We analyzed 15,412 AI-generated Pinterest pins. The most popular style isn't the one that performs best.

marstay — Wed, 23 Sep 2026 10:52:03 +0000

I run URL2Pin, a platform that turns a product URL (Amazon, Etsy, a blog post) into ready to publish Pinterest pins: AI image, copy, hashtags, scheduling. It's a solo bootstrapped SaaS, nothing huge, but it's been running long enough to generate a dataset I haven't seen published anywhere else: real usage data from real creators actually running Pinterest accounts, not a survey or a vendor's cherry picked case study.

So I pulled it apart. Here's what's actually true, with the numbers.

The dataset

15,412 pins, 439 creators, March to September 2026. I'm reporting user weighted percentages (average across creators) rather than raw pin counts, because usage is concentrated: the top creator alone accounts for about 21% of all pins, and the top 3 for about 46%. Pin weighted numbers would just be describing a few power users. Median usage is 10 pins per creator.

Finding 1: Amazon affiliate content dominates, by a wide margin

52.3% of creators have generated at least one pin for an Amazon product. Etsy, by comparison, sits at 2.9% (only 4.9% of creators have ever touched it). That's not a close race. For a tool that supports both marketplaces equally, the actual usage is almost entirely Amazon affiliate shaped. If you're building anything in this space, Etsy content is a rounding error next to Amazon. (More on what that actually looks like in practice: Pinterest for Amazon Affiliates.)

Finding 2: everyone uses the same style, and it's not the best one

83% of creators use the same visual pin layout (a clean, minimal style we call minimal_elegant). That's an enormous default bias signal: it's very likely just the first option in the list.

But when we matched pins to their actual Pinterest performance (impressions, clicks, saves), a different style, a clean, question headline format, outperformed it on every metric that matters:

Style	Save rate	Clicks/pin
`question_style`	0.41%	N/A
`clean_appetizing`	0.23%	0.373 (best)
`grid_3_images`	0.16%	N/A
`minimal_elegant` (83% of usage)	0.10%	0.313
`minimal_typography`	0.04%	N/A

The style 83% of people default to has roughly a quarter of the save rate of the best performing one. Defaults are sticky, and most people never go looking for a reason to change them.

Finding 3: more pins per link doesn't split your reach, it adds to it

I expected posting multiple pins for the same URL to cannibalize reach (same audience, same link, diminishing returns). It doesn't work that way. The first pin for a link gets a median of 18 to 21 impressions; later pins for the same link still pull 12 to 14. A solo pin, the only one ever posted for that link, gets just 10.

In other words: posting three pins for one link outperforms posting one pin, per pin. Pinterest's algorithm seems to reward a link having multiple entry points more than it penalizes repetition. (We wrote this up in more detail in the Pinterest Affiliate Pin Generator guide.)

Finding 4: there's a real ceiling on posting volume, and it's lower than you'd think

Reach per pin peaks when creators post 11 to 20 pins per day (median 23 impressions) and then falls off a cliff at 21+ per day (median drops to 13 impressions, and the save rate, already thin at 0.15% platform wide, bottoms out at 0.09%, the worst of any posting volume bucket).

So "more content" only works up to a point, and that point is a lot lower than the quota based pricing most tools (including, honestly, my own) are built around encourages people to hit.

Why this matters if you're building anything Pinterest adjacent

Pinterest tooling content online is mostly opinion: "best practices" nobody's actually tested against real accounts. This is 15,412 real pins from 439 real creators, and three of the four findings above contradict what I would have guessed going in. If you're building for this space, or just curious, happy to share more of the methodology or raw breakdowns in the comments.