Nestlé: Reduce sugar without losing the customer

A lower sugar recipe can disappoint customers, increase ingredient costs or create production difficulties. A commercially successful recipe can also deliver too little nutritional improvement to justify its positioning. The central decision is therefore which combination of benefits and trade-offs is acceptable for a specific category, brand and market. Broad statements about health-conscious consumers do not establish willingness to buy a particular reformulated confectionery product.

Fishbone analysis and creative problem solving connect consumer behaviour, regulation and internal processes. Acceptance becomes a design requirement: the category team should define the customer experience, nutrition objective, cost envelope and compliance requirements before selecting a formulation route.

Sweetener blends and AI-assisted formulation are part of a broader option set. Nestlé’s July 2023 announcement describes enzymatic reduction of intrinsic sugars in selected ingredients, while its nutrition research article describes AI-assisted recipe development. These disclosures establish that relevant capabilities exist; they do not prove that either route will work for a particular confectionery pilot.

A proposed comparison would test gradual recipe adjustment, permitted sweetener or bulking combinations, and suitable processing technologies. Each should be assessed for taste, texture, stability, ingredient availability, line compatibility and fully loaded cost. High-intensity sweetening and fibre functionality are separate formulation requirements; replacing sweetness does not necessarily replace sugar’s structural properties. Ingredient suitability and claims require product-level evidence.

A broad immediate rollout offers faster portfolio coverage but exposes more brands to an untested taste change. A separate healthier sub-brand can limit disruption to a core recipe, but may fragment demand and require additional marketing expenditure. A staged change to a selected existing product provides a practical learning route when distribution, customer cohorts and manufacturing trials can be observed. It is the preferred initial option, subject to baseline data and category economics.

Initial qualitative screening can prioritise experiments, but cannot establish financial merit or predict consumer behaviour. The decision process must make trade-offs visible: a route that improves sensory acceptance but materially worsens cost or stability returns to development. A route that reduces sugar without enough nutritional improvement to support the intended proposition should also be reconsidered. Explicit thresholds make the investment decision more defensible than a general rating of attractiveness.

The proposed pilot begins with one market and a clearly specified product, chosen for measurable repeat purchase and an observable manufacturing process. Nutrition and regulatory specialists establish the permitted formulation and claims. R&D compares candidate recipes with the current product using blinded sensory assessment and stability testing. Operations then tests yield, waste and throughput. Any use of AI should be evaluated against measured formulation outcomes, including cases where predictions fail.

A limited market test should compare repeat purchase, complaints, price response and contribution margin against a baseline, adjusting for promotion and distribution differences. Pre-agreed stop rules help prevent favourable reinterpretation of weak results. This remains a proposed validation design rather than completed consumer research.

Resistance often reveals a valid operating concern. Marketing may question taste consistency, procurement may flag supply reliability, and factories may anticipate process losses. A proposed Healthy Reformulation Team should include these functions alongside nutrition, R&D and regulatory affairs, with a category sponsor authorised to resolve trade-offs. An issue register should record the concern, required evidence, owner and decision date rather than treating alignment as attendance at workshops.

Value measurement should combine sugar per standard serving and per 100 grams with consumption and sales context. Commercial measures should include repeat purchase and contribution after reformulation costs; manufacturing measures should include scrap, yield and complaints. A lower sugar figure alone does not establish better population health, stronger demand or profitable growth. The purpose of the scorecard is to prevent improvement in one dimension from concealing deterioration elsewhere.