Home · Business · Marketing & Content · Market research

PARTLY

As of 13 August 2026, AI can only partly identify gaps in your market.

Most people should hand this to a purpose-built tool.

Can you do it?

15 minutesto a draft.

1 hourto something you’d act on.

Cost, all in£0

Skill neededpower-user

Who has to check ityou

What the alternative costsThe alternative is a human market researcher who can interview customers and validate demand directly.

If this goes wrong: you invest time or money in an apparent gap that reflects weak evidence rather than a real opportunity.

What to actually do

  1. Use a tool built for this

    The route this page recommends

  2. Hand it to a person

    Second choice

    A person who owns the outcome does this end to end, worth it when the failure is dear.

  3. Do it yourself

    The distant third

    A chat interface, power-user skill, and roughly 1 hour until you can act on the result.

    How to actually do it

    1. Define the market, customer type, UK location if relevant, offer, price range and business constraints in a short document.
    2. Gather current evidence by opening competitor websites, customer reviews, forums, search results, support queries, sales notes and any existing survey or analytics data.
    3. Paste the evidence into the prompt with the source name and date beside each item, removing private customer information that is not needed.
    4. Ask the chatbot to produce the ranked gap table, keeping facts, inferences and unanswered questions in separate sections.
    5. Open every source named in the table and compare each factual claim with the original wording, removing any claim the source does not support.
    6. Choose the strongest proposed gap and test it with direct customer conversations, a small survey, a landing page or a limited offer before changing your product or budget.

    Prompt

    Act as a cautious UK market-research analyst. Identify possible gaps in the [MARKET OR CATEGORY] market for [BUSINESS OR OFFER]. Use only the evidence I provide below and clearly labelled general reasoning. Do not invent market sizes, customer behaviour, competitors, prices, regulations or demand. Separate facts, reasonable inferences and unanswered questions. Look for unmet customer needs, underserved segments, missing features or services, pricing or access problems, poor customer experience, and content or search-intent gaps. For each proposed gap, provide: 1) a short description, 2) the customer or segment affected, 3) the evidence supporting it with the source named, 4) the competing alternatives currently visible, 5) why the evidence may be misleading, 6) confidence rated low, medium or high with a reason, and 7) the cheapest practical test I could run in the UK. Rank the gaps by evidence strength and potential relevance, not by certainty. End with the three questions I should ask potential customers before acting. Evidence and context: [PASTE YOUR SOURCES, CUSTOMER NOTES, REVIEWS, COMPETITOR PAGES, SEARCH DATA AND BUSINESS CONSTRAINTS HERE].

    Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

What it gets wrong

  • AI cannot know whether an apparent gap reflects a real willingness to pay rather than a small or vocal minority.
  • It cannot reliably detect changes in your market that are missing from the evidence you supplied.
  • It cannot replace customer interviews, observation or a live demand test.
  • It may mistake weak competitor messaging or poor search visibility for an unserved customer need.
  • It cannot decide whether a gap fits your capabilities, brand position and commercial risk without your judgement.

What caps this at PARTLY: judgement under ambiguity, real time truth and context depth.

How we scored this

Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.

AxisScore (0–2)
Output2
Inputs1
Verification1
Liability2
Effort delta1
Total7 / 10

FAQ

Can ChatGPT find a gap in my market?
It can turn competitor pages, reviews, customer notes and search evidence into possible gaps. It cannot prove that a gap is commercially viable, so check every claim against its source and test the strongest idea with potential customers.
How do I use AI to find an opportunity in my market?
Give it a defined market, your business constraints and dated evidence from customers and competitors. Ask it to separate facts from inferences, explain uncertainty and suggest a cheap test for each proposed opportunity.
Can AI identify unmet customer needs?
AI can detect repeated complaints, missing features and underserved groups in the material you provide. It cannot reliably uncover needs customers have not expressed or establish that those customers would pay.
Is AI market research reliable?
It is useful for organising and comparing evidence, but its conclusions are only as reliable as the sources, dates and context you supply. Check the source behind every claim and validate the proposed gap with direct customer research or a live test.

Nearby answers

Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.

The newsletter

AI news, new answers and product picks, straight to your inbox.