadata enrichment and ad audiences.
ThinkAnalytics describes ThinkMediaAI as the platform behind three solutions: ThinkMetaData, ThinkAdvertising and ThinkDiscovery. Only the last overlaps with TasteRay. If your project covers all three, you are comparing a platform with a component, and the component will lose that comparison. If your project is discovery, read on.
ThinkAnalytics and TasteRay side by side
The ThinkAnalytics figures below are its own statements. We did not verify them, and they describe a kind of vendor: one built around viewing behaviour across TV platforms.
| Question | ThinkAnalytics | TasteRay | Source |
|---|---|---|---|
| Audience | "TV and streaming platforms", including TV networks | Streaming services, broadcasters, telecoms, pay TV and stores | ThinkAnalytics, Enterprise |
| Scope | Recommendations and search, metadata enrichment, and ad audiences on one platform | Recommendations and explanations. No metadata enrichment or ad products | Platform, TasteRay docs |
| Main signal | Individual and household viewing behaviour | What a viewer says, with any history you choose to send | ThinkDiscovery, TasteRay docs |
| Content knowledge | ContentDNA, stated at 40K+ attributes across mood, theme, subject, setting and character types | Your names and descriptions in each request. Enterprise can train on your content | Platform, Enterprise |
| Discovery surfaces | Rows, carousels, search, curated rails, FAST personalization, A/B testing and orchestration | A ranked list per request. Your team builds rails, testing and search | ThinkDiscovery, TasteRay docs |
| Stated scale | 8B+ daily recommendations, 500M+ viewers, 47 languages, 20+ years (vendor figures) | We state limits: up to 200 requests a minute on Pro, custom on Enterprise, a few seconds per response | Platform, TasteRay docs |
| Integration model and data requirements | Not listed in their docs (as of Nov 2026) | Public REST docs. No training data needed to start | ThinkDiscovery, TasteRay docs |
| A reason per result | Not listed (as of Nov 2026) | why_match and key_factors on every result | ThinkDiscovery, TasteRay docs |
| Public price | Not listed (as of Nov 2026). The site offers "Talk To An Expert" | Free, $49 and $199 a month. Enterprise custom | ThinkDiscovery, TasteRay prices |
| Trying it | Through a conversation with the vendor | A keyless playground, then a free key, then a pilot | Playground |
Platform or component
- A platform makes sense when
- One team owns discovery, metadata and ad yield together, and wants one data model and one contract across them. Consolidating vendors is the goal in itself.
- A component makes sense when
- You already have a recommender for your rows and metadata from another supplier, and there is one gap: a viewer who wants to ask for something in their own words and read why it was offered.
- What a component costs you
- Your team or integrator builds the screens, joins the data and runs the experiments. TasteRay ships a REST API with no TV SDK, so those costs are real on a set-top box.
Questions to put to any discovery vendor
We would put these in an RFP, and we answer them on our own pages. They work equally well for ThinkAnalytics, TasteRay or anyone else on your list.
- What do you need from us before the first useful result, in records and in weeks?
- What does a viewer see on day one, before any viewing history exists?
- What does a title get if nobody has watched it yet?
- Can a viewer ask in their own words, in each language we broadcast in?
- Can you show a reason a viewer could read, and who checks it for wrong facts?
- How is a household told apart from the person holding the remote?
- What leaves our environment, who receives it, and how long is it kept?
- What are the response times at the percentile we care about, and where is that in the contract?
- If a pilot shows no movement against a control group, how do we exit?
Who you would work with
TasteRay is a smaller company with a narrower product, and you would deal directly with people who have years of experience running product, technology and growth at scale. Mike Jaskolski led product and technology for Nokaut and Morizon, both of which went public, and builds the engine, trained on more than 500,000 long-form reviews. Mike Wolniak built and sold the VML hub in CEEMEA. See the team.
When to choose ThinkAnalytics
- You are an operator or broadcaster that wants recommendations, metadata enrichment and ad audiences on one platform.
- You need household-level modelling from viewing behaviour across your own TV estate.
- You want FAST curation, editorial orchestration and A/B testing from your discovery vendor.
- A vendor stating very large scale and 47 languages is what your procurement process requires.
When to choose TasteRay
- Your gap is a viewer asking in their own words and reading why, and everything else is covered.
- Your editors want to test on your own titles before anyone schedules a sales call.
- You want to start with a pilot on one surface and one part of the catalog.
- Your team can build the screens and hold a cache in front of a call that takes a few seconds.
ThinkAnalytics is a separate company; see its own site. TasteRay is not affiliated with it.
Sources and further reading
- ThinkAnalytics homepage (ThinkAnalytics, opened 5 November 2026)
- ThinkMediaAI platform (ThinkAnalytics, opened 5 November 2026)
- ThinkDiscovery (ThinkAnalytics, opened 5 November 2026)
FAQs about ThinkAnalytics Alternative
What does ThinkAnalytics do?
ThinkAnalytics offers ThinkMediaAI, a data platform for TV and streaming with three solutions as of November 2026: ThinkMetaData for enrichment, ThinkAdvertising for audiences, and ThinkDiscovery for recommendations, search and curated rails. It states 8B+ daily recommendations and 500M+ viewers. Those figures are its own, and we have not verified them.
How much does ThinkAnalytics cost?
ThinkAnalytics does not list a price on the pages we opened in November 2026, and its site routes you to a conversation with an expert. Ask for a quote scoped to the solutions you need. For comparison, TasteRay publishes free, $49 and $199 monthly tiers, and quotes Enterprise individually.
Can TasteRay replace a ThinkAnalytics deployment?
Only partly. TasteRay covers one job in that scope, which is ranking titles against what a viewer says and explaining the result. It offers no metadata enrichment, ad audiences or FAST curation, and it has no TV SDK. A smaller step is to add TasteRay beside what you already run, starting with a pilot.