Best ad spy tools UK: an honest comparison

Search "best ad spy tools" and a striking number of the results are published by vendors who appear in their own rankings. That is not a conspiracy, it's just incentives. We capture ads for 164 UK brands across Google, Meta, TikTok and LinkedIn, and have AI-qualified 50,982 creatives, so we can do something generic reviewers can't: test what these tools are actually up against. This guide covers where the free official libraries are genuinely sufficient, where paid tools earn their fee, the structural ceilings we've measured in our own data, and — since we sell in this market too — exactly what we don't do.

Start here: you may not need a paid tool at all

The single most under-reported fact in this category is how much the free official libraries now give you. Vendor listicles rush past them in a paragraph because there's no commission in "use the free thing". For a large share of UK marketers doing competitor research — checking what three or four named rivals are running right now — the free route is not a compromise, it's the correct answer.

What the free libraries cover

Library	Covers	Genuinely good at	Hard limit
Google Ads Transparency Center	Search, Display, YouTube, Shopping	Every verified advertiser, free, no login, UK region filter	Retrieval ceiling around 900 creatives per advertiser
Meta Ad Library	Facebook, Instagram	Complete for live ads, full-text search of ad copy	No memory — ads vanish when they stop
TikTok Commercial Content Library	TikTok	Keeps stopped ads, has an export	UK, EEA and Switzerland only — no US inventory
LinkedIn Ad Library	LinkedIn	Roughly a year of memory, all regions show creatives	Impressions and targeting for EU-targeted ads only

We have deep guides on each of these, written from testing rather than press releases: the Google Ads Transparency Center guide, the Meta Ad Library, the TikTok Commercial Content Library and the LinkedIn Ad Library.

The honest test for whether free is enough

Free libraries are sufficient if all of these are true:

• You care about who you already know — a named list of rivals, not discovery of unknown advertisers.
• You care about now, not history. You're not trying to reconstruct last year's Black Friday.
• Your rivals are normal-sized advertisers, not giants running thousands of concurrent creatives.
• You're happy to look manually, and nobody needs a spreadsheet or a weekly diff.

Break any one of those and you've found the actual reason paid tools exist. Not "more data" in the abstract — these four specific failures.

The structural limits nobody in the listicles verifies

This is where our own data earns its place. These aren't opinions about interfaces; they're ceilings we hit repeatedly while building coverage.

1. The Google retrieval cap: about 900 creatives per advertiser

The Transparency Center does not tell you when you've seen everything. Past roughly 900 creatives for a single advertiser, retrieval stops — you get the most recent ads and no signal that more exist. Reviewers never notice because they test the tool on advertisers small enough to fit underneath it.

We notice because we're above it. We hold 3,176 Google creatives for EE alone — more than three times what a single pass of the public tool will surface. 48 of our tracked brands exceed the 900 mark, including B&Q at 3,108, ASOS at 3,073, Cinch at 2,579 and carwow at 2,528. You can see the live teaser for the brand that taught us this on our EE ad tracking page.

The practical consequence: for a big-spending advertiser, "I checked the Transparency Center" and "I know what they're running" are different statements. Our guide to ads missing from the Transparency Center covers every cause and its fix.

2. Meta has no memory, and that's unfixable from the outside

When a commercial ad stops running on Meta, it leaves the public library. Not archived — gone. This means no tool, ours included, can retrospectively show you a campaign nobody was recording at the time. Any vendor implying they can reconstruct arbitrary Meta history is selling you a dataset that was built by someone watching continuously, or selling you fiction. The only defence is continuous capture starting before you need it. See Meta Ad Library alternatives for the full treatment.

3. LinkedIn's good data is EU-only

LinkedIn publishes impressions and targeting breakdowns only for ads targeted to the EU/EEA, under the Digital Services Act. For a UK marketer researching a UK-only campaign, the creative and run dates are there but the numbers are not. This is a legal boundary, so no tool can route around it — if a vendor shows you LinkedIn impressions for a UK-only campaign, ask where the figure came from.

4. There is no public Reddit ad library

Reddit runs a real UK ad business and publishes no searchable creative library. Any "Reddit ad spy" feature is scraped observation, not a library feed, with the coverage gaps that implies. We wrote up what's actually possible in the Reddit ad library guide, and the short answer is "less than you'd like".

The fifteen tools the roundups keep listing

Search "best ad spy tools" and you'll meet roughly the same fifteen names in a slightly different order each time, usually with a "best for" label the author invented. Below is that set with the two facts those labels tend to obscure: which ad platforms each one actually covers, and what it costs to get in. We've also answered the question that decides most purchases — whether you can hand it a list of named rivals and have it watch them, as opposed to searching a big shared pool of ads.

Tool	What it's built for	Ad platforms, per the vendor	Watch a named brand list?	Entry price, 4 Nov 2026
Foreplay	Swipe files and creative briefs for ad teams	Meta, TikTok, LinkedIn	Yes — "Spyder", Meta only	$59/mo, but Spyder starts at $175/mo
BigSpy	Cheap breadth across many channels	11 platforms, gated by tier	Quota is counted in creatives, not brands	Free tier; $69/mo covers Facebook, Instagram, AppLovin only
AdSpy	Deep searchable Facebook/Instagram archive	Facebook, Instagram	No — search at query time	$149/mo flat, which it calls introductory
PowerAdSpy	Ad intelligence with AI-parsed creative fields	11 networks, gated by tier	Yes — entry tier allows one brand	$69/mo, Facebook and Instagram
Minea	Ecommerce product and influencer discovery	Meta; TikTok, Pinterest, Snapchat listed "coming soon"	Brand tracker included, depth not documented	$49/mo
PiPiADS	TikTok and Meta ad and product research	TikTok, Meta family	Yes — max 50 targets, Facebook only, not TikTok	Pricing page didn't render for us; API from $200
Panoramata	The whole funnel — ads, email, landing pages	Meta, TikTok, YouTube, LinkedIn, Pinterest, Google Display	Yes — 20 competitors at entry	€99/mo, capped at 6 months of history
Atria	Generating your next ad from rivals' plus your own data	Meta, TikTok	Yes — follow 50 brands	$129/mo on annual billing ($219 monthly list)
AdSpyder	Multi-platform ad search plus ad generation	9 platforms, including Google Search	Claimed, but not documented	$10/mo, capped at 1,000 ads
Dropispy	Dropshipping product research	Facebook, Instagram	No watchlist found	$29.90/mo — from its homepage; it has no pricing page
Winning Hunter	Winning products and ads for ecommerce	Meta, TikTok, Pinterest, Google	Yes — 30 brands, from the $49 tier up	$30/mo on annual billing
GetHookd	A Meta ad library your AI can query	Facebook, Instagram	Yes — "Brand Spy"	$19/mo annual, $29 monthly
SocialPeta	Mobile app and game creative intelligence	Mobile networks plus major social	Not documented	Quote only — no public price
AdHeart	Meta-first ad search, affiliate lineage	Facebook, Instagram, Threads, TikTok (beta)	Saved-search alerts; no brand watchlist documented	$70/mo
TrendTrack	Shopify store intelligence, ads as one module	Meta, Google Search and Shopping, TikTok	Yes — 2 brands at entry	$49/mo
AdRival (us)	Benchmarked ad data on a curated UK brand set	Google, Meta, TikTok qualified; LinkedIn captured but not qualified	Yes, if the brand is in scope — otherwise we seed it first	£99 one-off report; £99/mo tracker

Four things fall out of that table that the "best for" labels never say out loud.

• This is overwhelmingly a Meta category. Seven of the fifteen cover Facebook and Instagram and essentially nothing else. Google Search ads — still where a lot of UK high-intent budget goes — appear in only a handful. If your rivals' money is in Google, most of this market simply doesn't see it.
• The headline price is rarely the real price. Foreplay's competitor-tracking module starts at nearly three times its entry tier. BigSpy's advertised eleven platforms need its $499 plan; its $69 tier is three. PowerAdSpy's cheapest plan tracks exactly one brand. Read the tier, not the banner.
• Watching named rivals is now table stakes, not a differentiator. We expected this to be the dividing line and it isn't: most of the set does it, some of it well. Anyone selling you "we track your competitors" as the unique bit is describing the category, not their product.
• Not one of them publishes a baseline. Every tool in this table will show you your rival's ads. None of them tells you whether what you're looking at is normal — and that turns out to be the thing that decides whether a finding is worth acting on.

Two deliberate omissions. We've left out per-tool feature scores, because behind a login we can't verify them and several vendors claim AI tagging in marketing copy that their own documentation doesn't describe as a filterable field — ours included, which is why we show counts below instead of adjectives. And we've left out star ratings, because on a page written by a competitor they would be worthless.

The gap: nobody tells you what normal looks like

This is the honest version of "why us", and it is narrower than a vendor page would usually admit.

Suppose you do the work. You pick a tool, point it at four rivals, and find that one of them is running 31% of its creative with a hard deadline. Now what? On its own that number means nothing. It's only actionable once you know that across every brand we track the figure is 13.1% — at which point 31% stops being a data point and becomes a rival leaning on urgency more than twice as hard as the market, which is either an opportunity or a warning depending on your category.

That second number is the one no tool in the table above sells you, because producing it requires something different from a search index: a stable brand panel, the same fields extracted the same way across all of it, and a willingness to publish the aggregate. We publish ours — offer share, hook mix, discount mechanics and urgency, broken down across 22 sectors, in our competitor ad analysis guide, free, with no signup. Offer intensity alone runs from 60% of ads in mattresses to 11% in fintech, which is the difference between a rival discounting hard and a rival behaving exactly like its sector.

So the slot we'd claim in anyone's table isn't "tracks competitors" — fifteen products do that. It's knowing whether what you found is normal, on UK brands, with Google included rather than assumed away.

What the paid categories actually do

Rather than rank products — the thing every other page does, badly, in its own favour — here are the four genuinely different jobs paid tools do. Match the job to your problem and the shortlist writes itself.

Creative discovery and swipe (affiliate/DTC lineage)

Tools in the mould of AdSpy, BigSpy, Minea and PiPiADS exist to find unknown winning creatives — search a keyword or product, surface ads from advertisers you've never heard of, sorted by longevity or engagement proxies. Strong for dropshipping and performance-creative inspiration. Weak for "what are my four named UK rivals doing", which is a different question, and historically weakest on UK-specific coverage because the inventory skews US and APAC. Verify UK depth on your own category before paying.

Spend and share-of-voice estimation

Tools like Adbeat and SimilarWeb model spend and placement share. Useful for board slides and media planning. The caveat that belongs on every one of these numbers: outside the DSA-mandated disclosures, UK ad spend figures are modelled, not reported. We go into why in estimating competitor ad spend — treat modelled spend as a directional ratio, never a fact.

Native and push intelligence

Anstrex and AdPlexity cover Taboola, Outbrain, MGID and push traffic — a genuine blind spot in every official library, since none of these networks publishes one. If that's your channel, this category has no free substitute.

Structured, qualified, longitudinal tracking

This is where we sit, and we'll describe it plainly. The libraries give you images and copy; they don't give you fields you can count. We run every captured creative through an AI qualification pass that extracts offer type, discount percentage, hook, urgency, audience and 14 more fields, so questions like "which rivals discounted above 30% in October" become a query rather than an afternoon of scrolling. Of our 50,982 successfully qualified creatives, 17,044 carry a detectable offer and 6,439 have a discount percentage above zero recorded.

We should be careful about claiming this as ours alone, because several competitors describe something similar. PowerAdSpy's own pages say every ad is parsed into type, intent, hook and offer type; SocialPeta lists hook, selling point and CTA pattern; Winning Hunter tags persona, angle and funnel stage. What we couldn't verify from any vendor's public documentation is whether those tags are exposed as fields you can filter and count on, rather than analysis shown per ad — which is a different product. The fair statement is that we're not the only ones in structured territory, we're the ones publishing the totals, and nobody in the set documents a per-ad urgency score.

Where we're weaker than the alternatives

A comparison that only lists our strengths isn't a comparison. In plain terms, choose something else if:

• You need an API. We don't have one. No programmatic access, no webhooks. Foreplay, Atria and PiPiADS all publish documented APIs with per-tier call allowances, and TrendTrack, Winning Hunter and GetHookd ship MCP connectors so an AI assistant can query them directly. If you want ad data in your own warehouse or your own agent, any of those beats us outright.
• You're price-sensitive and just browsing. Several tools in the table start between $10 and $49 a month, and AdSpyder's entry tier is $10. Our tracker is £99/mo. If the job is casual exploration rather than a decision you're going to spend money on, we are the wrong shape and the free libraries are a better starting point than either.
• You want to sign up and start now. There's no self-serve signup. Enquiries go through a form and a human replies. Several tools in this category will have you logged in within a minute.
• You need real-time. Our refresh is currently run manually rather than continuously. For promo detection on a weekly cadence that's fine; for same-hour alerting it isn't.
• Your target isn't a tracked UK brand. Our coverage list runs to 223 brands, but 164 have ads captured and 158 have ad-level qualification. If your rival isn't in scope, we'd have to seed them — fine for a report, but it's a lead-time conversation, not a search box.
• You only need one look. If you want to check one rival once, the free libraries will do it today for nothing, and we'd rather tell you that than sell you a report you didn't need.

Our LinkedIn position deserves the same candour: we hold 1,712 LinkedIn ads as creatives, but our AI qualification currently runs on Google, Meta and TikTok only. So the "25 dimensions" claim applies to those three, not to LinkedIn.

So does our discount data. Whether an ad carries an offer at all is read straight off the creative and is reliable. The depth of that offer is less so: 84% of the discount percentages we hold appear verbatim in the ad's copy or the on-image text we transcribe, and the other 16% are the model inferring the brand's prevailing offer. That's good enough to rank sectors and spot escalation; it is not good enough to quote a single rival's exact discount without opening the ad. Every structured-data tool in this category faces the same problem — we'd rather state our error rate than imply there isn't one.

How to choose, in the order that actually works

1. Spend an hour in the free libraries first. Check your real rivals. Most people discover their question is already answered, and the ones who don't now know precisely which limit they hit — which makes the next step obvious instead of guesswork.
2. Name the limit you hit. No memory? Need the 900+ tail? Need non-EU LinkedIn numbers? Need native networks? Each points at a different category above, and only one of them at us.
3. Trial against a brand you already know well. The fastest way to judge any tool, ours included, is to point it at an advertiser whose activity you can independently verify and count what it misses. Vendors publish global totals; what matters is depth on your category in the UK.
4. Check the refresh cadence and the memory, not the feature list. For competitor work, "how far back does it go and how often does it update" beats every other specification. A long feature list on a shallow archive is worth less than two fields on a continuous one.

The short version

For most UK marketers researching named competitors, the free official libraries are genuinely sufficient, and the pages telling you otherwise mostly sell the alternative. Paid tools become correct when you hit a specific structural wall: Meta's missing memory, Google's ~900-creative retrieval ceiling, LinkedIn's EU-only disclosure, the absence of any Reddit library, or the need for countable fields rather than screenshots. Work out which wall you're against before you look at pricing — and be sceptical of any vendor, us included, whose comparison table happens to crown itself.

Across the fifteen tools above, the differences that actually matter are narrower than the marketing suggests: most are Meta-only, most now track named brands, the entry tier usually excludes the feature you wanted, and none of them will tell you whether the behaviour you've found is normal for the sector. That last one is the gap we work in, and it's also the reason our sector baselines are published free — if the benchmark is only useful to people who pay for it, it isn't much of a benchmark.

If structured, qualified tracking of UK brands is the wall you've hit, our brand pages such as EE and ASOS show the live teaser of what we hold, and our competitor ad analysis guide covers the method whether or not you ever buy anything. If the specific job is offers rather than creative, how to track competitor promotions sets out the free manual version first.