Mintel Teams with Dragonfly AI to Add Predictive Attention Intelligence to GNPD — the market‑intelligence firm announced Tuesday that every product entry in its Global New Products Database (GNPD) will now carry a “packaging performance score” powered by Dragonfly AI’s neuroscience‑backed attention analytics. The move aims to give CPG brands a data‑driven view of what consumers actually notice on shelf, before a product reaches market.
What the partnership delivers
The integration embeds Dragonfly AI’s patented visual‑attention model directly into Mintel’s GNPD, a repository that tracks tens of thousands of new product launches each month across more than 100 categories. When a brand uploads a new SKU, the system generates a score that predicts how likely the package is to be seen, remembered, and acted upon. The score is presented alongside traditional attributes such as price, category, and launch geography, turning a static catalog into a dynamic decision‑support tool.
How the technology works
Dragonfly AI’s engine is built on a decade of neuroscience research conducted with Queen Mary University of London. It simulates early‑stage visual processing in the human brain, measuring factors like contrast, colour hierarchy, and focal points before conscious attention kicks in. The model has been validated in multiple labs and is claimed to predict real‑world noticeability with a reported ≈ 85 % correlation to eye‑tracking studies. By feeding these predictions into GNPD, Mintel gives marketers a “first‑look” assessment of packaging effectiveness without the need for costly physical testing.
Why it matters for marketers
In an era where shelf space is increasingly crowded and first‑impression time is measured in fractions of a second, the ability to forecast attention can shave weeks off the creative‑approval cycle. Gartner estimates that by 2027 > 70 % of marketers will rely on AI‑driven creative testing to inform launch decisions. For enterprise marketing teams, the new score translates into three practical benefits:
- Prioritisation – Brands can rank hundreds of concepts quickly, focusing resources on the few that are predicted to break through.
- Risk reduction – Early‑stage insight reduces the likelihood of costly redesigns after a product has been produced.
- Cross‑market insight – Because GNPD covers global launches, the attention score can be compared across regions, helping multinational teams harmonise packaging strategies.
Competitive context
The packaging‑analytics space has seen a flurry of entrants, from pure‑play AI startups to legacy market‑research firms adding visual‑testing modules. Compared with rivals such as Cortexica and EyeQuant, Dragonfly AI differentiates itself through its biologically inspired model rather than purely pixel‑pattern analysis. Moreover, the direct embedding of scores into GNPD gives Mintel a distribution advantage; competitors typically require separate API calls or manual uploads, adding friction for end users.
Implications for the broader AdTech ecosystem
While the partnership is rooted in CPG packaging, the underlying technology has crossover potential for digital ad formats, especially in Connected TV (CTV) and Over‑the‑Top (OTT) environments where visual attention is a premium commodity. Companies like Google and Amazon have already incorporated attention‑based bidding signals into their ad‑exchange platforms. Mintel’s move signals a broader trend: data‑rich, AI‑enhanced signals moving from offline media (shelf) into programmatic buying, giving DSPs and SSPs richer context for inventory valuation.
Enterprise adoption considerations
- Data‑management alignment – Syncing GNPD’s score with a Customer Data Platform (CDP) or Data Management Platform (DMP) to enrich audience segments.
- Privacy compliance – Since the model uses anonymised visual data, it sidesteps first‑party/third‑party data tensions, but firms must still document AI‑generated insights under emerging AI‑model‑transparency regulations.
- Creative‑testing cadence – Teams may shift from iterative A/B tests to a “predict‑first, test‑later” approach, accelerating time‑to‑market.
Overall, the Mintel‑Dragonfly AI collaboration offers a concrete example of how AI can move from post‑hoc measurement into pre‑emptive decision‑making, a shift that could reshape how brands allocate spend across media channels.
Market Landscape
The packaging‑attention market sits at the intersection of traditional market research and next‑gen AI. IDC predicts that worldwide spending on AI‑enabled marketing technology will surpass $150 billion by 2026, driven largely by demand for predictive analytics that cut cycle time. At the same time, privacy‑first regulations (e.g., GDPR, CCPA) are nudging firms toward first‑party signals, making attention scores derived from visual content an attractive alternative to third‑party cookie data.
Within AdTech, platforms such as The Trade Desk and Adobe Advertising Cloud are already testing attention‑based metrics for video inventory. The Mintel‑Dragonfly AI integration could serve as a template for similar partnerships that bring offline visual intelligence into programmatic ecosystems, especially for retail media networks that blend shelf and digital experiences.
Top Insights
- Predictive attention scores give marketers a quantifiable visibility metric at the concept stage, shortening the creative‑approval cycle by up to 30 %.
- Dragonfly AI’s neuroscience‑backed model outperforms pure pixel‑analysis tools, delivering an 85 % correlation with real‑world eye‑tracking data.
- Embedding the score in Mintel’s GNPD creates a single source of truth for both product innovation and media‑planning teams, reducing data silos.
- The partnership illustrates a broader industry shift: AI‑driven pre‑emptive insights are moving from digital ad formats into offline media such as shelf packaging.
- Enterprise marketers can leverage the score within CDPs and DMPs to enrich audience segmentation without relying on third‑party cookies.
