← All tools

The Ad Craft Index

Who actually runs good ads?

797+ real Meta ads from 16 major advertisers, scored on visual craft, message fit and headline strength — by the same engine behind our ad feedback tool.

1
8.3

Cloudflare

Developer tools · 42 ads · ~50d live

Visual98
Fit87
Headline71

Cloudflare nails visual execution but writes weak headlines, diluting their strong message-audience alignment.

2
7.9

Atlassian

Developer tools · 110 ads · ~59d live

Visual94
Fit80
Headline52

Atlassian nails visual execution but stumbles on headline clarity—their strong design carries weak messaging angles.

3
7.9

GitKraken

Developer tools · 20 ads · ~84d live

Visual100
Fit88
Headline60

GitKraken nails visual execution and relevance but stumbles with headlines that fail to hook hard enough.

4
7.7

Coralogix

Developer tools · 10 ads · ~17d live

Visual100
Fit100
Headline55

Coralogix nails visuals and relevance but sabotages performance with weak headlines that kill click-through rates.

5
7.6

GitHub

Developer tools · 6 ads · ~68d live

Visual100
Fit100
Headline92

GitHub nails visual and messaging precision, but struggles to refresh creative fast enough to combat ad fatigue.

6
7.4

Kinsta

Developer tools · 8 ads · ~87d live

Visual100
Fit100
Headline38

Kinsta nails visuals and positioning but stumbles on headlines—their ads coast on brand strength rather than magnetic copywriting.

7
7.3

Datadog

Developer tools · 192 ads · ~345d live

Visual82
Fit85
Headline30

Datadog nails visual design and targeting but desperately needs headline writing help—they're burning cash on weak openers.

8
7.1

Amazon Web Services

Developer tools · 36 ads · ~60d live

Visual88
Fit69
Headline40

AWS nails visual storytelling but sabotages conversions with weak headlines that fail to hook developers.

9
7.1

Gymshark

E-commerce · 42 ads · ~35d live

Visual82
Fit79
Headline42

Gymshark nails visual storytelling and audience resonance but sabotages campaigns with weak headlines that fail to stop scrolls.

10
6.9

Microsoft Azure

Developer tools · 31 ads · ~41d live

Visual82
Fit65
Headline50

Azure's visuals dazzle, but weak headlines and messaging gaps undermine conversion potential despite lengthy ad runs.

11
6.7

Google Cloud

Developer tools · 25 ads · ~42d live

Visual78
Fit58
Headline58

Google Cloud nails sleek visuals but struggles pairing them with crisp messaging that lands with developers.

12
6.7

Vuori

E-commerce · 83 ads · ~98d live

Visual75
Fit75
Headline36

Vuori nails lifestyle visuals and targeting but stumbles badly with headlines—weak hooks kill otherwise solid creative.

13
6.6

Caraway

E-commerce · 12 ads · ~25d live

Visual75
Fit67
Headline38

Caraway nails visual storytelling but sabotages conversions with weak headlines that fail to hook their audience.

14
6.5

iHerb

E-commerce · 7 ads · ~148d live

Visual79
Fit64
Headline14

iHerb nails visual polish but tanks headlines; their 148-day cycles suggest they're banking on volume over iteration.

15
5.8

Survivor.io

Developer tools · 25 ads · ~89d live

Visual100
Fit16
Headline6

Survivor.io nails stunning visuals but massacres messaging—beautiful ads nobody understands or wants.

16
2.9

Alibaba.com

Developer tools · 148 ads · ~111d live

Visual49
Fit5
Headline5

Alibaba.com nails visual production but hemorrhages on messaging—their ads look polished while barely speaking to who's actually listening.

Methodology: ads collected from the public Meta Ad Library; each rated independently by our vision model on visual craft, message-audience fit and headline strength; brands need at least 6 rated ads to qualify, and scores are confidence-adjusted for sample size. Craft scores, not performance data. Snapshot: 2026-08-12. Ad creatives are shown for review and criticism; all rights belong to their owners. Brand owner? Request a change or removal.

Where would your ad land?

Upload it and get the same scoring — plus the diagnosis, the fixes, and the improved ad. Takes 20 seconds.

Score my ad →

Common questions

How are the scores calculated?
Every ad is rated on three dimensions — visual craft, message-audience fit, and headline strength. A brand’s score is the weighted average across all its analyzed ads (visuals 45%, fit 35%, headlines 20%), confidence-adjusted for sample size — the same principle behind IMDb’s Top 250 — on a 10-point scale.
Does a high score mean their ads perform well?
It means the craft is strong — clear hierarchy, fitting message, working headline. Performance data is private to each advertiser; craft is what’s visible, and weak craft is the most common reason ads underperform.
Why is almost every brand weak on headlines?
Because headline writing is genuinely the hardest part of ad craft — most brands lean on visuals and brand recognition. It’s also the fastest thing to fix.