Our partners, platforms and the research we build on.
We build our own tools at Freak. They're only as good as the data behind them, so we source the best we can find. These are the platforms we use, the directories that list us and the published research our image analysis is measured against.
Data partners
Every tool in Freak Lab starts with data. We license it from the platforms below because they collect it at a scale no agency or SaaS company can match. Our approach is flexibility, finding the richest and most powerful data sets before putting them into our algorithmic frameworks.
- SEARCH AND AI VISIBILITY
Semrush
Keyword, competitor and backlink data across Google and the AI assistants. It feeds the keyword research in Freak Copy, our position tracking and the search plans we build for clients.
- AMAZON MARKET DATA
Jungle Scout
Amazon sales estimates, search volumes and category data. We use it to size a category, find the search terms that sell and track competitors before we write a listing or plan a launch.
- AI ASSISTANT VISIBILITY
Profound
Shows how brands appear in answers from assistants like ChatGPT, Perplexity and Google AI Overviews. We use it to measure AI visibility for product content and to see which sources the assistants trust.
- CREATOR DATA
Influencers.Club
Creator profiles, audience data and verified contacts across Instagram, TikTok, YouTube and more. It feeds creator discovery and vetting in Freak Creators and our seeding programmes.
Where you can find us listed
Independent directories where brands compare agencies before they get in touch.
- AGENCY MARKETPLACE
DesignRush
A B2B marketplace that matches brands with agencies. Freak is a DesignRush Verified Agency, listed among the digital marketing agencies in London.
- TRADE PRESS
LBB
Little Black Book covers the creative and production industry. Our company profile sits alongside the interviews our team has given.
Neuromarketing research we build on
Our Image Analyser predicts where a shopper looks on a product image, what engages them, excites them and how it makes them feel. Its scores rest on more than a century of eye tracking and consumer behaviour research, the core of neuromarketing.
| Year | Who | What it showed |
|---|---|---|
| 1879 | Javal | Named the jumps the eye makes when reading saccades |
| 1898 | Huey and Delabarre | Early eye trackers that recorded the eye's movements while reading |
| 1935 | Buswell | Photographed 200 people looking at pictures and found their instructions changed where they looked |
| 1967 | Yarbus | Eye Movements and Vision showed the viewer's task steers the eye |
| 1980 | Treisman and Gelade | Simple features such as colour register early and all at once, before attention |
| 1998 | Itti, Koch and Niebur | The first working computer model of visual saliency |
| 2004 | Pieters and Wedel | Eye tracked more than 3.6k consumers looking at 1,363 magazine ads |
| 2006 | Nielsen Norman Group | The F shaped pattern people use to scan web pages |
| 2009 | Judd and colleagues, MIT | 1,003 images eye tracked by 15 viewers, the start of the MIT saliency benchmark |
| 2012 | Pieters and Wedel | People get the gist of an ad in a single glance of a tenth of a second |
| 2022 | Jiang and colleagues | The first eye tracking dataset of ecommerce images |
| 2025 | Hosseini and colleagues | SUM, a saliency model with separate modes for photos, ecommerce images and interfaces |
Two lessons run through all of it. Each step made attention data bigger and cheaper, which is what made prediction possible. The viewer's task also keeps deciding where the eye goes, which is why we weight every score by where the image will be seen.
We've worked with eye tracking since 2019, when we built VR eye tracking for J&J. The five measures we score and how we weight them are in neuromarketing and AI image analysis.
Why we build our own tools on top
The Freak approach is hyper customisation. We build you the exact tools you need on top of the best data sets, processed through our Freak data layers. If you want to see how that works on your own range, get in touch.
- Semrush, Jungle Scout, Profound and Influencers.Club supply the data behind Freak Lab
- Freak is a verified agency on DesignRush with a company profile on LBB
- Our image analysis rests on more than a century of neuromarketing research
- We use the data inside our own tools, not the platforms' dashboards
Sources
- A feature integration theory of attention, Treisman and Gelade, Cognitive Psychology, 1980
- A model of saliency based visual attention for rapid scene analysis, Itti, Koch and Niebur, IEEE, 1998
- Gaze bias both reflects and influences preference, Shimojo and others, Nature Neuroscience, 2003
- Processing fluency and aesthetic pleasure, Reber, Schwarz and Winkielman, Personality and Social Psychology Review, 2004
- Graph based visual saliency, Harel, Koch and Perona, NIPS, 2006
- F shaped pattern for reading web content, Nielsen, Nielsen Norman Group, 2006
- The central fixation bias in scene viewing, Tatler, Journal of Vision, 2007
- Eye tracking for visual marketing, Wedel and Pieters, Foundations and Trends in Marketing, 2008
- Biasing simple choices by manipulating relative visual attention, Armel, Beaumel and Rangel, Judgment and Decision Making, 2008
- Competitive brand salience, van der Lans, Pieters and Wedel, Marketing Science, 2008
- Does in store marketing work?, Chandon and others, INSEAD working paper, 2008
- Did Javal measure eye movements during reading?, Wade and Tatler, Journal of Eye Movement Research, 2009
- Learning to predict where humans look, Judd, Ehinger, Durand and Torralba, ICCV, 2009
- Faces and text attract gaze independent of the task, Cerf, Frady and Koch, Journal of Vision, 2009
- The stopping power of advertising, measures and effects of visual complexity, Pieters, Wedel and Batra, Journal of Marketing, 2010
- Visual fixations and the computation and comparison of value in simple choice, Krajbich, Armel and Rangel, Nature Neuroscience, 2010
- A study on the impact of product images on user clicks for online shopping, Goswami, Chittar and Sung, WWW, 2011
- Relative visual saliency differences induce sizable bias in consumer choice, Milosavljevic, Navalpakkam, Koch and Rangel, Journal of Consumer Psychology, 2012
- Ad gist, ad communication in a single eye fixation, Pieters and Wedel, Marketing Science, 2012
- The role of visual complexity and prototypicality regarding first impression of websites, Tuch and others, International Journal of Human Computer Studies, 2012
- The visual depiction effect in advertising, Elder and Krishna, Journal of Consumer Research, 2012
- Simultaneous modeling of visual saliency and value computation improves predictions of economic choice, Towal, Mormann and Koch, PNAS, 2013
- A review of the findings and theories on surface size effects on visual attention, Peschel and Orquin, Frontiers in Psychology, 2013
- Color psychology, effects of perceiving color on psychological functioning in humans, Elliot and Maier, Annual Review of Psychology, 2014
- The influence of banner advertisements on attention and memory, Sajjacholapunt and Ball, Frontiers in Psychology, 2014
- SALICON, saliency in context, Jiang, Huang, Duan and Zhao, CVPR, 2015
- The buffer effect, the role of color when advertising exposures are brief and blurred, Wedel and Pieters, Marketing Science, 2015
- Beyond memorability, visualization recognition and recall, Borkin and others, IEEE VIS, 2015
- DeepGaze II, reading fixations from deep features trained on object recognition, Kümmerer, Wallis and Bethge, arXiv, 2016
- Meaning based guidance of attention in scenes as revealed by meaning maps, Henderson and Hayes, Nature Human Behaviour, 2017
- Learning visual importance for graphic designs and data visualizations, Bylinskii and others, UIST, 2017
- Saliency benchmarking made easy, Kümmerer, Wallis and Bethge, ECCV, 2018
- Combining virtual reality and mobile eye tracking to provide a naturalistic experimental environment for shopper research, Meißner and others, Journal of Business Research, 2019
- The visual environment and attention in decision making, Orquin, Lahm and Stojić, Psychological Bulletin, 2021
- DeepGaze IIE, calibrated prediction in and out of domain for state of the art saliency modeling, Linardos and others, ICCV, 2021
- DeepGaze III, modeling free viewing human scanpaths with deep learning, Kümmerer, Bethge and Wallis, Journal of Vision, 2022
- TranSalNet, towards perceptually relevant visual saliency prediction, Lou and others, Neurocomputing, 2022
- Does text attract attention on e-commerce images, Jiang and others, CVPR, 2022
- UEyes, understanding visual saliency across user interface types, Jiang and others, CHI, 2023
- SUM, saliency unification through Mamba for visual attention modeling, Hosseini and others, WACV, 2025
- Modeling saliency dataset bias, Kümmerer and others, ICCV, 2025
Questions clients ask us.
Which data platforms does Freak use?
Semrush for search and AI visibility data, Jungle Scout for Amazon market data, Profound for how brands show up in AI assistants and Influencers.Club for creator data. Their data feeds our own tools in Freak Lab alongside each client's sales, reviews and product information.
Do you resell these platforms?
No. We license them for our own work. The value is in how the data is used, which is why it runs through our own tools and frameworks rather than the platforms' dashboards.
Is Freak a verified agency?
Yes. Freak is a DesignRush Verified Agency, listed among the digital marketing agencies in London. Our company profile is on LBB too.
Is Freak affiliated with MIT?
No. Like most teams working on visual attention we use research published by MIT and other universities. We credit it because our image scores rest on it. The papers are linked in the sources.
More from the Resource Centre.
Neuromarketing and AI image analysis, and how we weight the score.
The research behind our neural marketing framework, how we got here from VR eye tracking in 2019 and the weights we give each measure for search, gallery and A+ images.
Amazon listing copy built from evidence, and what it did for Feliway and Douxo.
How we turn reviews, search data and approved claims into Amazon copy that ranks, converts and answers the questions shoppers ask.
AI shopping agents trust brands that tell one product story everywhere.
AI agents read every version of your product page. How we build one consistent, claims checked product story for Amazon, brand sites, retailers and feeds.
Got a brief like this?
Tell us where you want to grow, and we'll show you how we'd get there.