Why Beauty Brands Are Flying Blind on Consumer Feedback And What Digitised Products Change

Beauty is one of the most data-intensive consumer categories in existence. Brands invest heavily in market research, consumer panels, social listening platforms, and retailer sell-through reporting. And yet the most commercially critical question, what actually happens between the moment a consumer picks up the product and the moment they decide whether to buy it again, is answered almost entirely by inference, not observation.

The reason is structural. Most beauty products are sold through retailers, department stores, pharmacies, and e-commerce platforms that own the consumer relationship. The brand receives a sell-in number, how many units left the warehouse. What happens after that is reconstructed from proxies: retailer reorder rates, returns data, social media sentiment, and periodic consumer surveys. None of these tell the brand what it actually needs to know: how the specific formulation in that specific package is being experienced by the specific consumers who bought it.

That observation, from one of the most senior data and AI leaders in the global beauty industry, describes the problem with precision. The current state of beauty consumer intelligence is social scraping harvesting signals from platforms that the brand does not own, that capture only the consumers who choose to post publicly, and that filter feedback through the unpredictable dynamics of social algorithms. The result is a distorted picture of consumer experience: amplified by extreme reactions, suppressed in the middle, and structurally unavailable for the quiet majority of consumers who use their products daily and say nothing at all online.

The structural gap between sell-in and consumer experience creates specific blind spots in the information beauty brands use to make decisions about formulation, allocation, marketing, and product development. Each one costs margin, misallocates investment, and delays the detection of signals that should be shaping strategy.

• Usage frequency and product completion rates: A serum that consumers use daily for 60 days is a different commercial proposition from one that sits on a shelf for six months before being discarded half-empty. Most brands cannot distinguish between these two outcomes from sell-in data alone. Reorder rate is a proxy — but it arrives weeks after the usage behaviour has already shaped the consumer’s relationship with the product.
  • What brands miss: Early signals of product abandonment, by market and channel, before they compound into declining repurchase rates.
• Geographic variation in product experience: A formulation developed for European skin and humidity conditions may perform differently in South East Asian climates, not because the formula is wrong, but because the consumer’s experience of it varies with environmental factors the lab cannot fully anticipate. Without direct consumer signal from products in use, this variation only becomes visible when it reaches the scale of a distributor-level sales decline. By then it is already a problem, not a signal.
  • What brands miss: Market-specific formulation feedback before it becomes a commercial issue, and the intelligence to adjust allocation or introduce market-specific variants proactively.
• The moment of repurchase decision: The most commercially important moment in a beauty consumer’s relationship with a product is the moment she decides whether to buy it again. Brands have no direct visibility into this decision. The reorder data arrives weeks after it was made, and the sell-in reporting shows the outcome but nothing about the process what competing products she considered, what triggered the decision, or whether the experience of the original product was the primary driver.
  • What brands miss: The behavioral signals in the weeks before a repurchase decision that could be acted on: a personalised offer, a care communication, a loyalty reward, rather than observed retrospectively in aggregate data.
• Formulation feedback from real-world use: Consumer panels and clinical testing provide controlled feedback on formulations under defined conditions. Real-world use is not controlled — it reflects the full complexity of skin types, application habits, climate, and product combinations that no panel can replicate. The gap between controlled testing outcomes and real-world consumer experience is where most formulation development surprises live, and most brands have no systematic way to close it.
  • What brands miss: Real-world formulation signals from diverse consumer populations, at scale, in the conditions where the product is actually used not the conditions under which it was tested.

The common failure across all current sources: Every signal source above tells the brand what happened units moved, consumers returned, sentiment trended positive. None of them tells the brand what is happening with the products that are right now sitting in consumers’ hands across the world, being used or abandoned, for reasons the brand cannot currently observe. This is the fundamental data gap that digitised products close.

The shift that digitised products create is not incremental it is structural. Moving from indirect, delayed, aggregate signals to direct, real-time, item-level signals changes not just the speed of decision-making but the quality of the decisions themselves. The whitepaper research that includes leaders from Coty and other major beauty groups describes this shift in terms that the category’s data leaders find compelling precisely because it resolves the most persistent limitation of their current intelligence infrastructure.

“Products generate data as they move, are used, serviced or resold. Information is tied to real items rather than estimated from proxies. Direct signal arrives earlier and with less ambiguity.”
— Digitised Products: Product Identity as Infrastructure, Selinko Toppan & Pivot & Co.

For beauty specifically, the signal that digitised products generate is behavioral in a way that no other source approaches. A consumer who taps their skincare product ten times in thirty days is using it daily. A consumer who tapped twice in the first week and has not tapped since has likely abandoned it, a signal that can trigger a targeted engagement before she decides not to repurchase. A cluster of consumers in a specific market who tap consistently in the morning but rarely in the evening reveals something about how the product is actually being used in that market that no sell-in report can show.

“Digitised products are not a channel or a tool. They are an operating model connector that links supply chain, commercial, marketing and data into one system.”
— Nicolas Comestaz, Vice President Global Data & AI CoE, Coty — quoted in Digitised Products: Product Identity as Infrastructure, Selinko Toppan & Pivot & Co.

This framing, digitised products as an operating model connector rather than a channel feature, is important for beauty brand leadership to internalise. The commercial value of an NFC chip in a moisturiser bottle is not the authentication it provides (though that matters for counterfeiting) or the content it delivers (though that builds engagement). It is the data infrastructure it creates: a direct, persistent connection between the brand’s intelligence systems and the consumers who are using its products right now, in conditions the brand has never previously been able to observe.

The behavioral intelligence produced by a digitised beauty product NFC tap is more specific, more timely, and more actionable than anything the current signal landscape provides. Here are the four most commercially significant signal types and what they replace.

1. Usage frequency and engagement depth: From repurchase rate to real-time usage signal
   Each tap is a usage proxy. The frequency, timing, and consistency of taps from a specific product unit reveals how integrated the product is into the consumer’s routine,  far more precisely than asking them in a survey or inferring from a reorder date. A consumer who taps every morning for eight weeks has a fundamentally different product relationship than one who tapped twice and stopped. Both might eventually repurchase, or not, but the signals arrive weeks earlier than any sell-through data, in time to act on them.
   • Without digitised products: Repurchase rate available 6–8 weeks after usage period ends. No visibility into engagement depth or abandonment trajectory.
   • With digitised products: Usage frequency observable in real time. Abandonment patterns flagged within days. Engagement interventions possible before the repurchase decision crystallises.
2. Geographic and market-level behavioral variation: From aggregate sell-in to market-specific intelligence
   Tap event data, aggregated across product units in a specific market, reveals how consumer engagement with a formulation varies by geography — which markets show high daily usage, which show early abandonment, which show seasonal variation. This intelligence is currently accessible only through expensive market-by-market consumer research; digitised products generate it continuously as a by-product of consumers interacting with their purchases.
   • Without digitised products: Geographic variation visible only through distributor sell-through disparities weeks delayed, requiring interpretation. Market research commissioned separately.
   • With digitised products: Market-level usage pattern comparison available from the same platform that manages authentication. No additional research investment required.
3. Refill and repurchase journey intelligence: From outcome data to process visibility
   A consumer who taps her skincare product daily, then stops tapping, then taps a new unit of the same product a few weeks later, has left a complete repurchase journey in the tap record. The brand can see the completion of the first unit (engagement drop-off as the product runs out), the gap, and the return. If the new unit carries the same NFC infrastructure, the next repurchase journey begins. Over time, the brand accumulates usage lifecycle data — how long specific products actually last in real consumer use that no sell-in data can provide and that changes how distribution and production decisions are made.
   • Without digitised products: Repurchase journey invisible between the two transactions. Why a consumer returned, or did not, is a matter of inference, not evidence.
   • With digitised products: Complete journey visible: usage lifecycle, gap between completion and repurchase, loyalty trajectory. Actionable at the individual consumer level in real time.
4. Counterfeit and grey market signal for product quality intelligence: From assumption to evidence on where product issues originate.
   Beauty counterfeiting is a significant and growing problem, with documented health and safety consequences for consumers who unknowingly purchase fake formulations. Digitised products detect counterfeit items at the point of consumer interaction. But the geographic and volume data on counterfeit scan attempts also generates product quality intelligence: where counterfeit products are appearing, at what scale, and in which channels. A cluster of failed authentication scans in a specific market is not just a brand protection signal, it is market intelligence about where the brand’s consumer protection problem is most acute, and where authorised products are likely being displaced by substitutes.
   • Without digitised products: Counterfeit activity discovered through consumer complaints, retailer reports, or brand protection investigations weeks or months after the issue has scaled.
   • With digitised products: Counterfeit scan cluster detected in real time by market and channel. Consumer safety risk addressable before it reaches the scale of complaints or media attention.

The integration of NFC into beauty packaging is simpler and less disruptive to existing production processes than most brand teams assume when they first encounter it. The chip does not require packaging redesign, new materials, or new production equipment in most cases  it is an addition to the existing packaging specification that is handled by the brand’s existing packaging suppliers.

1. NFC chip embedded or applied at packaging production: A unique NFC chip is embedded in the bottle, jar, or compact during packaging production in the base, the cap, the label, or the outer carton. Each chip receives a unique cryptographic identity at this stage, linked to a cloud record storing the product’s formula variant, batch, production date, and allocated market. This is the item-level identity that all downstream intelligence is built from.
2. Consumer taps, authentication and engagement in under 3 seconds: The consumer holds their phone near the NFC area of the product. No app required. The chip generates a unique cryptographic message validated by the brand’s cloud backend. Authentication confirmed. The consumer sees a branded page: ingredient provenance, usage guidance, loyalty rewards, or whatever experience the brand has configured for this product at this lifecycle stage.
3. Tap event logged direct consumer signal generated: The tap event is logged: chip identity, timestamp, location (where consented by the user), and lifecycle stage. Each tap adds to the behavioral intelligence record for this product unit and, in aggregate, for this formula, this market, and this consumer segment. The brand’s intelligence platform receives real-time behavioral data from products in active use a signal source that has never previously existed for most beauty categories.
4. Brand intelligence platform surfaces actionable patterns: Aggregate tap intelligence is available in the brand’s intelligence dashboard: usage frequency by market, formula, and channel; abandonment trajectory patterns; counterfeit detection clusters; repurchase journey data; and engagement depth by consumer segment. These patterns are available in real time, not weeks after the fact and they are observable at item level, not only in aggregate sell-in numbers.
5. Personalised consumer engagement triggered by behavioral signals: The same platform that generates intelligence also enables action on it. A consumer whose tap frequency drops, a potential abandonment signal, can receive a targeted engagement through the next tap: a personalised tip, a loyalty reward, a reformulation message, or a refill offer timed to when the product is likely running low. The response is triggered by real behavior, not a campaign calendar.

The EU Digital Product Passport is expanding to cover cosmetics alongside textiles, electronics, and other product categories. For beauty brands, the compliance requirements are directly aligned with the consumer intelligence infrastructure that digitised products create making DPP compliance a by-product of commercial deployment rather than a separate regulatory investment.

The recall use case that changes the economics of DPP investment: For beauty and personal care brands, the ability to identify precisely which units are affected by a formulation recall, rather than withdrawing entire production batches is a commercially significant capability that digitised product infrastructure enables as a by-product of the same item-level identity used for authentication and consumer engagement. A brand that can notify only the consumers whose specific lot is affected, and precisely track the recall response, recalibrates the cost-benefit analysis of the entire DPP investment.

The most important strategic framing from the whitepaper research contributed by a VP-level data leader at one of the world’s largest beauty groups  is that digitised products are not a channel or a tool. They are an operating model connector. This distinction matters for how beauty brand leadership should think about the investment decision.A channel is something the brand uses to reach consumers email, social, retail. Its value is defined by reach and conversion. A connector is infrastructure that links previously separate functions in this case, supply chain, commercial, marketing, and data through the shared signal generated by the product in the consumer’s hands. The value of a connector compounds across functions. When the same product tap that generates a brand protection alert also enriches the CRM record, also updates the demand planning model with real usage data, and also triggers a personalised consumer engagement that is not a marketing feature. It is an operating model capability.

The compounding value argument: A beauty brand that deploys digitised products for authentication in year one gets brand protection. A brand that integrates that authentication infrastructure into its CRM in year two gets first-party consumer data. A brand that feeds that consumer data into its demand planning model in year three gets more accurate allocation. A brand that uses that allocation intelligence to reduce overstock and markdowns in year four generates margin improvement. The investment is made once, at the product level, and the value compounds as the infrastructure is connected to more functions. This is the case Nicolas Comestaz is making when he describes digitised products as an operating model connector.