Competitor Ad Spy Tools: How to Research Meta Ads Competitors

Competitor ad research should produce a traceable creative hypothesis. A public advertisement can reveal its message, format and offer; it cannot prove the advertiser’s profit. Choose a tool for the fields and saving workflow you need, then evaluate the resulting idea in your own account.

What public research can establish

Save the retrieval date, country, category, advertiser identity, record URL, creative, copy, CTA and displayed status or dates. Describe a long-running record as a repeated observation, not a profitable winner: low spend, intermittent delivery and a brand campaign can all fit the same visible record.

Meta’s EU transparency announcement describes additional delivery and targeting information and a one-year archive for ads targeting the EU. That scope does not establish identical fields worldwide, exact conversion results, or competitor ROAS.

Named tools and their limits

These are functions described by Foreplay and BigSpy, not a paid-account accuracy test. Foreplay’s Lens concerns analysis of your ads; do not confuse connected-account results with competitor revenue. Provider popularity scores, estimated impressions or audience labels require a definition and method before comparison. None establishes audited profit.

A dated public record, with missing fields intact

On 2 October 2026, BigSpy’s public homepage displayed a Brawl Stars sample card with provider labels for popularity, impressions, duration and a last-seen date rendered as 26-08-24. This is a vendor showcase observation, not a verified Meta Ad Library record or evidence of a successful advertiser.

For usable inspiration, obtain the actual creative and destination from an accessible record. Keep the vendor sample separate from that evidence. Never fill absent fields using a similar brand name.

Turn research into an original brief

Hypothetical brief for a fictional puzzle app: test whether a short demonstration of one solvable challenge produces more activated players than a general entertainment montage. This is our proposed hypothesis, not an inference about Brawl Stars performance. Use your own gameplay, permissions and truthful product facts.

• Baseline: current montage. Treatment: challenge demonstration. Keep offer, destination, geography and measurement consistent.
• Define activation before launch, deduplicate installs and measure activation within a declared window. Record spend and contribution from your own business.
• Agree the decision rule and affordable loss limit before inspecting outcomes; inconclusive evidence leaves the hypothesis unresolved.

Use the creative-testing framework to structure the experiment. Research can reduce repetition in ideation; it does not guarantee lower CPA or higher ROAS.

Keep the economic calculation honest

Illustrative calculation, not observed results: if your own comparable ROAS changed from 1.5 to 3.5, the ratio would be 3.5 ÷ 1.5 = 2.33 times and the relative increase 133.3%. This would still need consistent revenue, spend and attribution definitions, and an experiment to attribute the change to research.

Make the routine manageable

• Choose a small relevant advertiser set and one question; review accessible records weekly as an operating choice.
• Separate observed inspiration, possible differentiation and unanswered questions. Recheck the destination when preparing your brief.
• Evaluate the tool on records recovered, fields documented and time saved; evaluate campaigns on your own outcomes.

Prepare the resulting campaign draft in NovaStorm, review the creative, destination and proposed test, and check that it remains paused before launch. Competitor-library collection and your experiment measurement remain separate workflows.