{"id":41669,"date":"2026-05-21T10:55:50","date_gmt":"2026-05-21T10:55:50","guid":{"rendered":"https:\/\/adspyder.io\/blog\/?p=41669"},"modified":"2026-05-21T11:24:22","modified_gmt":"2026-05-21T11:24:22","slug":"build-ad-swipe-file","status":"publish","type":"post","link":"https:\/\/adspyder.io\/blog\/build-ad-swipe-file\/","title":{"rendered":"How to Build a Swipe File Using Competitor Ad Intelligence \u2014 May 2026"},"content":{"rendered":"
\n
How-To Guide<\/span><\/div>\n

<\/p>\n

<\/p>\n

\n

Quick Answer<\/p>\n

To build an ad swipe file that improves campaigns: search competitor ads by domain on your target platform, filter for ads live 30+ days, save into named project folders, tag each one by hook type and format, and refresh before every brief. AdSpyder’s Ad Library<\/a> \u2014 360 million+ ads across 10 platforms \u2014 gives you the live competitor intelligence a static swipe file never can.<\/p>\n<\/div>\n

<\/p>\n

Most creatives and media buyers keep their “swipe file” in a Slack thread, a camera roll, or a folder of screenshots they haven’t opened since last quarter. That’s not a swipe file \u2014 it’s an archive of good intentions.<\/p>\n

A real swipe file is searchable, curated by platform, and updated before every brief. It’s the difference between staring at a blank doc and walking into a briefing with 20 examples of what your competitors are actually spending budget on right now.<\/p>\n

This guide gives you the exact workflow: where to source competitor ads at scale, what to filter for, how to organise what you save, and how to turn that library into campaign briefs that reflect real market intelligence \u2014 not gut feel.<\/p>\n

<\/p>\n

\n
\n

360M+<\/p>\n

ads searchable<\/p>\n

in AdSpyder’s Ad Library across 10 platforms<\/p>\n<\/div>\n

\n

87%<\/p>\n

of research on 3 platforms<\/p>\n

Google (50%), Facebook (22%), YouTube (15%) \u2014 where to build your swipe file first<\/p>\n<\/div>\n

\n

~54%<\/p>\n

of ads run under 7 days<\/p>\n

tested and killed \u2014 the noise a date filter removes from your swipe file<\/p>\n<\/div>\n

\n

43 days<\/p>\n

median LinkedIn ad life<\/p>\n

the longest of any platform \u2014 B2B swipe files need a dedicated section<\/p>\n<\/div>\n<\/div>\n

AdSpyder platform data, May 2026. Ad archive analysis across 24,650 sampled ads (Google Search, Bing, Display, YouTube, Meta, LinkedIn).<\/p>\n

<\/p>\n

\n

Search 360M+ competitor ads. Save to named projects. Brief smarter.<\/p>\n

Filter by advertiser domain, date range, and format across 10 platforms.<\/p>\n

Explore the Ad Library \u2192<\/a><\/p>\n<\/div>\n

<\/p>\n

\n

In This Guide<\/p>\n

What is an ad swipe file?<\/a>
\nWhy static swipe files fail<\/a>
\nWhere to search first<\/a>
\nWhat to save (and skip)<\/a>
\n6-step workflow<\/a>
\nHow to organise it<\/a>
\nActivating in briefs<\/a>
\nManual vs intelligence-led<\/a>
\nPre-brief checklist<\/a>
\nFAQ<\/a><\/div>\n<\/div>\n
\n

<\/p>\n

What Is an Ad Swipe File \u2014 and What It Is Not<\/h2>\n

An ad swipe file is a curated reference library of competitor ads you’d actually use as inspiration for a live brief. The term “swipe” comes from direct-response copywriting \u2014 Gary Halbert kept physical folders of proven sales letters so he could study what worked before writing new copy. The same logic applies to paid media: before you brief creative, you look at what’s running and converting in your market.<\/p>\n

What it’s not: a collection of ads you saved because they looked good. A swipe file without context \u2014 the platform, the advertiser, the reason you saved it \u2014 decays fast. Six weeks later you have no idea why that screenshot is in your downloads folder, and you’re briefing from instinct again.<\/p>\n

Three things separate a useful swipe file from a digital junk drawer: source quality<\/strong> (are these from competitors actually spending budget?), metadata<\/strong> (do you know the platform, format, hook type?), and freshness<\/strong> (is this from last quarter or last week?).<\/p>\n

\n

On curated ad libraries<\/p>\n

Sites like adswipefile.com or swiped.co give you a hand-picked set of ads \u2014 useful for general inspiration, useless for competitive intelligence. They can’t tell you what your specific competitors are running right now, what’s been live for 30+ days (a signal of profitability), or what ad formats dominate your exact vertical. That’s what a competitor-sourced swipe file does differently.<\/p>\n<\/div>\n

\n

<\/p>\n

Why Static Swipe Files Fail Creatives and Media Buyers<\/h2>\n

The core problem with a folder of screenshots: it has no memory. You don’t know when an ad ran, how long it stayed live, or whether the advertiser kept spending behind it. You’re looking at a frozen moment, not a signal.<\/p>\n

There’s a second problem: format bias. Most manual swipe files over-index on whatever the person building them personally finds interesting \u2014 usually striking visual creative or clever copy. “Looks good in a folder” and “performed well in market” are completely different filters.<\/p>\n

The useful filter is longevity. An ad a competitor ran for 30+ days, or kept cycling for 3+ months, almost certainly has budget behind it because it’s working. AdSpyder’s archive data makes that filter available at scale \u2014 and the numbers make the case for using it.<\/p>\n

<\/p>\n

\n
\n

~54%<\/p>\n

test-and-kill cohort<\/p>\n

of all ads run less than 7 days across platforms we sampled<\/p>\n<\/div>\n

\n

22%<\/p>\n

evergreen cohort<\/p>\n

survive 30+ days \u2014 the creative that earned its place<\/p>\n<\/div>\n

\n

43 days<\/p>\n

median LinkedIn lifespan<\/p>\n

longest of any platform \u2014 B2B ads run an order of magnitude longer than B2C<\/p>\n<\/div>\n<\/div>\n

Source: AdSpyder ad archive analysis, 24,650 sampled ads across Google Search, Bing, Display, YouTube, Meta, LinkedIn \u2014 AdSpyder platform data, May 2026.<\/p>\n

The implication is direct: bias your swipe file toward ads that have been running for 30 days or longer. On most platforms, fewer than 1 in 4 ads survives that long. The ones that do are either proven direct-response performers or core brand-awareness plays \u2014 both worth studying before your next brief.<\/p>\n

\n

One caveat on longevity as a signal<\/p>\n

AdSpyder measures “last seen” dates \u2014 an ad’s observed duration is a floor, not the exact run length. An ad still live today shows lastSeen = today. The direction of the signal is reliable; treat specific day counts as a guide, not a guarantee.<\/p>\n<\/div>\n

\"Why<\/p>\n

\n

<\/p>\n

Where to Search First: Platform Priority for Your Swipe File<\/h2>\n

A common mistake: building a swipe file that doesn’t mirror where you actually buy. If 80% of your spend is on Google Search and Meta, a TikTok-heavy swipe file gives you creative inspiration you can’t use \u2014 and a false read on your competitive landscape.<\/p>\n

AdSpyder’s search logs across 88,000+ competitor ad searches show exactly where practitioners do competitive research. Build your swipe file to match:<\/p>\n

\n\n\n\n\n\n\n\n\n\n\n
Platform<\/th>\nShare of research<\/th>\nAds indexed<\/th>\nBest signal for swipe file<\/th>\n<\/tr>\n<\/thead>\n
Google Search<\/td>\n50%<\/td>\n165M+<\/td>\nCopy hooks, headline patterns, CTA verb<\/td>\n<\/tr>\n
Facebook \/ Instagram<\/td>\n22%<\/td>\n55M+<\/td>\nVisual formats, offer framing, emotional trigger<\/td>\n<\/tr>\n
YouTube<\/td>\n15%<\/td>\n2.5M+<\/td>\nVideo scripts, pre-roll hooks, brand narrative<\/td>\n<\/tr>\n
LinkedIn<\/td>\n2.7%<\/td>\n860K+<\/td>\nB2B messaging, evergreen ad patterns<\/td>\n<\/tr>\n
Display \/ Bing<\/td>\n~5%<\/td>\n23M+<\/td>\nBanner formats, retargeting creative<\/td>\n<\/tr>\n
Amazon \/ TikTok \/ Twitter<\/td>\n~1%<\/td>\n27M+<\/td>\nAdd only if these are active spend channels<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

Source: AdSpyder Ad Library search distribution, 88,035 searches \u2014 AdSpyder platform data, May 2026.<\/p>\n

Google, Facebook, and YouTube account for 87% of all competitive ad research. Build depth on those three first. For B2B, LinkedIn earns its own dedicated section \u2014 its 43-day median ad lifespan means the creative that’s still running at 90 days is almost certainly a proven performer.<\/p>\n

One more thing the data settles: the format question for Meta. Across AdSpyder’s Meta historical archive of 44 million ads, 88% are static images and 12% are video<\/strong>. A Meta swipe file that’s video-heavy is misrepresenting what actually dominates in that market. Build yours to reflect reality \u2014 mostly image, with a dedicated video subsection for Meta, YouTube, and TikTok.<\/p>\n

\n

Don’t build swipe file coverage for platforms you don’t buy on<\/p>\n

A TikTok swipe file is wasted effort if your budget is on Google and Meta. Add a platform only if it’s in your active media plan \u2014 otherwise you’re creating research overhead with zero briefing payoff.<\/p>\n<\/div>\n

\n

<\/p>\n

What to Save \u2014 and What to Skip<\/h2>\n

Saving everything is the same as saving nothing. The filter you apply when adding to your swipe file matters more than the sourcing itself.<\/p>\n

Save this:<\/p>\n

\n