The Product Title Study: What 95,000 Titles Reveal About What Sells
Data updated July 28, 2026
Written by Christopher Krassnig, Founder & CEO of ZenoX Media. Every number on this page comes from the accounts ZenoX runs, measured, not surveyed.
The headline finding
Top-selling products carry shorter, cleaner titles than products that never sell: 50.6 characters on average versus 56.3. We analyzed 95,149 fashion product titles to see which words earn their place.
95,149
Product titles analyzed
115 fashion stores, June 2026
50.6 vs 56.3
Average title length in characters
top sellers vs zero sellers
1.11
Revenue index for 'floral' in the title
11% above its fair share of revenue
0.88
Revenue index for 'waterproof'
spec words underdeliver in fashion
Winners Have Shorter Titles. Losers Have Clutter.
Line up the top 10% of products by revenue against the products that sold nothing, and the title difference is consistent: winners average 8.4 words and 50.6 characters. Zero sellers average 9.5 words and 56.3 characters.
The clutter is specific. Zero sellers stuff more variant data into the title: 18.3% carry a color word versus 12.2% of winners, and 45.1% carry size tokens versus 34.4%. Even gender labels follow the pattern, 56.2% versus 47.6%.
One honest caveat: this is correlation. Head products get hand-tuned titles; forgotten variants inherit auto-generated ones. But the direction of the lesson does not change. The titles that sell look clean, lead with the product, and push variant details into the fields Google built for them.
TITLE CLUTTER: TOP SELLERS VS ZERO SELLERS
- Top 10% by revenue
- Zero sellers
The Words That Earn Their Place
We indexed every common title word against its fair share: an index of 1.00 means products carrying that word earn exactly the revenue share their ad spend predicts. Above 1.00, the word over-delivers.
The over-deliverers are style and fabric words: 'vintage' and retro-era markers at 1.12, 'floral' at 1.11, 'cropped' at 1.10, 'linen' at 1.06. Words that sell a look, not a spec sheet.
The under-deliverers are technical spec words: 'UV400' at 0.89, 'waterproof' at 0.88. In fashion, specs do not sell; style does. Even 'set' and 'two-piece' products, a proven social commerce format, land at 0.94 on Google, where searchers hunt for a specific item more than an outfit.
TITLE WORD REVENUE INDEX (1.00 = FAIR SHARE)
| Title trait | Products | Revenue index |
|---|---|---|
| Vintage / retro era | 1,790 | 1.12 |
| Floral | 4,032 | 1.11 |
| Cropped | 874 | 1.10 |
| Linen | 1,596 | 1.06 |
| Cotton | 579 | 1.03 |
| Wide-leg | 2,146 | 1.01 |
| Oversized | 1,791 | 1.01 |
| Satin / silk | 988 | 0.99 |
| Knit | 2,362 | 0.98 |
| Wedding / party | 3,252 | 0.98 |
| Set / two-piece | 5,931 | 0.94 |
| Leather | 2,597 | 0.93 |
| UV400 | 334 | 0.89 |
| Waterproof | 268 | 0.88 |
How to Write Titles That Match the Data
Lead with what the product is, in the words a buyer would search. The pattern that wins: product first, style second, brand last, and nothing else fighting for space.
Move color and size out of the title and into their own feed attributes. Google reads those fields directly; repeating them in the title costs clarity and, per this data, correlates with the losing pile.
Spend your remaining characters on style and fabric words with proven pull: era markers, prints, materials people search by feel. Skip spec jargon unless you sell in a category where specs are the purchase driver. Fashion is not that category.
Methodology
95,149 fashion products with real Google Ads spend across 115 accounts, titles in five languages classified with translated patterns.
Rolling 30-day windows pulled June 1 to July 1, 2026
- For each title trait (a word or pattern like 'linen', 'oversized', or a color name) we compared the trait's share of niche revenue to its share of niche ad spend. The result is an index where 1.00 means the trait earns exactly its fair share. We only report traits carried by at least 200 products.
- The index compares shares, not returns. It never exposes any store's actual performance.
- Title structure stats compare the top 10% of products by revenue (9,514 products) against products with zero attributed revenue (68,206 products).
- Correlation, not causation: winners have clean titles partly because head products get hand-polished, while long-tail variants accumulate clutter. The patterns still tell you what winning titles look like in practice.
Questions People Ask About This Study
Does a 1.11 Index Mean Adding 'Floral' Boosts Revenue 11%?
No. It means products carrying that word collectively earned 11% more revenue share than their spend share predicts. The word marks products that win; pasting it onto a product that is not floral does nothing except mislead shoppers.
Why Only Fashion?
It is our largest single-niche dataset, big enough to clear the 200-products-per-trait floor with room to spare. Other niches follow as their samples reach publishable size.
Should Titles Always Be Under 51 Characters?
No. The finding is about clutter, not a magic length. Google truncates around 70 characters in most placements, and winners simply spend fewer of theirs on variant data a feed field should carry.
Keep reading
More from the ZenoX team.
On the call, we look at your account live. If we can grow it, we will show you how. If we cannot, we will tell you that too, and point you at someone who can.
We work with brands serious about scale. If that is you, let's talk.
