As of 13 August 2026, AI can only partly analyse UK reviews of your competitors.

What to actually do

1. Hand it to a person

   The route this page recommends

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

2. Use a tool built for this

   Second choice

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. Open the review platforms and competitor pages you are allowed to use, and collect a clearly labelled UK-only sample with the competitor, platform, rating, date and review text for each entry.
   2. Remove names, email addresses, order numbers and other unnecessary personal details, then check that copying or exporting the reviews complies with the relevant platform terms.
   3. Paste the cleaned dataset into a spreadsheet, give every review a source label, and mark duplicate, non-UK or unclear entries instead of silently deleting them.
   4. Open a chatbot and paste the supplied prompt followed by the labelled review dataset, then ask it to produce the themes, competitor comparison, evidence and sample limitations in separate sections.
   5. Compare every reported theme and quotation with the original review rows, correcting counts, source labels and wording where the draft is wrong.
   6. Ask a colleague who knows the market to challenge whether the sample is representative and whether each proposed implication follows from the evidence.
   7. Turn only the checked implications into testable marketing or product actions, and keep the source dataset and analysis together so the conclusions can be updated when new reviews arrive.

   Prompt

   Analyse the UK competitor reviews pasted below. Use only the supplied reviews and metadata, and do not invent reviews, quotes, ratings, dates, customer details or competitor facts.
   
   Tasks:
   1. State the number of reviews, competitors, platforms and review dates represented, if those details are supplied.
   2. Group the reviews into recurring positive and negative themes.
   3. For each theme, give the number of supporting reviews only when it can be counted from the supplied text, and include short exact quotations with their source labels where available.
   4. Compare the competitors by theme, separating direct observations from your interpretation.
   5. Identify unmet customer needs, common purchase reasons and repeated complaints.
   6. Flag contradictory evidence, possible duplicate reviews, unclear wording, missing metadata and limits caused by the sample.
   7. End with no more than five cautious marketing or product implications. Phrase them as hypotheses to test, not established facts.
   
   Use British English. Do not claim that the reviews represent all UK customers. Do not infer a customer's location, identity, protected characteristic or motive unless the review explicitly states it. Do not recommend publishing a competitor comparison without checking the original sources and current facts.
   
   Review dataset:
   [PASTE THE REVIEWS HERE, INCLUDING COMPETITOR, PLATFORM, RATING, DATE AND REVIEW ID WHEN AVAILABLE]

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