Beyond a marketplace brand: 9 beauty launch architectures for 2026

The central finding

A new beauty venture often begins with a platform question: what could we sell on Amazon, Ozon, Wildberries, Tmall or Shopee? A founder then chooses products, a manufacturer, packaging and a target cost. That remains a viable approach. It is one possible starting point among several.

The environment in which demand first emerges can influence formula, claims, pack size, price, margin, evidence, initial order quantity, logistics and content. Choosing it after production can lock the business into a product poorly suited to its commercial route.

McKinsey’s June 2026 report assigns e-commerce 28% of global beauty sales in its channel classification. That category combines retailer websites, brand websites, marketplaces, social commerce and agentic commerce. The report identifies social commerce as the fastest-growing channel and projects approximately USD 4 billion in US TikTok beauty sales for 2026. This is an outlook, not a completed annual sales result.

A launch environment is an input to business design, not simply a destination for a finished product.

Sources [01]

What is a launch architecture?

A launch architecture connects five questions: where demand originates, what creates trust, where the transaction occurs, why customers return and whether the economics can sustain the route.

Discovery × Trust × Transaction × Retention × Economics

The multiplication sign is a strategic metaphor, not a mathematical model or a validated scoring formula. A serious weakness in one part can undermine the whole: demand does not repair negative contribution, professional credibility does not solve supply failures, and a viral first batch does not establish repeat purchasing.

Two constraints surround that system: Regulation, which determines what a product may be and claim in a particular market, and Supply, which determines whether production and logistics can support it. Product feasibility and these constraints should be assessed iteratively, alongside commercial design.

1. 01Discovery
2. 02Trust
3. 03Transaction
4. 04Retention
5. 05Economics

Constraints: regulation and supply capability.

Why architectures rather than exclusive channels?

Alse.Beauty 9 Launch Architectures is an original analytical framework, not an official or universally accepted industry taxonomy. Its categories deliberately use different axes. They identify the factor most strongly shaping an early business.

Marketplace-first centers on platform demand; Creator-first on a source of trust; DTC on direct relationships; Dermocosmetic on evidence; Retail-ready on distribution requirements; Export-first on a destination market.

A venture can combine Creator-first, DTC and Export-first, or Professional Beauty, Dermocosmetic and Retail-ready. Identify a primary architecture in which to test demand first, followed by secondary routes that can scale demonstrated value.

Which commercial system should first prove that this product is needed?

Nine architectures at a glance

The same cream may require a different commercial design depending on its first route to the customer. The comparison below identifies the dominant principle and main risk; it does not rank one architecture above another.

Combining architectures without losing direction

A multichannel business is not channel-neutral. Channels can have different roles and timing, and expanding too early can multiply inventory and operational costs.

• Creator → DTC → retail
• Professional → DTC → pharmacy
• Social commerce → marketplace → retail
• Export-first → retail-ready
• Hospitality → DTC → specialist retail
• Dermocosmetic → pharmacy → marketplace

These are illustrative pathways, not validated recipes. Use the primary architecture to test demand, trust, pricing, contribution and repeat behavior. Each additional channel should answer a new commercial question.

AI as a layer across the nine architectures

AI can participate in discovery, consideration, comparison, transaction and service. In this framework we treat it as a cross-cutting capability, rather than adding a tenth architecture.

Possible roles include marketplace shopping assistance, DTC advice, retailer discovery, an expert’s knowledge interface, multilingual product information and professional support. Each requires a defined scope, reliable information and appropriate human oversight.

Sephora announced a US ChatGPT pilot on 24 March 2026 for discovery, recommendations and shopping. Payment and checkout inside that experience were assigned to future updates. A launch capability and an announced roadmap must remain distinct.

Sources [04]

What makes a beauty brand readable by AI?

A useful product record connects identity, category, ingredients, purpose, directions, applicable warnings, substantiated claims, evidence, pack size, variants, price and availability. Compatibility should be stated only when justified.

Google says AI Overviews and AI Mode do not require special AI markup, a new schema type or new machine-readable AI files. Ordinary search fundamentals still matter: crawlability, helpful content, accessible text, useful images, internal links and structured data consistent with the visible page.

This guidance is specific to Google Search; it is not a promise of inclusion or ranking in every AI system. Machine-readable knowledge is an operational discipline, not an SEO shortcut.

Sources [11]

Design product and route together

A linear process may run from idea to formula, packaging, production and only then a search for customers. A commercially informed process starts earlier with the destination market, discovery, trust and channel economics, and uses those findings to shape product requirements.

Market → discovery and trust → route economics → product requirements → formula, claims and packaging → production

Social commerce may change sensory and demonstration needs; marketplaces the price-to-pack relationship; professional use the format and protocol; retail the wholesale margin and cash cycle; export the label and regulatory preparation; hospitality the refill design; evidence-led positioning the testing program.

This is iterative work. Commercial intent cannot override formulation feasibility, safety or regulation, and research may require revisiting earlier decisions.

Seven questions before the first production run

1. Discovery

Where will the customer first encounter the product: search, a creator, a practitioner, a shelf, a hotel or an AI interface?

2. Trust

What supports belief: appropriate evidence, a recommendation, verified reviews, experience or a product trial?

3. Transaction

Where does the first purchase occur, and who controls its terms and customer relationship?

4. Retention

Why would the customer buy a second, third or fifth time? Test performance, routine, service and replenishment rather than assuming loyalty.

5. Economics

Does contribution withstand the actual cost of reaching and serving that customer?

6. Regulation

What can this product contain, do and claim in the selected jurisdiction?

7. Supply

Can manufacturing and logistics support an unexpectedly successful pilot without compromising quality?

Use the answers to define the formula brief, pack format, minimum order quantity, pricing and channel plan. Unresolved questions become research tasks before commitments are made.

Cost of Route to Market

A low factory cost does not establish strong unit economics. Four similar serums sold DTC, through a retailer, through affiliates or through an overseas distributor can generate very different contributions at the same consumer price.

Include platform and payment fees, retailer or distributor discounts, acquisition, affiliates, shipping, warehousing, returns, samples, promotions, support and the working-capital cycle. Avoid counting the same cost twice and separate fixed launch investment from recurring per-unit costs.

Cost of Route to Market is Alse.Beauty’s analytical frame for comparing the commercial cost from production to initial and repeat purchasing. It is not a standardized accounting metric. Define the period, unit, net revenue basis, cost allocation and cash timing before comparing scenarios.

A practical output is a scenario model with explicit assumptions and sensitivity ranges, rather than one unexplained “profit” number.

Marketplace remains one credible route

A brand can originate in platform search, a creator community, social content, a direct relationship, professional use, evidence-led positioning, export distribution or a hospitality contract. It can later add other routes.

In which environment should the product first demonstrate that customers need it?

That answer can change the formula brief, claims, packaging, price, cost target, initial batch, research program, manufacturing location, content, supply chain and funding needs. A marketplace is a legitimate architecture without being the mandatory center of every venture.

Research conclusion

The nine architectures identify different ways of proving early commercial viability. They are compatible lenses, not mutually exclusive company types or a ranking of guaranteed opportunities.

Platforms, interfaces and advertising costs will change. The durable questions remain: how demand forms, why people trust, where they purchase, why they return, whether the economics work, whether the proposition is lawful and whether supply can deliver.

Design a beauty business around a specific mechanism for creating, testing and scaling demand.

Method and status of the framework

Alse.Beauty 9 Launch Architectures is an original analytical framework for comparing launch scenarios, product requirements, channel economics, scaling risks and international expansion. It is not an official industry taxonomy or an empirically validated prediction model.

Source-based review updated on 28 September 2026. Market estimates and forecasts remain attributed to their authors. Announced launches are described using the dates and scope in the source; platform access and actual transactions were not independently tested.

The financial discussion is a planning framework, not an investment return forecast. Regulatory summaries support strategic scoping and do not replace product-specific professional advice. Relevant rules and service availability can change.

Illustrations are AI-generated editorial concepts, not official materials, product photographs or interface reproductions from the companies mentioned. No sponsorship or affiliation is implied. The images do not encode quantitative data; the tables and text carry the exact model.

Research sources

Primary corporate announcements, official regulatory materials and Google documentation support the factual statements. McKinsey provides the cited market estimates. Alse.Beauty’s proposed model and business interpretation are distinguished from those sources.