The Future Wins When It Finds a Familiar Shelf

What does a liquor store selling THC drinks and an artificial intelligence startup raising millions have in common?

At first glance, almost nothing. One belongs to a regulated consumer market, the other to the technology sector. One is poured into a can, the other is presumably embedded in software. Yet both point toward the same important shift in how new categories become ordinary: the breakthrough is rarely the invention itself. It is the moment the invention finds a trusted place in an existing habit.

A liquor store reporting that THC drinks represent nearly 10% of its sales is not merely reporting a successful product line. It is revealing that cannabis consumption is being reorganized through a familiar retail ritual. Meanwhile, a five million dollar seed investment in an AI company is not simply a bet on clever algorithms. It is a bet that intelligence can be inserted into an existing workflow and become as unremarkable as a search bar, a spreadsheet, or a point of sale system.

The deeper question is this: How does society decide that something unfamiliar is safe, useful, and normal enough to adopt?

The answer has less to do with novelty than with placement.

New categories do not win by being new

Entrepreneurs often describe innovation as a race to create something unprecedented. But consumers and businesses do not adopt novelty in the abstract. They adopt it through context. A product must answer not only, “What can this do?” but also, “Where does this belong in my life?”

This is why the same underlying idea can fail in one setting and flourish in another. A new beverage sold through an unfamiliar channel asks customers to learn a product, a ritual, and a purchasing environment at the same time. A THC drink sitting beside established beverages in a liquor store asks for only one additional decision: whether to choose it.

That difference is enormous.

The liquor store acts as a contextual translator. It tells shoppers that the product belongs to the world of beverages, occasions, taste, and responsible selection. The customer does not need to build a new mental category from scratch. The store has already done much of the interpretive work.

AI faces a similar challenge. “Artificial intelligence” is a vast and intimidating category. It evokes autonomous systems, job displacement, science fiction, and technical complexity. But an AI tool placed inside a familiar task can shrink that psychological distance. Instead of asking a worker to “use AI,” it might help draft a document, sort information, generate a design, or automate a repetitive step. The technology becomes less important than the job it quietly improves.

The fastest route to mass adoption is often not a new behavior. It is a familiar behavior with a new capability hidden inside it.

This suggests a useful distinction between invention value and integration value. Invention value concerns what a product can do in isolation. Integration value concerns how easily it fits into an existing system of habits, trust, language, and distribution.

Many companies obsess over the first and underestimate the second. They ask whether the product is technically impressive. They ask whether it is cheaper, faster, or more powerful. They ask whether the market is large. They ask less often whether the product has a natural home.

Yet a product without a home has to sell both itself and the behavior required to use it.

The three stages of normalization

A useful way to understand emerging categories is to separate adoption into three stages: possibility, permission, and placement.

1. Possibility

This is the technical or commercial breakthrough. Something becomes feasible that was previously too expensive, too difficult, or too restricted. AI systems become capable enough to perform useful work. Cannabis beverages become viable as a consumer format. Capital begins to flow toward companies that believe these possibilities can become businesses.

Possibility attracts inventors and investors, but it does not guarantee customers. A technology can be possible for years before it becomes practical.

2. Permission

Permission is the social and psychological signal that says, “People like me can use this.” It may come from regulation, trusted retailers, respected brands, visible peers, or simple repetition.

A customer who sees THC drinks sold through a conventional liquor store receives a different signal from a customer encountering them in an isolated specialty shop. The product has entered a recognized commercial environment. It may still require education and caution, but it no longer feels entirely outside the familiar order.

AI products require permission too. A manager may be intrigued by artificial intelligence but reluctant to authorize it inside a company because of privacy, accuracy, or reputational concerns. Adoption accelerates when the tool is endorsed by an existing workflow, a credible vendor, a respected colleague, or a clear policy. Trust is not an accessory to the product. It is part of the product.

3. Placement

Placement is the moment the new capability appears exactly where a decision is already being made. A THC drink is available where someone is already choosing a beverage. An AI feature appears where someone is already writing, searching, designing, or analyzing.

Placement reduces what might be called behavioral distance: the number of unfamiliar steps between interest and use.

Imagine two products with identical effects. Product A requires a new account, a separate device, a new vocabulary, a training session, and a new routine. Product B appears inside a tool people already use, with familiar controls and a clear benefit. Product B will usually win, even if Product A is technically superior.

The lesson is not that quality is irrelevant. Quality matters once a product is in the consideration set. But distribution and context determine who enters that set in the first place.

Why ten percent matters more than it seems

A statement that THC drinks account for nearly 10% of sales is striking because it describes more than consumer curiosity. Curiosity produces occasional purchases. A meaningful share of revenue suggests that the product has begun to participate in the store’s operating reality.

It may affect inventory decisions, shelf space, staff knowledge, customer questions, promotions, and the kinds of occasions the store serves. In other words, the category is not merely present. It is becoming legible to the business around it.

This is a critical threshold for any emerging market. Before a product reaches it, the retailer treats the item as an experiment. After it reaches it, the retailer may begin treating the item as a department.

The same transition can occur inside an organization adopting AI. At first, AI is a side project. A few employees experiment with it privately. Someone creates a prototype. A team uses it to save time on an annoying task. These activities are interesting but fragile.

The category becomes durable when it changes resource allocation. The company updates its processes, revises job descriptions, establishes review standards, trains employees, and designs systems around the new capability. AI has moved from an exciting tool to an operating layer.

This leads to a broader principle:

A new category becomes real when institutions begin making room for it before customers explicitly demand that room.

Retailers allocate shelf space. Companies allocate permissions. Managers allocate budgets. Consumers allocate attention. Those acts of allocation are stronger evidence of normalization than enthusiasm alone.

The hidden business is not the product. It is the translation layer

Emerging markets often create opportunities for companies that do not invent the underlying technology or substance. These companies build the translation layer between novelty and ordinary life.

A liquor store can translate a new intoxicating product into an established beverage occasion. It can explain dosage, flavor, timing, and expected effects. It can help customers compare options without requiring them to become experts in cannabis policy or chemistry.

An AI company can perform a parallel service for knowledge work. It can translate a general purpose model into a workflow that reflects a specific profession, vocabulary, risk tolerance, and standard of quality. The winning product may not have the most powerful model. It may have the clearest understanding of what a particular user is trying to accomplish.

This is why narrow applications can be strategically valuable even when they appear less ambitious than general platforms. A general system offers capability. A specialized system offers judgment about when and how to use capability.

Consider the difference between a kitchen stocked with every possible ingredient and a meal kit designed for tonight’s dinner. The stocked kitchen has greater theoretical possibility. The meal kit has less friction. It tells the user what belongs together, what to do next, and what a successful result should look like.

The same is true of new consumer products. A shelf full of unfamiliar options may overwhelm customers. A retailer that organizes those options around clear use cases, such as social occasions, low sugar preferences, or alcohol alternatives, turns abundance into guidance.

The strategic opportunity lies in reducing ambiguity.

The risk of confusing visibility with adoption

There is, however, a danger in reading early traction too optimistically. A product can become visible without becoming deeply adopted. It can occupy shelf space, attract media attention, and generate trial purchases while failing to produce repeat behavior.

This distinction matters for both THC drinks and AI tools.

For a beverage, the key questions include whether customers understand the effects, whether the experience is predictable, whether the price feels reasonable, and whether the product fits a recurring occasion. A one time purchase may reflect novelty. Repeated purchases reveal habit.

For AI, a demo can be impressive while daily use remains disappointing. The system may produce errors, require too much supervision, or create new risks that outweigh the time saved. A successful pilot is not the same as a successful operating model.

A practical adoption funnel therefore looks like this:

1. Awareness: People know the category exists.
2. Trial: They are willing to experiment once.
3. Reliability: The experience is consistent enough to trust.
4. Routine: The product becomes part of an existing habit.
5. Infrastructure: The surrounding institution changes to support it.

The most revealing metric is rarely awareness. It is movement from routine to infrastructure. That is when a category stops depending on novelty and begins generating its own momentum.

Investors and operators should therefore ask a more demanding question than, “How many people tried it?” They should ask, “What behavior, budget, or workflow has changed because this exists?”

A practical framework for builders and buyers

The intersection of these markets offers a compact framework for evaluating any emerging category. Call it the Familiarity Fit Matrix. Before launching or adopting a new product, examine four forms of fit.

Habit fit

What existing behavior does this product improve or replace? If the answer is unclear, the company may be asking customers to create an entirely new routine.

Trust fit

Which person, institution, or environment gives users permission to try it? Trust can come from a retailer, a professional association, a familiar brand, or transparent safeguards.

Workflow fit

Where does the product appear at the moment of need? The closer it is to the decision, the less education and effort adoption requires.

Consequence fit

What happens if the product fails? Products used in low consequence settings can tolerate experimentation. Products affecting health, money, privacy, or reputation require stronger controls and clearer expectations.

The best products score well across all four dimensions. They improve a familiar behavior, arrive through a trusted channel, fit into an existing workflow, and make the consequences of error understandable.

For a founder, this framework may reveal that the next breakthrough is not another feature. It may be a better channel partner, a clearer use case, or a safer default setting.

For an executive, it may reveal that the obstacle to AI adoption is not employee resistance. It may be poor workflow design. People are often willing to use a tool when the tool is placed where their work already happens and when responsibility remains clear.

For a retailer, it may suggest that education, merchandising, and staff confidence are not support functions. They are part of the product experience itself.

Key Takeaways

• Design for a familiar behavior before designing for a large market. Identify what people already do and make the new capability feel like a natural extension of it.
• Treat distribution as part of the product. The channel that introduces an unfamiliar category also determines how customers interpret its safety, legitimacy, and usefulness.
• Measure repeat behavior, not just trial. Awareness and first purchases are signals of curiosity. Recurring use is evidence of value.
• Look for institutional room making. Shelf space, budgets, policies, training, and workflow changes show that a category is becoming durable.
• Build the translation layer. Users often need help understanding not what a technology can do, but when to use it, how to evaluate its output, and where it belongs in daily life.

The most consequential innovations may therefore be less spectacular than we expect. They may not arrive as dramatic replacements for everything that came before. They may appear as a new option on a familiar shelf, a new button inside a familiar workflow, or a new ritual that feels obvious in retrospect.

That is the paradox of normalization. The more successfully a new category is integrated, the less visible its novelty becomes.

A THC drink reaching a meaningful share of liquor store sales and an AI startup attracting early capital are different kinds of signals, but they point toward the same commercial truth: markets expand when unfamiliar capabilities are given familiar homes.

The future will not be adopted all at once. It will be smuggled into the present through trusted channels, ordinary habits, and small changes in where decisions happen. The companies that understand this will stop asking only how to invent the next thing. They will ask the more valuable question: What existing human ritual is ready to carry it?