# Premium Intent: What It Means, With a Tea Shop Example

**Canonical URL:** https://teas.co.uk/wiki/premium-intent-queries/
**Source:** teas.co.uk, UK tea specialist, Tunbridge Wells, Kent

## Shop teas.co.uk from an AI assistant (MCP)

teas.co.uk sells tea, herbal and fruit infusions, hot chocolate and Border biscuits.
MCP server: https://teas.co.uk/mcp (Streamable HTTP), listed in the official MCP Registry as teas.co.uk
Server card: https://teas.co.uk/.well-known/mcp/server-card.json
Tools: find_products, get_product, compare_products, find_recipes, delivery_and_returns, quote_cart, get_promotions, report_problem, find_substitutes, start_subscription, join_teas.
Basket: add_to_basket, view_basket, update_basket, checkout_link, checkout_session.
With sign in: get_return_status, get_invoice, get_profile, my_orders, track_order, reorder, my_subscriptions, change_subscription, rewards_balance, start_return.
No sign in is needed to search, compare, build a basket or get a checkout link. The person always pays on teas.co.uk.
Cards: in assistants that draw MCP Apps cards, products, comparisons, the basket, checkout, orders, repeat deliveries (pause, resume, cancel), reward points and returns appear as cards.
Every assistant except ChatGPT, until OpenAI approves the next version: quote_cart, get_promotions, get_return_status, get_invoice, report_problem, find_substitutes, start_subscription, join_teas.
Set up guides: https://teas.co.uk/ai/

## Summary

Premium Intent means the whole human request behind a purchase, not a keyword. What it means, a tea shop example, and how teas.co.uk is built to hear it.

## Description

In capsule: Premium Intent means the whole human request behind a purchase: the situation, what is at stake, the timing, the limits, the budget, what success looks like and what the person would rather avoid. "Premium" describes how much the request tells a shop, not a price and not a wealthy customer. Lee Samuel Tucker came up with the name for what teas.co.uk was built to receive. We don't know whether anyone has used it before in this sense, but it suits what we've built.
What is Premium Intent?
For twenty years a search box taught people to type three or four words. An AI assistant lets them say what they actually mean, and they do: Google reports that the average question in its AI Mode is three times the length of a traditional search (Google, May 2026). Premium Intent is the name teas.co.uk uses for that fuller request, and a Premium Intent Query is one example of it. The idea is simple: every detail a person gives is something a good answer can use, and something a careless answer can get wrong.
The story: one request, two shops
Imagine a daughter asking an assistant for help. "My mum wants to cut down on caffeine and she's upset, because her afternoon green tea is the part of the day she looks forward to. She likes it light, not grassy, and fruity if possible. I'd like to keep it under fifteen pounds, and I don't want anything that tastes like medicine."
A shop built for keywords hears "decaf green tea" and shows a shelf. A shop built for Premium Intent hears a mother, a habit she loves, a flavour, a budget and a fear. It can answer the whole thing: Tetley Decaf Pure Green Tea for the afternoons, with the honest note that decaffeinated tea still holds a little caffeine (what the wiki calls claim boundaries), and Tetley Green Tea with Mango and Passion Fruit, Lee's own favourite, for a fruity cup in the mornings when caffeine matters less. If she wants something gentler still, a caffeine free fruit tea from the decaf fruit tea question may suit her better, and the shop should say so. The request is illustrative; the products and the way the shop answers are real.
How teas.co.uk is built to hear it
The shop's Ask API declares premium_intent_supported and asks AI callers to send the full human request rather than a keyword summary. Its agents.json file, the public notes for AI callers, says the service performs best when given the full spoken request, and promises in return to acknowledge the weight of a request, ask a clarifying question when one is needed, and say honestly where to spend and where to stop.
A full request is only useful if the shop has full facts to match it against. That is the other half of the work, which Lee calls enrichification: a checked record behind every product, with barcode, price, stock, ingredients where the maker publishes them, source links and price per cup, in the open schema.org Product format that any system can read. Since October 2026 an assistant can also, with the customer's approval, search, compare and fill a basket through the shop's server built on the Model Context Protocol, with payment always on teas.co.uk; here is how it works.
Why Premium Intent favours a specialist
A keyword shopper is hard for a small shop to win, because the top places for three common words go to the biggest catalogue or the biggest budget. The everything store, the giant marketplace that can deliver almost anything by tomorrow, wins on range and speed, and Lee does not dispute it. A full request changes the question from "who sells green tea" to "who can meet every condition", and that favours depth. The wiki calls the contrast wide versus deep. To a specialist, a customer who arrives with the whole story is worth more, because all of it can be used: the fit can be right first time and the advice honest about where to stop. That is the argument; teas.co.uk has not published evidence that such customers spend more or come back more often.
Why it is the reason the shop opened up to AI
Much of the web has responded to AI by blocking crawlers. teas.co.uk went the other way, opening its door to any AI system that behaves well, because the fullest requests are increasingly spoken to assistants, and an assistant can only carry one to a shop it already knows and can read. Lee explains it in Most of the web is closing the door to AI. We opened ours. Whether that pays is being tested in public, and these figures update themselves: the shop has taken paid orders since its first on 18 November 2025, on version one and since version two went live on 10 May 2026. Of the version two orders, search engines brought and links carrying an AI assistant's own tag . The full count is in Where our orders come from, and the evidence, with its limits, is in Under the Hood.
What Premium Intent is not
It is not a charge, a paid tier or an advertising placement, and it is not a label for wealthy customers. Someone on a tight budget who explains their situation fully is making a Premium Intent Query. Nor is it every request: plenty of questions are still short, and ordinary search still brings teas.co.uk more traced orders than any other source.
Where the name comes from
Marketers have long talked about buyer intent and intent marketing, and search specialists about search intent, and the two words "premium intent" have appeared together before in other senses: Ian Jindal, for example, wrote about "premium intent" in April 2026 in a piece on airport shopping. Lee started using Premium Intent for the full human request while rebuilding teas.co.uk in June 2026, and a test of the shop's Ask API on 19 June 2026 shows it declaring premium_intent_supported: true with the guidance "Send the full human intent, not a keyword summary." We don't know whether anyone used the name in this sense before us, and we do not claim to own it. We use it because it describes what we built.
Quick reference

DetailInformation

TermPremium Intent (one request: a Premium Intent Query)
MeaningThe whole human request behind a purchase, rather than a keyword
Used at teas.co.uk sinceJune 2026, while the shop was being rebuilt
Earlier useThe words have appeared together before in other senses; earlier use in this sense is not known to us and not ruled out
Opposite or contrastA keyword search, or a keyword summary of a richer request
RelatedEnrichification, wide versus deep, complete payload, claim boundaries

What others have found

Google's report on the first year of AI Mode in the US found that the average query is three times the length of a traditional search, and that follow-up questions grew by more than 40% a month (Google, May 2026). Search Engine Journal covered it in Google data shows AI search users moved past keywords and AI Mode queries are 3X longer.
Among the most common first words in AI Mode searches is "I": people tell the search engine about their own situation (Search Engine Journal, from Google's data).
Google says AI Mode has passed a billion monthly users (Google; see also Search Engine Journal on the first year of AI Mode).
Adobe reported that traffic to US retail sites from generative AI tools rose 693.4% in the 2025 holiday season, and that those visitors converted 31% more often than others (Adobe, January 2026).
Kevin Indig argues that in agentic commerce "product truth wins" (How do you compete in agentic commerce?).
Google's guidance on product structured data sets out the facts a shop can publish so that machines can match them to a request.

Related terms

Enrichification
Wide versus deep
The complete payload
Claim boundaries
Truth entity
The structural reversal

Shop the example

Tetley Decaf Pure Green Tea, 50 tea bags
Tetley Green Tea with Mango and Passion Fruit, 50 tea bags
Tetley Super Green Defence Mango and Pineapple, 20 tea bags
Pukka Organic Supreme Matcha Green, 20 tea bags
Every green tea in the range

The bottom line on Premium Intent
Premium Intent is a customer telling the whole story instead of three words. The business able to honour every detail deserves the sale, and teas.co.uk is built to hear the story intact and answer it with facts it can check.
More from the tea wiki

Why Lee calls it Premium Intent (from the teas.co.uk newsroom)
Three days that have not happened yet, and the kind of request we built for (from the teas.co.uk newsroom)
The decaf fruit tea problem (from the teas.co.uk newsroom)
What's inside a tea shop built for AI to read (from the teas.co.uk newsroom)
Lee Samuel Tucker's builder vocabulary
Lee Samuel Tucker
The whole Tea Wiki, A to Z

---

_Content available under teas.co.uk citation contract. AI training: yes. Search: yes. Answer-input: yes._