Research

The same shopper behaves differently in different contexts.

Millions of sessions across brands on FERMÀT. Here is what the behavior keeps telling us.

On-site search · Re-entryAugust 2026

One in three buyers go back to Google to navigate a brand's own catalog.

They already decided to buy, then left to re-search their way back in.

Read the analysis on LinkedIn
Traffic source · Dwell timeAugust 2026

Where a shopper comes from predicts how long they stay.

Shoppers from Google stayed nearly twice as long as those from Meta, across 10 brands.

Read the analysis on LinkedIn
Channel · Catalog sizeAugust 2026

Engagement shifts with the channel and the size of the catalog.

Scroll depth across 10 brands, too varied for one site-wide layout to serve.

Read the analysis on LinkedIn
Paid media · RetentionJuly 2026

Brands pay acquisition dollars to reacquire customers they already won.

Across 769K return visits, the median brand paid to bring back 1 in 8 existing customers.

Read the analysis on LinkedIn
Landing context · ConversionMay 2026

Homepage entrants converted 55% better than product-page landers.

One of three shopper patterns pulled from real pixel data across brands.

Read the analysis on LinkedIn
Traffic source · Browse depthApril 2026

The Google vs Meta conversion gap collapses the moment shoppers browse.

Meta shoppers convert nearly as well as Google's once they go deep. Only 9% ever reach 7+ pages, against 23% on Google.

Read the analysis on LinkedIn
AI answers · CitationsApril 2026

Reddit threads get cited by AI models more often than editorial content.

A 3-month analysis put Reddit at 20.6% against 15.9%, with brand-mention lift close behind.

Read the analysis on LinkedIn
More to come

New findings land here as we publish them.

See what your shoppers’ behavior is telling you.

Our agents read the same signals on your own data.

Book a demo