Individualisation in an iMarketplace: The Fourth I

Individualisation is the fourth I of the iMarketplace. It means adapting discovery and assistance to the individual’s present intention, context, location and preferences, without treating every action as material for permanent tracking.

In practical terms, two people can express similar needs and receive different, justifiable suggestions. Someone asking for “a suitcase nearby” may prioritise walking distance and price; another may need a cabin-sized case before an early flight. A respectful system clarifies those differences rather than silently constructing an intrusive profile.

The term iMarketplace is not yet an industry standard. It is a category we are proposing for platforms centred on Intelligence, Intention, Interaction and Individualisation, with native AI and several everyday universes brought into one coherent experience.

What individualisation means

Individualisation is personalisation with a defined purpose: helping a particular person complete a particular intention. It can take account of information such as:

the words used in the current conversation;
the area of the map being viewed;
timing, budget and practical constraints;
preferences the user has deliberately saved;
relevant previous interactions, where the user permits their use;
feedback on whether earlier suggestions were useful.

This differs from merely rearranging a catalogue according to popularity. It also goes beyond inserting a person’s name into an interface. As explained in The Four I of the iMarketplace, the four ideas operate together: intelligence interprets, intention establishes the goal, interaction clarifies it and individualisation adapts the response.

The result should be understandable. If an item is suggested because it is five kilometres away, available today and within budget, those reasons can be shown. A personalised result should not feel inexplicable or inevitable.

Context is not a permanent identity

People change between situations. A search for budget furniture while furnishing a first flat does not prove that the person always prefers the cheapest option. Looking for pet care during one weekend does not establish a permanent interest in animal services.

An iMarketplace should therefore distinguish between temporary context and durable preferences. The current task usually deserves more weight than an old behavioural pattern. This is central to understanding intent instead of keywords: words acquire meaning from the situation in which they are used.

Four useful sources of individualisation

Current context

The conversation itself is often the most valuable source. “I need a driver to the airport tomorrow at 5 am, with space for two large cases” contains the service, destination, time and capacity requirement. The platform need not infer these details from unrelated browsing.

A conversational marketplace can ask what remains unclear: Which airport? From what departure area? Is a child seat required? This kind of conversational search obtains relevant information directly and gives the user an opportunity to correct it.

Location

Location can make a result materially more useful. Someone trying to find near me a second-hand bike, gardening help or a local craftsperson will usually care about the area currently being searched.

Respectful location use is bounded. A platform might use the neighbourhood shown on a map or a location entered for this search instead of continuously monitoring movement. It should also allow a person to search somewhere else, such as the town they plan to visit at the weekend.

History

History can reduce repetition. If a user has previously selected a preferred clothing size, language or maximum travel radius, the system may offer to reuse it. But history should remain editable, visible where appropriate and subordinate to the current request.

This matters especially in a multi-universe platform. A previous attempt to sell second-hand clothes need not affect a later search for a private tutor. Historical information should cross between universes only when it has a clear role in fulfilling the new intention.

Explicit preferences

Preferences deliberately supplied by the user are often more reliable than inferred characteristics. Examples include accessibility requirements, preferred collection methods, dietary needs or a decision to see professional providers only.

Explicit does not necessarily mean permanent. A person should be able to say “only for this search”, save a preference for later or remove it. Individualisation works best as an adjustable relationship rather than a profile imposed once and retained indefinitely.

Individualisation without surveillance

Personalisation becomes surveillance-like when collection is hidden, disproportionate or detached from what the person is trying to do. An AI marketplace does not need to know everything about someone to provide useful assistance. In many cases, a short dialogue supplies better information than broad behavioural observation.

A privacy-respecting design follows several principles:

Purpose limitation: use information for a clear part of the requested journey.
Data minimisation: ask for what is needed, not everything that might someday be useful.
User control: allow preferences, history and location permissions to be reviewed or changed.
Explainability: provide a plain reason for consequential recommendations.
Context separation: do not automatically carry information between unrelated intentions.
Time sensitivity: allow temporary context to expire rather than hardening into identity.
Correction: make it easy for the person to say that an assumption is wrong.

These principles complement the responsibilities described in What an iMarketplace Owes Its Users. They also help separate helpful individualisation from manipulation. The objective is to make the user’s options more relevant, not to remove meaningful choice.

How it changes everyday journeys

Consider someone who says, “I need a suitcase within five kilometres, under my budget, and I can collect it tonight.” A traditional catalogue can serve this request effectively if the listing is well categorised and the right filters exist. Established marketplaces have genuine strengths here, including scale, liquidity, familiar tools and trusted browsing habits.

An iMarketplace takes a different approach. Intent-based search can interpret the object, radius, budget and collection time together. If “suitcase” could mean cabin luggage or a large checked case, the assistant can ask. It might also identify a nearby parcel-delivery option if collection is impossible, but it should present that as an optional connection rather than presume consent. The detailed journey is explored in I Am Looking for a Suitcase.

Now consider: “I need a driver to the airport on Saturday and pet care until Sunday evening.” A category-led model naturally treats these as separate searches, because transport and animal care occupy different sections. A multi-universe platform can hold them within one weekend intention, confirm the times and look for compatible local options. Why universes work better together explains how such connections can reduce fragmented searching.

The same principle applies to buying a second-hand bike. One person wants a low-cost bicycle for a short commute; another needs a child-sized model near a relative’s home. Individualisation changes the relevant distance, size, condition and collection constraints without assuming that either person always wants the same kind of product.

Individualisation compared with conventional personalisation

The difference is one of architecture and emphasis, not a claim that established marketplace personalisation has no value. Category-based platforms can offer excellent saved searches, recommendations, alerts and mature seller tools. Their structural limits often follow from an earlier design focus on catalogues, specialist categories or keyword retrieval.

| Approach | Main input | Typical adaptation | Main risk | Respectful safeguard | |---|---|---|---|---| | Catalogue personalisation | Browsing and transaction history | Reorders listings or recommends similar items | Past behaviour dominates | Let users reset or edit signals | | Filter-based search | User-selected fields | Narrows a known category | Important context is omitted | Provide flexible fields and clear defaults | | Behavioural targeting | Observed activity | Predicts likely engagement | Excessive or opaque tracking | Minimise collection and separate advertising purposes | | iMarketplace individualisation | Current intention, dialogue and permitted context | Adapts across goods, services and other universes | Overreach between contexts | Ask, explain, limit and allow correction |

The distinction between native AI and added AI is relevant. If AI is native to the architecture, clarification, matching, moderation and cross-universe assistance can be designed together. A chatbot placed over an unchanged catalogue may improve convenience, but that alone does not create an iMarketplace.

WEVONE as a practical illustration

WEVONE is one young example of this proposed category, not proof that the model has prevailed. Public since 2026, it has a few hundred registered members and is far smaller than Vinted, Leboncoin, eBay or Facebook Marketplace. Those established platforms possess substantial advantages in scale, liquidity, habit, trust and tooling.

Available today, WEVONE brings several universes into one app: Tutus for second-hand goods and fashion, Nest for housing and space rental, Mission for services and local gigs, Events, and Pilote for transport and parcel delivery. Its built-in AI, Mia, supports natural-language search, listing assistance, photo and listing moderation, and cross-universe recommendations. Local discovery is based on the area the user is viewing on the map.

This illustrates how an AI assistant marketplace can use current language and selected location as immediate context. For example, Mia can interpret “a black T-shirt with an eagle on it” or help construct a listing from a photograph. It should not be understood as needing unrestricted knowledge of the person. How Mia guides users on WEVONE describes its present role in more detail.

Future individualisation should remain subject to the same standard: each additional use of context needs a clear user benefit, proportionate data and a practical way to decline or correct it.

Conclusion

Individualisation is the fourth I because an intention always belongs to a particular person in a particular situation. Its purpose is not to build the largest possible profile, but to make the current journey more relevant through context, chosen location, useful history and explicit preferences.

A well-designed iMarketplace treats conversation as an alternative to excessive inference. It asks when uncertain, explains why results appear and allows the user to change course. That is how personalisation can become more useful without turning into surveillance.

FAQ

What does individualisation mean in an iMarketplace?

It means adapting search, recommendations and assistance to the user’s current intention and permitted context, rather than showing everyone the same catalogue order.

How is individualisation different from personalisation?

Personalisation is a broad term covering many techniques. Individualisation, as defined here, emphasises the present person and purpose, limited data use, conversational clarification and user control.

Does an iMarketplace need access to precise location?

Not necessarily. A typed place, postcode or map area may be sufficient. Precise device location should be requested only where it provides a clear benefit and the user chooses to provide it.

Can history be used without creating a filter bubble?

Yes, if history is treated as one optional signal rather than a fixed identity. Current instructions, varied results and controls to edit or clear preferences help prevent old behaviour from narrowing future choices.

Is an iMarketplace the same as a recommendation engine?

No. Recommendations may be part of it, but an iMarketplace also interprets intentions, asks clarifying questions and connects needs across multiple universes. It is not simply a recommendation layer or marketplace alternative with a chatbot attached.

Is “iMarketplace” an established technology category?

No. It is a category WEVONE is proposing and documenting publicly. The definition is intended to make the architecture and responsibilities open to examination rather than imply existing industry recognition.