AdsWinningStart with the decision your team needs to make
Competitor research is most useful when it resolves a choice. Should your next ad lead with convenience, price or a product demonstration? Which objection should it address? Does the first frame need to show the product or the outcome?
Give your dot the product, audience, target market and that decision. AdsWinning accepts a website, brand or keyword as the starting point. The research can then look for relevant advertising examples instead of producing an unbounded list of companies.
For example, a smart-ring brand might ask how advertisers explain recovery to people who have never owned a wearable. That is an illustrative brief, not a claim about a completed search. The useful comparison is the way the ads teach the benefit, not simply how many smart-ring ads exist.
Choose sources that can answer the question
AdsWinning supports ad research across Meta, TikTok, Google and LinkedIn. Start with the platforms relevant to the intended creative. Meta and TikTok are the default ad sources in the product; other sources can broaden a brief when their coverage fits.
Specify the market. An offer shown in one country may have a different price, delivery promise or eligibility elsewhere. If a returned example is from another market, label it as an adjacent reference instead of presenting it as local competitive evidence.
Reddit and YouTube can add language, questions and objections from supporting material. They answer a different question from a paid-ad library. A discussion about a product is context; it is not proof that the company ran an ad using those words.
Coverage can be partial. A source returning no usable matches means the search did not produce evidence under that scope. It does not establish that a competitor never advertises there.
Compare the same fields for every ad
Ask the dot to build a small evidence table from the returned research. A consistent structure makes a comparison easier to audit and stops a striking image from carrying more weight than the evidence supports.
| Field | What to record |
|---|---|
| Source and market | Advertiser, platform, source link and the market actually represented. |
| Hook and format | The visible opening claim and creative format. Quote only text actually captured. |
| Offer and action | The price, incentive or benefit shown, plus the requested next action. |
| Evidence and unknowns | Available signals, missing fields and any limitation that changes the interpretation. |
When video playback or analysis is unavailable, the dot should describe the captured frame or metadata it actually inspected. A thumbnail does not establish the full script, scene sequence or spoken claim.
Turn observations into testable patterns
Suppose several relevant ads open with the same customer objection. That gives your team a reason to investigate the objection. It does not prove that copying the wording will improve results.
Ask the dot to distinguish three layers: the observation, its interpretation and the proposed test. An observation might be that an ad shows a product being used in a small kitchen. The interpretation might be that space is an objection. The proposed test could be an original demonstration of your product fitting into a compact setup.
AdsWinning’s saved creative, findings and source evidence make that chain inspectable. Your team can challenge the interpretation without losing the underlying example. That is the practical advantage of using an ad research workspace as the input to a dots task.
Use the included follow-up to resolve the largest remaining uncertainty: “Which of these examples explicitly addresses setup difficulty, and what visible evidence supports that?” A follow-up with one purpose is more useful than asking for a second unrelated report.
Bring back a brief that survives review
Request a shortlist of relevant examples, two supported patterns, one original angle to test and a clear account of coverage gaps. If there are fewer useful examples than requested, keep the smaller set and explain the gap.
Keep the source links with the recommendations. Avoid turning a source’s reported reach, spend band or ranking into an invented profitability claim. Different currencies, markets and reporting scopes should not be combined into a single performance league table.
You can inspect saved research in the workspace. If you choose to publish it, AdsWinning provides a public read-only view. Publication is a separate action: review the material first, especially anything you put into the research prompt. A public link is readable by its recipients and later completed work can appear there.
Start with the free actual research allowance. When the comparison is useful, give the dot the creative brief workflow to develop the next deliverable.
Ready to use
Give this task to your dot
Replace the bracketed fields with your brief. This prompt requests research and a draft for review.
Use the available free browser research run at https://adswinning.com to investigate [competitor or category] in [market] for [our product and audience]. Check the allowance and stop before paid work. Use [selected ad platforms]. Compare up to five relevant ads by advertiser, source URL, observed hook, offer, format and available evidence. Keep organic supporting context separate. Identify two repeated approaches and one original angle for us to test, if the evidence supports them. State coverage gaps and unknowns. Do not infer ROAS or sales. Do not purchase or publish anything.
Questions before you start
Can I find every ad a competitor has run?
No completeness guarantee is made. Results depend on source coverage, the search scope, market and current availability. Record what was searched and what was returned.
Can my dot identify the most profitable competitor ad?
Public creative and exposure signals do not establish profit. Use them to select examples and form hypotheses; validate outcomes with your own campaign data.
How do I start without buying credits?
Use the eligible anonymous browser allowance: one Standard research run and one follow-up. Check the displayed allowance first; signed-in and connected research use account credit rules.