An ad spy tool shows you the ads your competitors are running right now: the hooks, the formats, the offers, and how long each ad has been live. Free ad libraries from Meta, TikTok, Google, and LinkedIn give you the raw feed; paid ad spy tools add search, filters, alerts, and spend estimates on top of it.
Spying is the easy part. Anyone can scroll the Meta Ad Library for an hour and screenshot pretty ads. The operators who get paid for this do something different: they treat competitor creative as performance data, extract the patterns, and turn them into briefs. This guide compares the best ad spy tools in 2026, free and paid, then shows the workflow that turns what you find into ads that convert. It sits alongside our deeper playbook on how to analyze competitor ads.
The best ad spy tools in 2026 compared
| Tool | Platforms | Price | Best for |
|---|---|---|---|
| Meta Ad Library | Facebook, Instagram | Free | Every active Meta ad, searchable by brand or keyword |
| TikTok Creative Center | TikTok | Free | Top ads by region and industry with engagement data |
| Google Ads Transparency Center | Google, YouTube | Free | Search and YouTube ads by advertiser |
| LinkedIn Ad Library | Free | B2B competitor creative | |
| PiPiAds | TikTok | From $77/mo | TikTok product research with ad-level filters |
| Foreplay | Meta, TikTok | $79 to $249/mo | Save-to-folder swipe files and team workflows |
| BigSpy | Meta, TikTok, YouTube, more | From $9/mo | Cheapest cross-platform database |
| AdSpy | Facebook, Instagram | $149/mo | Huge historical Meta database with Boolean search |
| Minea | Meta, TikTok, Pinterest | Free tier, from $49/mo | Dropshipping product and ad research |
| Atria | Meta, TikTok | $99 to $399/mo | Creative analysis and brief workflows |
| Superscale AI | Meta, TikTok | Free plan (1,000 credits, no card) | Agent that researches competitor ads, then produces and publishes your own |
Three tiers hide inside that table, and knowing which tier you need saves real money.
Tier 1: the free ad libraries. Every major platform is forced to show its active ads. This is the ground truth every paid tool builds on.
Tier 2: paid spy databases. Foreplay, AdSpy, BigSpy, PiPiAds, Atria. They index the libraries, then add Boolean search, saved folders, alerts, and run-length data. Worth it once you research competitors weekly instead of occasionally.
Tier 3: AI agents. Tools that do not stop at showing you ads. Superscale AI researches competitor ads in the Meta Ad Library and TikTok Creative Center, extracts winning patterns, writes scripts, produces the video and static ads, resizes them to 9:16, 1:1, and 16:9, publishes to Meta, TikTok, Instagram, and Google Ads, and iterates on whatever wins. Spying becomes an input to production instead of a separate research chore. The category is ranked in the best AI marketing agents.
Free ad spy tools: the official ad libraries
If you just want to know how to find competitor ads without paying for software, start here: the official libraries cover more than most paid tools admit.
Start here. At small scale the free libraries are genuinely enough, and you should exhaust them before paying anyone.
Meta Ad Library (the free Facebook ad spy)
The Meta Ad Library shows every active ad on Facebook and Instagram, searchable by advertiser name or keyword. No performance data outside political ads, but the metadata is rich: start date, active variations, placements, and formats. Run length is your performance proxy. Nobody pays to run a losing ad for four months. Our Meta Ad Library guide walks through the exact search workflow, filters, and what each signal means.
TikTok Creative Center
TikTok's free library goes one step further than Meta's: the Top Ads section actually ranks ads by performance within region and industry, with CTR and engagement benchmarks attached. For hook research it is the single best free resource in paid social, because you see what holds attention in the first two seconds, ranked.
Google Ads Transparency Center
Search and YouTube ads by advertiser. Less useful for creative patterns (search ads are text), more useful for offer and messaging research: what claims, prices, and guarantees competitors push in their highest-intent channel.
LinkedIn Ad Library and X Ad Repository
Both exist, both are thinner. LinkedIn's is worth a monthly scan if you sell B2B. X's repository is mostly relevant for compliance research.
TikTok ad spy: how to research TikTok ads
TikTok deserves its own section because the research motion differs from Meta. Meta creative rotates slowly; TikTok creative burns out in days, so recency matters more than run length.
The free path: TikTok Creative Center, filtered to your region and industry, sorted by CTR. Watch the first two seconds of the top 20 ads and write down every hook mechanic you see. That list is a week of briefs.
The paid path: PiPiAds (from $77/month) indexes TikTok ads with filters the Creative Center lacks, like ad duration, engagement rate, and landing page. It is built for ecommerce and dropshipping product research, so if that is your world, it is the specialist pick. If you run dropshipping video ads, this is where the winning product formats surface first.
The agent path: Superscale researches the TikTok Creative Center automatically as part of its competitor research, then produces the TikTok-native versions in the same system.
Ad spy tools vs ad intelligence software
The terms get used loosely. Ad spy tools show you creatives; ad intelligence tools add spend and performance context; and the newest layer, agentic ad intelligence, runs the competitor ad analysis for you and feeds it straight into production.
"Ad spy" and "ad intelligence" get used interchangeably. They should not be. Spying is one signal layer. Ad intelligence software combines four, and the difference explains why two teams can look at the same competitor ads and only one ships winners.
| Layer | Source | What you read | Cadence |
|---|---|---|---|
| 1. Your account performance | Native platforms, analytics | CPA, CTR, ROAS, fatigue curves | Daily check, weekly readout |
| 2. Competitor signal | Ad libraries, spy tools | Run length, format mix, hook patterns | Weekly scan |
| 3. Category signal | Trend reports, creative benchmarks | Emerging patterns not yet mainstream | Monthly review |
| 4. Predictive signal | AI scoring models | Predicted CPA bands before production | Per brief |
Layer 2 is the ad spy tool. Layers 1 and 2 together answer the question that matters: which competitor patterns should we test, given what already works in our own account? Layer 1 is its own discipline, covered in our creative analytics guide, and the search-side equivalent of this whole stack is PPC competitor analysis.
The output of real ad intelligence is not a swipe file. It is a testable hypothesis with a predicted cost attached: "the discussion-thread static layout outperforms poster-style for tax audiences on Meta, we expect CPA below control, and we expect the pattern to saturate within two quarters." That is a brief. A folder of screenshots is not.
How to score the ads you find
The goal of scoring is turning raw creatives into ad angles and concrete Facebook ad ideas you can brief today.
The biggest failure mode in ad spying is volume without structure. A hundred saved ads with no rubric produce nothing. Fifty ads scored against a fixed rubric produce a ranked test queue. This is the 0 to 15 scoring model we use:
| Dimension | 0 points | 1 point | 2 points | 3 points |
|---|---|---|---|---|
| Pattern recency | Everywhere for 2+ years | Mainstream in your vertical | 2 to 3 leaders using it | Emerging, under 3 brands |
| Competitor proof | No strong competitor runs it | One competitor, briefly | One competitor scaling 30+ days | Multiple competitors scaling |
| Brand fit | Needs a voice you cannot deliver | Significant stretch | Minor adjustment | Native to your voice |
| Production cost | $5,000+ | $1,000 to $5,000 | $200 to $1,000 | Under $200 |
| CPA prediction | 50%+ above control | Within 20% of control | 20 to 40% below control | 40%+ below control |
Thresholds: 11+ ships immediately, 7 to 10 goes in the B-tier test queue, 6 or below gets archived or revised. Two notes from running this rubric in practice. First, competitor proof means scaling duration, not prettiness: a plain ad running 60 days beats a beautiful ad that vanished in a week. Second, treat any predictive CPA score as directional, not bankable. The 2026 scoring models are useful for ranking briefs, not for skipping tests.
A good finding also has to be replicable (you can produce the pattern again), attributable (it traces to a measurable outcome), differentiable (it is distinct from adjacent patterns), and time-bound (you have a view on when it saturates). If a swipe-file entry fails those four checks, it is inspiration, not intelligence.
A weekly ad spy workflow
This is the cadence that turns research into shipped ads. Total cost: 6 to 8 hours a week across two roles, less with an agent handling production.
| Day | Activity | Owner | Time |
|---|---|---|---|
| Monday | Account performance readout | Media buyer | 30 min |
| Monday | Competitor scan (libraries or spy tool) | Creative strategist | 60 min |
| Tuesday | Draft briefs from scored patterns | Creative strategist | 90 min |
| Tuesday | Score briefs against the rubric | Strategist + AI tool | 30 min |
| Wednesday | Top-scored briefs go to production | Designer or AI agent | 4 to 8 hrs |
| Thursday | Human review and brand check | Creative director | 60 min |
| Thursday | Launch | Media buyer | 60 min |
| Friday | Early performance read | Media buyer | 30 min |
| Monthly | Category trend review | Strategist + leadership | 2 hrs |
The Wednesday row is where the economics changed. Production used to be the bottleneck that capped the whole loop at two or three tests a week. With an agent producing static ads and video from the scored briefs, the constraint moves back to judgment, which is where it belongs. Taxfix ran exactly this loop across 4 teams and 3 languages: 200+ ads shipped, +45% CTR, and a 20% to 21% CPA reduction.
Common ad spy mistakes
Four pitfalls account for most wasted research time:
- Copying instead of decoding. Cloning a competitor's ad gets you a worse version of their creative and a possible copyright problem. Extract the mechanic (the hook type, the format, the offer structure) and rebuild it in your voice.
- Confusing volume with intelligence. 100 saved ads without a rubric lose to 50 ads with structure, every time.
- Ignoring your own account. Competitor signal without your own performance data is half a picture. The pattern that works for them may already be fatigued for your audience.
- Treating predictive scores as results. Scoring ranks the queue. Testing decides. No 2026 model replaces the live read.
How to choose an ad spy tool
Four questions settle it faster than any feature list:
How often do you research? Occasional checks: stay free. Weekly cadence: a paid database earns its fee through search and alerts alone.
Which platform matters most? Meta-first teams want the Ad Library plus Foreplay or AdSpy. TikTok-first teams want the Creative Center plus PiPiAds. Cross-platform on a budget: BigSpy.
Who consumes the research? Solo founders need a feed. Teams need folders, comments, and brief handoffs, which is where Foreplay and Atria justify their price.
Does the research connect to production? This is the question most buyers skip. A spy tool that ends at a swipe file leaves the expensive part (making the ads) untouched. An agent like Superscale closes the gap: research, scripts, production, publishing, and performance reads in one loop, which is what the agentic platforms category is converging on. One more thing worth knowing before you commit to a paid database: public ad library access has been tightening, so tools that only wrap the libraries carry platform risk. Tools with their own data layer or a production loop age better.
FAQs about ad spy tools
What is an ad spy tool?
An ad spy tool shows you the ads your competitors are running right now: the hooks, formats, offers, and how long each ad has been live. Free ad libraries from Meta, TikTok, Google, and LinkedIn cover the basics. Paid tools add search, filters, alerts, and spend estimates on top.
What is the best free ad spy tool?
The Meta Ad Library. It shows every active ad on Facebook and Instagram, searchable by advertiser or keyword, at no cost. Pair it with the TikTok Creative Center and the Google Ads Transparency Center and you cover the three biggest ad platforms without paying anything.
What is the best ad spy tool for TikTok?
The TikTok Creative Center is free and shows top-performing ads by region and industry with engagement data. For deeper TikTok research, PiPiAds (from $77/month) adds TikTok-specific filters like ad duration and engagement rate. Superscale AI researches the TikTok Creative Center automatically as part of its competitor ad research.
Is there a free way to spy on Facebook ads?
Yes. The Meta Ad Library at facebook.com/ads/library shows every active ad running on Facebook and Instagram for free. You cannot see performance data or spend outside political ads, but run length is a strong proxy: an ad that has been live for months is almost certainly profitable.
How much do ad spy tools cost?
Free to roughly $399 per month. The official ad libraries cost nothing. Paid databases run from $9/month (BigSpy) through $77 (PiPiAds) and $149 (AdSpy) up to $249 and beyond for team plans like Foreplay. Superscale AI includes competitor ad research on its free plan with 1,000 credits and no card required.
Are ad spy tools legal?
Yes. Ad libraries exist because of platform transparency policies, and paid spy tools index publicly visible ads. What you do with the research matters: copying a competitor's ad outright can violate copyright, but studying formats, hooks, and offers to inform your own creative is standard practice.
What is ad intelligence software?
Ad intelligence software is the broader category that ad spy tools sit inside. Spying shows you what competitors run. Intelligence adds your own account performance, category trends, and predictive scoring, then turns those signals into testable creative briefs. Superscale AI automates that loop end to end.
Can AI analyze competitor ads automatically?
Yes. AI agents like Superscale AI pull competitor ads from the Meta Ad Library and TikTok Creative Center, extract the patterns that repeat across winners, turn them into scripts and finished video and static ads, then read performance back and iterate on the winners.
From spying to shipping
This is the step where ad spying becomes agentic creative research: AI research for ad creatives that ends in a launched test, not a swipe file.
The gap between teams that spy on ads and teams that win with the research is not access. Everyone sees the same libraries. The gap is the loop: scan, score, brief, produce, launch, read, repeat, every week. Superscale AI runs that loop as one system, from competitor research in the Meta Ad Library and TikTok Creative Center through production and publishing to performance reads. The free plan includes 1,000 credits with no card required; ad account integrations for Meta, TikTok, and Google unlock on the /mo99/mo Pro plan.
Related reading
- how to analyze competitor ads
- the best AI marketing agents
- Meta Ad Library guide
- dropshipping video ads
- creative analytics guide
