loading...
Qloo's Cultural AI maps hidden connections between human tastes – letting brands predict what their audience wants before they know it themselves.
ENTRY ANGLES
AI-powered recommendation engines that incorporate lifestyle and contextual data beyond product attributes · Computer vision systems that auto-tag product photos to improve product discoverability · Multi-dimensional recommendation systems that synthesize data across lifestyle preferences and external context
VERTICALS
CAPABILITIES
AI/machine learning for contextual understanding and personalization, Computer vision technology, Data integration across multiple domains and data sources
QLOO FOUNDER
“What podcasts do people who drink Heineken tend to enjoy?”
"What podcasts do people who drink Heineken tend to enjoy?" That's the kind of question Qloo was built to answer.
Or something more layered: "Which celebrities appeal to teenagers in Tokyo who are fans of Stranger Things?"
To answer those kinds of questions, the startup built what it calls a "Cultural AI" – a proprietary intelligence engine trained on the connections between human tastes.
The value here isn't just in the algorithms. It's in the scale of the underlying dataset: 4.5 million films, actors, and directors; 1.5 million TV series and hosts; 142.5 million songs and musicians; 550,000 fashion designers and brands; 2.95 million restaurants and cafes; 85 million brands and products; 27.5 million books; 1.55 million podcasts, apps, and video games; 4.7 million hotels and entertainment venues; 950,000 bars and clubs; 375,000 public figures and influencers; and 25,000 athletes and sports teams.
From this, Qloo can surface roughly 750 billion correlations between cultural entities and their audiences – all accessible via API.
Companies use that API to embed nuanced personalization into their own products, powered by non-obvious connections across user tastes.
Starbucks uses Qloo's API to generate music playlists for its cafes – tuned automatically to the musical preferences of people who live nearby. Netflix uses it to optimize advertising and merchandising for different audience segments by age and geography, surfacing actors who resonate specifically with each group. Hershey uses Qloo to build custom chocolate assortments for different buyer segments.
Now Qloo is expanding into real estate, finance, and insurance – backed by a fresh $25 million raise that brings total funding to $57 million.
Most personalization and recommendation systems in use today operate within a single domain.
An online clothing retailer, for example, recommends items based on what a shopper has previously bought at that same store, or what buyers of a similar age typically choose – because that's the only data available to its algorithms.
A podcast app suggests new shows based solely on what a listener already likes, maybe cross-referenced by age group or geography.
Those constraints prevent products from getting truly close to their individual users – which limits both conversion rates and retention.
Qloo dramatically widens that aperture. Imagine an online clothing store that asks a new visitor which music they listen to and what beer they drink – and then recommends a look based on those answers.
A second use case: figuring out where to advertise. Say you're launching an energy drink for young adults. Which podcasts should you sponsor? Which creators should you partner with? Which celebrity faces move the needle for your specific audience? These connections are often far from obvious.
A classic Nielsen study from 2017 found that the top-selling category among comedy podcast listeners was baby food. Think about it – new parents can't get out to stand-up shows, so they listen to comedy at home. Obvious in hindsight, invisible in advance.
Less intuitive: tech podcast listeners over-index heavily on wine and spirits, while film podcast listeners skew toward beer. And pet food ranks second among film podcast listeners. Why? Nobody knows without the data.
Research like Nielsen's is expensive to commission and quickly goes stale. Qloo produces comparable (and more granular, more current) insights at a fraction of the cost – down to specific brands, not just product categories, broken out by age and geography.
Worth noting: Qloo runs a consumer-facing product alongside its B2B API – TasteDive. Anyone can use it to discover what else they might enjoy based on what they already like. The twist: TasteDive, like Qloo itself, crosses category lines. It might recommend a set of podcasts to someone who loves Pink Floyd – and none of them would be about classic rock.
Qloo and TasteDive share the same underlying technology but serve distinct audiences: Qloo for product builders and enterprises, TasteDive for end users.
Recommendation systems are now table stakes. It's hard to imagine an online store without one – built in-house or powered by a third-party platform.
But AI is enabling a qualitatively new generation of recommenders that can get far more precise about individual users.
The fashion e-commerce industry is estimated to lose roughly $300 billion annually because its recommendation engines can't incorporate lifestyle context – only product attributes.
A query like "what should I wear on a first date in Miami" requires a system that understands Miami's climate, what reads as appropriate for a date, and current style trends.
That's exactly what Curated For You built – [covered here](/review/nenajdennye-300-milliardov-dollarov) in November 2022, having raised $5.9 million. Lily AI, [reviewed previously](/review/nedoponimanie-na-300-milliardov-dollarov), takes a different angle: computer vision assigns dozens of additional tags to product photos, helping shoppers actually find what they're looking for. That startup raised $51.9 million.
Qloo extends the capability further – adding dimensions that no single-domain system could surface. The query becomes: "What should I wear on a first date in Miami to impress someone who loves Heineken and Pink Floyd?"
The recommendation engine market was already sizable at roughly $3 billion in 2022 – and analysts projected growth to $54 billion by 2031, compounding at 37% annually. That trajectory, combined with AI's arrival in this space, makes the timing unusually good for new entrants.
The opportunity: build next-generation recommendation platforms that use AI to connect taste signals across categories in ways that existing systems simply can't.