The short answer
For a team swipe file, Foreplay ($59/month) and Swipekit ($26/month with 5 seats) are the main dedicated tools; Atria adds analytics on your own ads; Ad Radar boards hold only winning Meta ads. The tool matters less than what you tag: on our board, ads with strong proof reach the top tier 16% of the time against about 10% overall, a pattern you only see if your swipe file records proof level.
The tools at a glance
All prices from each vendor's pricing page, checked October 2026.
| Tool | Saves from | Team plan | Entry price | Notes |
|---|---|---|---|---|
| Foreplay | Meta, Instagram organic, TikTok ads and organic, LinkedIn, YouTube Shorts | Workflow: 5 users, $175/month | $59/month, $49 billed annually (1 user) | Chrome extension, mobile app, 200M+ community library, briefs |
| Swipekit | Meta, Instagram organic, TikTok ads and organic, LinkedIn | Basic already has 5 seats | $26/month | API and MCP on every plan, white-label share links, 14-day trial without a card |
| Atria | Atria library, Chrome extension, mobile Instagram | Core has 5 seats | $129/month billed annually | Also analyzes your own ad accounts and generates creative |
| Ad Radar | Winning Meta ads on its board; follow any brand by its Ad Library link | Agency: 5 seats, 25 client workspaces, $199 every 4 weeks | $49 every 4 weeks (launch price, list $99) | Boards with share links and CSV export, ten tags per ad, estimated spend |
| Notion, Airtable, Sheets | Anything you paste in | Depends on the app | See their pricing pages | Full control over fields, no capture or transcripts |
One tool is missing from older lists: MagicBrief. Its site says it shut down on July 31, 2026 and its team joined Canva. If that was your swipe file, see MagicBrief alternatives.
Which one fits
Foreplay if your team saves ads from many platforms every day and wants briefs and storyboards in the same place. It has the most complete capture tools today. We compare it with Ad Radar in Ad Radar vs Foreplay.
Swipekit if you want the same capture job for less and need seats more than features. Five seats on the $26 plan is unusual, and API plus MCP access means Claude or your own scripts can read your saves.
Atria if the swipe file is one part of a bigger workflow that includes reporting on your own ads.
Ad Radar if you don't want to curate at all. Every ad on the board already has $50k or more in estimated spend, so a board is a shortlist from the start. It is Meta only, and the Chrome extension for saving from anywhere is still in progress.
A spreadsheet if you are one person, save ten ads a week and care most about the fields. It is slower to fill and better at forcing you to think.
Where the dedicated swipe tools beat Ad Radar
We'd rather say this plainly than have you find out after a month.
- Capture from anywhere. Foreplay and Swipekit save an ad from the Meta Ad Library, TikTok or LinkedIn in one click with a Chrome extension. Ours is still being built.
- Other platforms. Both save TikTok and LinkedIn ads. Ad Radar covers Facebook and Instagram only.
- Ads that never scaled. Sometimes you want to save a small test because the idea is good. Ad Radar only shows ads past $50k in estimated spend, so a clever test won't be there.
- Storyboards. Foreplay turns saves into briefs with storyboards. Ad Radar turns a winning ad into a written brief, not a storyboard.
The fields that matter, from 2,443 winning ads
Most swipe files fail the same way: hundreds of saved ads, no way to tell which ones were real winners and which just looked good. The fix is tagging. To show which fields are worth the effort, here is how the share of ads reaching our top tier (the top 10% by winner score) changes with the value of each tag. Across the board it averages about 10%.
| Field | Value | Ads | Reach top tier |
|---|---|---|---|
| Proof level | Strong | 104 | 16% |
| Proof level | Medium | 1,475 | 10% |
| Hook pattern | Specific number opener | 411 | 13% |
| Hook pattern | Problem/benefit opener | 714 | 9% |
| Hook pattern | Question opener | 366 | 7% |
| Narrative angle | Cautionary tale | 129 | 14% |
| Narrative angle | Root cause | 840 | 8% |
| Awareness | Solution-aware | 496 | 12% |
| Awareness | Unaware | 825 | 9% |
| Hook mechanism | Contrarian reveal | 32 | 22% |
Two lessons. First, the most common value is rarely the strongest: problem/benefit openers are the biggest group on the board and sit below average. A swipe file without tags fills up with the common pattern because it is the one you see most. Second, proof level and hook pattern move the rate more than anything else in this table, and neither is captured by a screenshot. Contrarian reveals look strongest but rest on 32 ads, so treat that row as a lead to check, not a rule.
For how we define these tags, see hook pattern vs hook mechanism and awareness levels in ads.
Three entries, tagged the way a swipe file should be
This is what not to do if your dog has tear stains.
- Est. spend
- $2.1M
- Days live
- 209
- Format
- Video, 44s
- Variants
- 4
Top tier. Saved as "cautionary tale + solution-aware", it sits next to other warnings in the file instead of disappearing into a pet folder.
Give me one minute and you'll stop using dish detergent pods forever.
- Est. spend
- $303K
- Days live
- 110
- Format
- Video, 176s
A contrarian reveal on a household product, nearly three minutes long. The tags tell you why it is in the file: the mechanism, not the category. Ingredient claims are the advertiser's.
These are Hammy's victory laps, and since he doesn't have a tail, he wags his entire body whenever he's excited.
- Est. spend
- $345K
- Days live
- 124
- Format
- Video, 48s
A customer story about a dog that used to struggle to move. Tagged "strong proof", it is easy to pull up when you brief a testimonial ad.
A swipe file template that works in any tool
Use these fields whether you are in Foreplay, Swipekit, Notion or a spreadsheet. The first five come from the Ad Library; the rest you add.
- Brand and landing domain. The domain, not the page name: many brands run ads from creator pages.
- Start date and days live. Recheck monthly. An ad still running after 90 days is evidence.
- Variants. How many versions share the creative. Our post on ad variants as a scaling signal explains why this is the cheapest strong signal.
- Format and length. Image or video, seconds, production style.
- Hook, verbatim. The first spoken line or the headline. Never paraphrase.
- Hook pattern and mechanism. What the opener looks like and how it works.
- Awareness level. Who the ad is written for.
- Proof level. None, medium or strong, and what kind.
- Landing page type. Product page, listicle, quiz, advertorial, dedicated page.
- Why you saved it. One sentence. If you can't write it, don't save the ad.
Review the file weekly. Our weekly creative review post has an agenda, and the facebook ad swipe file guide covers how to turn saves into briefs.
Figures marked as estimated spend come from Ad Radar's model of engagement on public Meta Ad Library ads. They are estimates, labeled as such, and are best used to rank ads against each other.