Snapchat Ads Library: How to Search Competitors' Snap Ads in a 12-Month Window

The Snapchat ads library exists. Its window is narrow.

A media buyer at a US supplement brand hears that a competitor is all over Snapchat. She opens Snap's ad library, types the brand's name and waits. No ads come back. She decides the rumor was wrong.

It probably wasn't. Snap's library for commercial ads is the Snap Ads Gallery at adsgallery.snap.com, and Snap's own Ads Gallery API documentation describes it as a publicly available repository of ads and commercial content, built to retrieve ads that delivered in the European Union in the last 12 months. Snap's transparency hub widens that by one country and describes the gallery as the place to find European Union and Turkiye ads from the same 12 months. A competitor that only buys US impressions sits outside that description entirely.

So the library works, for a specific slice of the market, searched in a specific way, on the days it answers. This guide covers all three: where Snap keeps its two ad libraries, how to search the gallery and read what comes back, what the surfaces we checked never list, and how to turn a handful of competitor records into a creative test plan you can actually produce.

Where Snap keeps its ad libraries

Snap runs two public ad libraries, and most of the confusion around "the Snapchat ads library" comes from treating them as one product. They answer different questions, cover different ads and are built differently.

The Ads Gallery covers commercial ads

The Snap Ads Gallery is the one a marketer researching competitors wants. Its developer documentation lists three things the service retrieves:

• Ads that delivered in the European Union in the last 12 months.
• Ad details, which Snap names as the creative preview, the advertiser name, and the start and end date.
• Organic commercial content that is currently live on Snap, which the API searches by creator name rather than by brand.

The page states its scope in one line under the title. When we opened it on 17 September 2026, that line described ads delivered in the European Union and other jurisdictions over the last 12 months, plus commercial content that is live right now. Snap's transparency hub names the other jurisdiction: Turkiye. The country filter on the page matched it, listing the 27 EU member states and Turkey.

The Political Ads Library is a separate archive

Political and advocacy ads have their own home at snap.com/political-ads. Snap says the library gives the public details about all political and advocacy advertising running on its platform, which is a wider geography than the gallery's, and the page lists yearly archives from 2018 through 2026.

Each year downloads as a spreadsheet file. The 2026 file we downloaded on 17 September 2026 carries, among other columns:

• Money and reach: spend, with its currency, and impressions.
• Timing: the start and end date of each ad.
• Who paid: the paying advertiser and the organization behind the ad.
• Targeting: fields from gender and age bracket to interests and operating system.

That archive serves a different kind of research. It is where you look for a cause-led or issue campaign, and it is the only one of Snap's two libraries that puts a spend figure next to an ad. A skincare or supplement brand selling product will not normally be in it at all, because its ads are not political or advocacy advertising.

Why the gallery is European in the first place

The gallery is a compliance product, not a marketing tool, and that explains its shape. On August 23, 2023, Snap's newsroom set out the changes it was making to meet the EU's Digital Services Act, and one of them was a searchable library of ads shown in the EU. The same post said each record would show who paid for the ad, a visual of the creative, the campaign length, impressions broken down by EU country, and information about the targeting applied.

So the gallery is not new, and it was never designed as a global swipe file. Snap built it as part of meeting a European law, which is why its map starts and stops where it does. Read every limit in the rest of this guide in that light: the tool answers to transparency rules first, and marketers get to use it second.

What the 12-month European scope means for your research

A competitor sweep on Snap starts with one honest question: is this competitor inside the scope at all? Answer it before you type a single name, or you will read an empty result as a finding.

Who you can and cannot study

• A competitor that sells in Europe. Its paid Snap ads from the last 12 months fall inside the scope Snap describes. This is where the gallery is strongest.
• A US or LATAM brand that also runs campaigns in Spain, France or Germany. You see its European ads only. Treat them as the European cut, which may or may not match what the brand runs at home.
• A brand that only advertises outside the EU and Turkiye. Outside the scope Snap describes. An empty result says nothing about whether it runs Snap ads in its home market.

Twelve months is a rolling window

The window moves every day. A campaign that ended 13 months ago is outside the period Snap describes, so last year's holiday push will not be waiting for you once enough time has passed. If a seasonal campaign matters to your plan, look for it within the year and save what you find the same day.

This is the single habit that separates useful Snap research from wishful research. The gallery is a moving window, not an archive you can return to.

Dates, status and impressions are clues, not results

Every record carries a start date, the search accepts a status of active or paused, and Snap's 2023 announcement promised impressions by EU country. Together they are the closest thing to a performance signal the gallery offers:

• An old start date on an active ad means somebody keeps paying to run it.
• A large impressions total means the advertiser bought reach behind it, which is a budget decision, not proof it converted.
• A run that ended within days can be a test that lost or a planned short flight, and the record cannot tell you which.

Every one of those is an inference, and it deserves the same discipline creative analytics applies to your own account. It is also the same kind of inference our guide to cloning a competitor's ad without copying it builds on for Meta's Ad Library, where persistence is the main public clue.

How to search the Snap Ads Gallery, step by step

The interface is short, which is good news: there are only a handful of controls to learn. Here is what the page offered when we used it on 17 September 2026.

Start with the exact advertiser name

The Ads tab has a single search box, and its placeholder asks for the ad publisher's name, noting it "must be exact spelling." That instruction matters more than anything else on the page. Researchers at AI Forensics, who examined Snap's ad transparency tools in April 2026, reported that the gallery could be searched only by advertiser name. There is no search by keyword, product or ad text.

In practice, work through the name in this order:

1. Copy the brand's display name exactly as it appears on its public Snapchat profile. Our one successful search used a retailer's profile name and returned ads paid for by two differently named legal entities of the same group.
2. Try the legal entity name if that fails. The API field is the name of the advertiser paying for the ad, which can differ from the consumer brand. We could not confirm that a legal-name search works before publishing, because every retry was throttled.
3. Try the obvious variants: with and without a legal suffix such as Inc or GmbH, and with and without a country tag.

Narrow by country, status and date

Three filters sit above the search box:

Filter	What it offers	Use it to
Shown in	All, 27 EU states, Turkey	Isolate one market
Ad Status	All, Active, Paused	Split live from stopped
Anytime	A calendar date range	Bracket a launch or season

The API mirrors those filters. Snap's documentation says its brand search takes up to five parameters: country codes, a start date, an end date, a status, and the paying advertiser's name.

Start broad, with every country and every status, then narrow. A narrow first search that returns nothing teaches you nothing, because you cannot tell a wrong filter from a wrong name.

The Commercial Content tab searches creators, not brands

The second tab covers organic commercial content: creator posts that are live on Snap now, as opposed to paid ads. It has one box that asks for a content creator and no filters, and the API behind it takes a single parameter, the creator's name. A brand name is not what it matches on.

The useful move runs the other way. When a competitor's ads or social posts feature a creator, search that creator here to see what else they are publishing as commercial content, and whether the same brand appears among those posts.

What a gallery record shows, and what it does not

The fields Snap names

Two Snap sources describe a record, three years apart:

• The Ads Gallery API documentation, which carries no date and which we read on 17 September 2026: the creative preview, the advertiser name, and the start and end date.
• Snap's August 2023 newsroom post: who paid for the ad, a visual of the creative, the campaign length, impressions broken down by EU country, and information about the targeting applied.

For creative research the preview is the whole prize. It is the actual vertical asset a competitor paid to put in front of Snapchatters, and it is the thing you will study second by second.

What one real search returned

When a search did go through on 17 September 2026, we asked the documented endpoint for a large European fashion retailer's ads in Germany. It returned ten active records, and each one carried more than the documentation lists:

• The creative and its copy: the creative type (collection, web view or composite), the headline and the call to action.
• Reach: a total impressions count and a per-country breakdown, with a slot for Turkey next to the EU countries.
• Targeting: an age range and a gender on every record, plus the device operating system on four of them.
• Identity: the Snapchat profile name, the legal entity paying, and an internal ad name that, on four of the ten, spelled out the market, the operating system and a 6-10 second length band.
• Dates: a start date, with no end date on ads still running. Several started in February 2025 and were still active, so long-running ads keep their original launch date.

Three of the ten also carried a direct link to the ad's media file, which makes saving the creative easier when it is there.

What no surface we checked lists

We read Snap's API documentation, its 2023 announcement of the library, its transparency hub, the labels in the gallery's own interface and the ten records above. None of them names any of these for commercial ads:

• Spend. No budget or amount spent is listed for Ads Gallery records.
• Performance. No click-through rate, view rate, conversions or return on ad spend.
• Ads delivered elsewhere. Anything outside the EU and Turkiye is outside the scope Snap describes.

That is a statement about what those surfaces say, not a claim about what Snap stores. It is consistent with the political library being the place spend appears: Snap's political ad policy says it may publicly disclose ad content, targeting details, delivery, spend and other campaign information for political advertising. If you want to reason about what a competitor might be paying for its placements, our breakdown of what Snapchat ads cost shows why every outside figure is an estimate.

The two libraries side by side

	Ads Gallery	Political Ads Library
Covers	Commercial ads, EU and Turkiye	Political and advocacy ads
Window	Last 12 months	Yearly files since 2018
How you search	Exact advertiser name	Download by year
Spend	Not on surfaces checked	Column in archive file
Best for	Competitor creative	Issue and cause campaigns

When the gallery does not answer

The biggest practical limit on Snap's library is not a missing field. It is whether the service responds at all.

April 2026: unavailable 99.7% of the time

AI Forensics investigated Snap's ad transparency tools in April 2026 and published its findings on 13 May 2026. Three points matter to anyone planning research on the gallery:

• Availability. The repository was unavailable 99.7% of the time during its observation period, a problem it traced to rate limiting.
• The fix. After it shared the findings with Snap, a fix deployed on April 29 lifted availability from almost 0% to nearly 100%, by AI Forensics' account.
• What stayed open. Its two other findings, the name-only search and the absence of commercial-content labels on the web, had not been addressed when it published.

What we saw on 17 September 2026

We ran our own check before publishing this guide. Between 16:32 and 17:07 UTC on 17 September 2026 we sent 15 searches from a single connection: 13 straight to the documented search endpoint, spaced from 20 seconds to more than 15 minutes apart, and two typed into the gallery in Chrome.

• 14 came back HTTP 429, Too Many Requests.
• One went through, at 16:49 UTC, and it returned the ten records described above.
• The gallery never said so. On both throttled Chrome searches, its results area simply kept inviting us to search.

One connection on one afternoon proves nothing about the platform as a whole, and the limit may have applied to us alone. It does show that the failure mode is still possible after the April fix: a throttled search can look exactly like an empty one.

The API is open, and it is the same pipe

Snap's developer documentation says the Ads library endpoints are open for use by third parties and need no OAuth token. That makes a scripted weekly sweep tempting. Before you build one, know that on 17 September 2026 the gallery page sent its searches to that same documented endpoint, so a script does not get a separate lane.

• Space requests out. A burst of searches is the fastest way to a 429.
• Log throttles separately. A 429 is not an empty result and must never be recorded as one.
• Never write "no ads" on a throttled day. Retry later, then decide.

Turning a Snap ad record into a test plan

Finding a competitor's Snap ads is the easy half. The records are only worth the time if they end up as briefs you produce and test.

Capture the record the day you see it

Every record ages out of a 12-month window, and the service has had long stretches of unavailability. So anything worth studying is worth saving immediately. For each ad, write down:

• the advertiser name exactly as it matched,
• the country filter you used,
• the start date, end date, status and total impressions,
• a screen recording of the creative preview.

Read each ad against Snap's own brief

Snap publishes the standard its ads are judged by, and our guide to running Snapchat ads covers it in full. Snap says its ads generally use a vertical 9x16 format, recommends keeping Sponsored Snaps under 10 seconds, and on its Sponsored Snaps page tells advertisers to try UGC-style video and grab attention early.

That gives you three quick questions for every competitor ad:

1. How fast does it earn attention? Note what happens before the viewer could swipe, the window our guide to ad hook variations tests.
2. Does it look like a Snap or like a commercial? UGC-style ads that feel native are what Snap asks for.
3. How long does it run? A competitor ad far over 10 seconds that stays active is worth a closer look, because it runs well past the length Snap recommends for Sponsored Snaps and survives anyway.

A worked example: 14 records, 6 briefs

Here is the arithmetic on a hypothetical sweep of one competitor selling in Germany and France.

The search returns 14 ads from the last 12 months. Four are still active. Of the ten paused ads, six ran under two weeks and four ran more than two months. Set the six short runs aside and you have 8 ads that survived long enough to matter. Sort those by total impressions so the concepts with the most reach behind them get read first.

Group those eight by concept and they collapse to 3: a problem-first selfie, an unboxing, and a side-by-side comparison. Cross each concept with two angles you already believe in, say price and speed, and you have 6 briefs, the kind of matrix our ad creative strategy guide builds.

Six briefs is a real testing round. It is also the point where most teams stall, because six briefs means six scripts, six shoots or creator deals, and weeks of back-and-forth before anything is live. Our guide to how many ad creatives you need explains why your conversion volume, not your list of ideas, decides how many of the six you can test at once, and ad creative testing covers how to read the results once they are in.

When the competitor never appears

If a brand only advertises outside Europe, the gallery cannot help, and the right move is to research the same brand elsewhere. Meta widened its Ad Library in March 2019 to bring in all active ads any Page is running, and TikTok's Top Ads gallery is covered in our TikTok Creative Center guide. Vertical creative a brand runs on Snap may also run on those platforms, so check them before you conclude a concept is invisible.

How Novoads solves the brief-to-video step

Novoads turns a written brief into a finished UGC-style video ad: you write a script or upload a product image, pick an AI actor, and download a vertical 9:16 video ready for Snap. That is exactly the step the gallery leaves you at, holding six briefs and no footage.

Each brief from the worked example becomes a script, and each script becomes a render you can put into the next test round. If a brief calls for someone holding your product, create a custom actor from a photo of a person holding it, or start from one of the Discover product templates; a stock actor from the library has empty hands.

Novoads starts at $49/month (Starter, 50 credits per month). All plans are published on /pricing, and you can start producing your Snap briefs here. Cancel anytime.

A library shows what ran. Your tests decide what runs next.

Snap's ads library is narrow by design. It sees Europe and Turkiye, it forgets after 12 months, it asks for exact names, and on some days it barely answers at all. None of that makes it useless. It makes it a tool you have to use deliberately: confirm the competitor is in scope, save every record the day you find it, and treat dates as clues rather than results.

The advantage never goes to the team that finds the most competitor ads. It goes to the team that turns six records into six live tests before the window closes.