Home · Business · Product & Strategy · Competitor analysis

PARTLY

As of 13 August 2026, AI can only partly analyse your competitors' Google Ads.

This still needs a person who signs their name to it.

Can you do it?

15 minutesto a draft.

1 hourto something you’d act on.

Cost, all in£0

Skill neededchat-fluent

Who has to check ityou

What the alternative costsThe supplied tool data gives no priced human or software alternative for this analysis.

If this goes wrong: you mistake an observed advert for a successful strategy, change your campaigns and waste budget before the error becomes clear.

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, chat-fluent skill, and roughly 1 hour until you can act on the result.

    How to actually do it

    1. Open the Google Ads Transparency Centre and collect the competitor adverts you can find, including the advert text, format, landing-page URL and the date you observed each one.
    2. Search the same commercial terms yourself and save screenshots or copied text for adverts that appear, noting the search term and your location.
    3. Export or copy your own Google Ads auction insights, advert performance and relevant landing-page content without including unnecessary personal or customer data.
    4. Open each competitor landing page and record its offer, price if publicly shown, call to action, proof points, form or checkout route and any obvious audience qualification.
    5. Paste the collected evidence into the prompt, keeping each competitor and source clearly labelled, then ask the chatbot to produce the comparison and tests.
    6. Compare every observed claim in the draft against the saved adverts, pages and your account data, remove unsupported performance assumptions and ask a colleague to challenge the proposed tests.
    7. Choose a small set of tests that fit your budget and tracking, define the success and stopping conditions in your campaign plan, then implement them in your Google Ads account rather than copying competitors wholesale.

    Prompt

    Analyse the competitor Google Ads evidence below for a UK business.
    
    Business and offer:
    [describe the business, target customer, location and main offer]
    
    Our current campaigns:
    [paste relevant ad copy, keywords, landing pages, performance data and auction insights]
    
    Competitor evidence:
    [paste adverts, search-result observations, screenshots, landing-page text, URLs and collection dates]
    
    Analyse only the evidence supplied. Separate every point into observed, inferred or unknown. Do not claim to know a competitor's spend, bids, targeting, conversion rate or profitability unless the evidence directly shows it. Compare messaging, offers, calls to action, apparent audience, keyword intent, landing-page structure and trust signals. Identify patterns, meaningful differences, evidence gaps and risks of copying an unverified tactic.
    
    Return:
    1. A comparison table for each competitor.
    2. The strongest evidence-backed patterns.
    3. What cannot be concluded from this evidence.
    4. Gaps that would materially change the analysis.
    5. Practical campaign tests for our business, each with a hypothesis, required change, success measure and stopping condition.
    6. A short priority order based on likely learning value, not claimed competitor performance.
    
    Use plain British English. Do not invent figures, sources or competitor intentions.

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

What it gets wrong

  • AI cannot see a competitor's private spend, bids, targeting settings, conversion data or profitability.
  • It cannot establish that a visible advert is effective rather than merely active.
  • It cannot reliably distinguish a deliberate positioning choice from a temporary promotion, experiment or stale landing page.
  • It cannot decide which competitive response fits your margins, brand constraints and commercial priorities without your judgement.
  • It cannot replace clean conversion tracking and controlled testing in your own Google Ads account.

What caps this at PARTLY: private data access, verification cost and judgement under ambiguity.

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
Liability1
Effort delta2
Total7 / 10

FAQ

Can AI see how much my competitors spend on Google Ads?
No. AI can analyse adverts and public pages you provide, but competitor spend, bids, targeting and conversion data are generally private. Treat any estimate as an inference, not a fact.
Can AI find my competitors' Google Ads?
Partly. You can give it adverts found through relevant searches or public advertising records, and it can organise and compare them. It cannot guarantee that the sample is complete or show every advert a competitor runs.
Can AI tell me which competitor Google Ads are working?
No, not from adverts alone. It can identify messages, offers and landing-page patterns, but only your own properly tracked tests can show whether a tactic works for your business.
What should I give AI to analyse competitor Google Ads?
Give it labelled advert text, screenshots, search terms, landing-page URLs and collection dates, plus your own relevant campaign and auction insights data. Tell it to separate observed evidence from inference and to state what remains unknown.

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.