Report Prepared for
Data Sources
SEO
Strong
Competitors
Strong
Accessibility
Improvement
Aesthetics
Improvement
Color Perception
Improvement
Perceptual Fluency
Improvement
Analytics
Requires Integration
User Research
Requires Integration
Ad Marketing
Requires Integration
Personalisation
Requires Integration
Usability Testing
Requires Upgrade
Consumer Psychology
Requires Upgrade
Prioritisation
Requires Upgrade
What-If Planning
Requires Upgrade


Atolls
eCommerce
Upgrade Required
Prioritisation
Conversion Planning
Consumer Trends
SUMMARY
We’ve developed a number of Machine Learning and AI models that purposefully track user journeys to easily identify areas causing conversion blockers, cognitive overload, predictive user actions, and overall user experience quality.

Cameron Henkes, Product Design and Consumer Psychology
RECOMMENDATION
We recommend:
1. Explore alternative UI designs that better align with Hick’s Law to reduce cognitive load.
Review upsell prompts to identify and resolve potential modal blindness issues.
Evaluate scroll depth to determine if sticky headers are necessary or simply adding friction.
Analyse the user journey in account sign-ups to pinpoint where drop-offs are most likely happening.
Use user research data to assess whether upsells to other brands align with user goals.
Explore top keywords commonly grouped by product, brand, or type. If users spend more time on the site, it’s worth testing upsells to other brands offering discounts within their interest range to boost session duration (e.g. Lands on Zalando, about to drop-off see upsell to a different brand with similar product offering). It might mean the difference between drop-off and conversion.
Brand = Zalando, Product=Garmin, Type=Donër
ASSUMPTIONS
The main goal of Atolls is to increase conversion from it's large traffic sources while exploring additional moneitisation opportunities.
SEO
Scanned
Site
pepper.com
SEO Score
92 / 100
Accessibility Score
79 / 100
Best Practices Score
100 / 100
Performance
52 / 100
Organic Traffic
6,092,152
SEO
Domain Ranking
62
Average Session Duration
12.29 M
Pages per Session
5.67
SEO Health
Your Top Keywords
pepper returns
13%
coolblue kortingscode
4.61%
thuisbezorgd kortingscode
3.58%
zalando kortingscode
3.58%
Site
promodescuentos.com
SEO Score
100 / 100
Accessibility Score
84 / 100
Best Practices Score
57 / 100
Performance
42 / 100
SEO
Domain Ranking
50
Organic Traffic
1,412,154
Average Session Duration
14:49 M
Pages per Session
7.56
SEO Health
Your Top Keywords
Site
hotukdeals.com
SEO Score
92 / 100
Accessibility Score
80 / 100
Best Practices Score
100 / 100
Performance
48 / 100
SEO
Domain Ranking
72
Organic Traffic
2,681,912
Average Session Duration
12.10 M
Pages per Session
6.58
SEO Health
Your Top Keywords
Site
dealabs.com
SEO Score
92 / 100
Accessibility Score
79 / 100
Best Practices Score
100 / 100
Performance
47 / 100
SEO
Domain Ranking
69
Organic Traffic
1,737,984
Average Session Duration
11:08 M
Pages per Session
7.19
SEO Health
Your Top Keywords
Site
preisjäger.at
SEO Score
100 / 100
Accessibility Score
81 / 100
Best Practices Score
100 / 100
Performance
61 / 100
SEO
Domain Ranking
47
Organic Traffic
111,316
Average Session Duration
9:53 M
Pages per Session
4.47
SEO Health
Your Top Keywords
Site
pepper.pl
Site
chollometro.com
SEO Score
92 / 100
Accessibility Score
72 / 100
Best Practices Score
100 / 100
Performance
57 / 100
SEO
Domain Ranking
56
Organic Traffic
1,367,776
Average Session Duration
11:27 M
Pages per Session
6.45
SEO Health
Your Top Keywords
Site
pepper.it
SEO Score
92 / 100
Accessibility Score
77 / 100
Best Practices Score
100 / 100
Performance
58 / 100
SEO
Domain Ranking
35
Organic Traffic
49,412
SEO Health
Your Top Keywords
66.66%
codice sconto zalando
33.33%
pcep certified entry level python programmer
0
pcep certified entry-level python programmer
0
Site
pepperdeals.se
SEO Score
92 / 100
Accessibility Score
79 / 100
Best Practices Score
100 / 100
Performance
60 / 100
SEO
Domain Ranking
27
Organic Traffic
39,309
SEO Health
Your Top Keywords
Accessibility
Improves
Active Users
100
Colour Blindness
Protanopia, Deuteranopia, Tritanopia
Simulates how a design appears to users with different types of color blindness, such as protanopia, deuteranopia, and tritanopia, ensuring inclusivity. The provided visualizations highlight how accessible the design is for color-blind users and indicate areas where color differentiation might fail.

Original

Protanopia

Deuteranopia

Tritanopianal
Based on Amazon Alexa Top 500 Global Sites
Aesthetics
Improves
Active Users
42
Neural Image Assessment (NIMA)
Predicting Human Opinion Scores
A machine learning model predicts the aesthetic and technical quality of images within a design. The mean score reflects the overall aesthetic quality, while standard deviation shows the consistency in quality across the interface.
Score
4.7318
Fair
Based on University of Trento, Trento, Italy
Color Perception
Improves
Page Loading
42
PNG File Size
Image Downloading
Evaluates the file size of PNG images, which impacts page load speed and performance. A "Fair" result indicates room for optimization to enhance loading efficiency.
Average PNG Size
438k
Fair
Based on University of Trento, Trento, Italy
Distinct RBG Values
Visual Complexity
Measures the diversity of colors used in the design by counting distinct RGB values. A high number of distinct RGB values indicates rich color diversity, while "Colorless" flags overly monochromatic designs.
Number of RGB Values
2293
Colourful
Based on University of Trento, Trento, Italy
Weighted Affective Valence Estimates (WAVE)
Color Likability
Assesses the emotional impact of colors based on their affective valence. "Fair" indicates moderate emotional resonance, requiring adjustments to evoke stronger reactions.
Average PNG Size
0.57
Fair
Based on Department of Psychology, University of California, Berkeley
Static Clusters
Color Interpretation
Groups similar colors that don’t change dynamically in a design. A "Fair" score suggests sufficient differentiation but may lack vibrancy.
Static Color Clusters
1527
Fair
Based on Department of Psychology, University of California, Berkeley
Luminance
Visual Complexity
Measures the variation in brightness levels across the design. High deviation suggests inconsistent brightness, which can cause visual discomfort or confusion.
Luminance Score
95.5227
Too High
Based on Department of Psychology, University of California, Berkeley
LAB Color Space
Visual Complexity
Evaluates the overall lightness and its variation across the design. High standard deviation highlights inconsistent lightness, requiring adjustments for balance.
Lightness Average
73.6011
Good
Lightness Score
37.0696
Too High
A (Green-Red) Average
1.4490
Meaningless
A (Green-Red) Score
8.1159
Meaningless
B (Blue-Yellow) Average
-0.5949
Meaningless
B (Blue-Yellow) Score
10.0074
Meaningless
Based on University of Trento, Trento, Italy
Colorfulness (Hasler and Süsstrunk)
Colors in Natural Images
Quantifies the richness and vibrancy of colors used in the design. A "Fair" score suggests adequate color vibrancy but potential for enhancement.
Colorfulness Score
26.7701
Meaningless
Based on Research by David Hasler and Sabine E. Suesstrunk
HSV Color Space
Colors in Natural Images
Analyzes the hue, saturation, and brightness levels in the design. Low saturation scores indicate muted colors, while high standard deviation in value suggests inconsistent brightness.
Hue Average
330.5742
Meaningless
Saturation Average
0.0484
Too Low
Saturation Score
0.1575
Too Low
Value Average
0.7471
Meaningless
Value Score
0.3650
Too Low
Based on Research by David Hasler and Sabine E. Suesstrunk
Distinct RGB Values per Dynamic Cluster
Color Perception Prediction
Analyzes the hue, saturation, and brightness levels in the design. Low saturation scores indicate muted colors, while high standard deviation in value suggests inconsistent brightness.
Distinct Hue Values
15.0531
Fair
Based on University of Trento, Trento, Italy
Color Harmony
Color Scheme Harmonic Distance
Evaluates the alignment of the color palette with harmonic templates for visual cohesion. A "Fair" result suggests a reasonably harmonious palette but potential for enhancement.
Distance
995.6681
Fair
Based on University of Trento, Trento, Italy
Perceptual Fluency
Improves
Time-to-Value (TTV)
42
Contour Density
Contour Pixels
Measures the complexity of visual contours in the design. "Fair" suggests moderate visual complexity but potential refinement for better clarity.
Score
0.0381
Fair
Based on University of Trento, Trento, Italy
Figure-Ground Contrast
Adjacent Color Difference
Evaluates the contrast between foreground elements and the background to ensure clarity. A "High" score confirms strong differentiation, aiding visual focus.
Figure-Ground Contrast
0.6912
High
Based on University of Bern, Switzerland
Contour Congestion
Adjacent Color Difference
Assesses the density of overlapping contours that can create visual clutter. "Fair" indicates moderate congestion, suggesting further decluttering for improved usability.
Figure-Ground Contrast
0.6912
Fair
Based on University of Calgary, Canada
Based on Department of Optometry and Neuroscience, University of California, Berkeley
Based on University of Bern, Switzerland
Subband Entropy
Interpretation of Imagery
Measures the randomness of visual information, impacting perceived clarity. "Fair" entropy suggests a balance between detail and simplicity but could be optimized for clearer visuals.
Subband Entropy
3.3137
Fair
Based on University of Trento, Trento, Italy
Department of Brain & Cognitive Sciences, MIT, Cambridge
Feature Congestion
Measurement of Visual Clutter
Quantifies the density of visual features that may overwhelm users. "Fair" indicates manageable congestion but room for simplification.
Feature Congestion
5.2739
Fair
Visualisation

Based on University of Trento, Trento, Italy
Department of Brain & Cognitive Sciences, MIT, Cambridge
Unified Model of Saliency and Importance (UMSI)
Predicting Human Attention
Highlights areas of the design that are most likely to draw user attention. The heatmap indicates focus points; bright areas suggest strong saliency, while dark areas may require enhancement.
Predicition

Predicition – Overlay

Department of Brain & Cognitive Sciences, MIT, Cambridge
Grid Quality
Identifiable Components
Evaluates the alignment and structure of visual blocks within the interface. "Meaningless" highlights poorly aligned or unstructured elements, suggesting a need for refinement.
No. Visual Blocks
46
Meaningless
No. Visual Blocks – No Children
33
Meaningless
No. Alignment Points
114
Meaningless
No. Alignment Points – No Children
84
Meaningless
No. Block Sizes
32
Meaningless
No. Block Sizes – No Children
21
Meaningless
GUI Coverage
0.0223
Meaningless
GUI Coverage – No Children
0.0190
Meaningless
No. Vertical Blocks
18
Meaningless
No. Vertical Blocks – No Children
13
Meaningless
Based on Hewlett-Packard Labs Research
Whitespace
User Cognitive Pressure
Assesses the presence and distribution of empty spaces in the design. "Meaningless" suggests ineffective use of white space, impacting balance and readability.
White Space
0.5045
Meaningless
Based on University of Trento, Trento, Italy
AIM Legacy Segmentation
Element Detection using Computer Detection
Divides the interface into distinct visual sections for easier analysis of layout and composition. The segmented image provides insights into the structural organization of the interface, highlighting potential improvements for clarity and usability.
Image

Based on Aalto University, Helsinki, Finland
Based on University of Trento, Trento, Italy
UIED Segmented Image
Element Detection with Deep Learning Models
Automatically segments a user interface into key components (e.g., headers, buttons, content areas) for a deeper understanding of its functional layout. The segmented image highlights areas where structural improvements can enhance usability and accessibility.
Image

Based on Australian National University, Australia
Predicted Human Attention
At different time intervals
Predicts which areas of the design will attract user attention at different time intervals. Bright areas in the heatmaps reflect high attention, suggesting effective placement of key elements.
0.5 Seconds

0.5 Seconds – Overlay

3 Seconds

3 Seconds – Overlay

5 Seconds

5 Seconds – Overlay

Based on Institute for Visualisation and Interactive Systems, University of Stuttgart
AI Chat
Requires Account Upgrade
You need to upgrade your account in order to access your AI Model trained on your data.
Contact Cameron