Frequently Asked Questions

We use a deep learning model called TranSalNet, which combines a DenseNet image feature extractor with a Transformer attention mechanism. It was trained on the SALICON dataset — approximately 10,000 images with real eye tracking fixation data collected from thousands of human participants. The output is a probability heatmap showing predicted visual attention.

Our saliency predictions achieve approximately 85–90% correlation with hardware-based eye tracking, validated against the MIT Saliency Benchmark. It's excellent for directional insight and creative optimisation — strong enough to catch major attention issues before you go live.

TranSalNet is trained on the SALICON dataset (Saliency in Context), the largest publicly available eye tracking dataset for saliency prediction. It contains approximately 10,000 images sourced from Microsoft COCO with fixation data collected from over 10,000 unique participants via webcam-based eye tracking. Each image was viewed by multiple people, so the heatmaps represent aggregated human attention — not one person's opinion, but the statistical consensus of where humans actually look.

Not yet. The model currently predicts general population attention — a 25-year-old and a 55-year-old are treated as the same pair of eyes. This is a known limitation shared by every synthetic eye tracking tool on the market today. Audience-segmented models are on our roadmap via an academic research partnership, and we'll announce this when it's ready.

Yes. Our Creative Effectiveness Report recommendations are grounded in published research, including Pieters & Wedel (2004) on attention capture in advertising (1,363 ads tested on 3,600+ consumers), Wedel & Pieters (2008) on eye-tracking research in marketing, and Xie et al. (2023) on TranSalNet's saliency prediction architecture. All citations are available in our methodology documentation.

We support JPEG, PNG, and WebP image files. You can upload screenshots of websites, social media posts, ad creative, packaging designs — anything visual. Video analysis is on our roadmap but not yet available.

Typically under 30 seconds from upload to a complete heat map and report. The AI inference runs on dedicated GPU infrastructure so results are near-instant regardless of image complexity.

A heat map shows you where people look. A Creative Effectiveness Report tells you what to do about it. The report includes visual saliency scoring, CTA and logo visibility analysis, gaze path prediction, clutter scoring, and specific research-backed recommendations to improve your creative. The full report is available on Agency plans and above.

Agency adds Zone Tagging (define specific regions of interest), Scan Path animation (visualise the predicted sequence of eye movement), the full Creative Effectiveness Report, PDF export, and priority support. Starter includes heatmap overlay and analysis with priority email support.

Yes. Upload variants into the same project and compare attention metrics side-by-side. This is available on all plans and is one of the most valuable workflows — test A/B variants before media spend, not after.

No. Never. Your images are processed through our pre-trained saliency model and the results are returned to you. The model does not learn from your data. Your creative assets remain your property and are never used for any purpose beyond generating your requested analysis.

Your data is stored on Supabase (database, SOC 2 Type II certified) and Cloudflare R2 (image storage, ISO 27001 certified). AI inference runs on Modal (SOC 2 Type II certified) where images are processed in memory and not persisted on inference servers. All data is encrypted in transit (TLS 1.3) and at rest (AES-256).

No. All data is isolated per customer at the database level. There is no shared access, no shared storage, and no cross-customer visibility.

Yes. Email privacy@veysalabs.io and we'll send you our standard Data Processing Agreement within 24 hours. It covers GDPR and PDPA requirements.

Not yet. We are aligned with the ISO 27001 framework and our infrastructure providers hold independent certifications (SOC 2 Type II, ISO 27001, PCI DSS). Full certification is on our roadmap and will be pursued when enterprise demand requires it.

Yes. We comply with both the Singapore PDPA and EU/UK GDPR. We have a published Privacy Policy, a standard Data Processing Agreement, and a subprocessor list. We use Standard Contractual Clauses for international data transfers and our infrastructure providers all hold relevant certifications.

The Beta plan gives you 1 seat with 25 analyses total (no monthly reset) and includes heatmap overlay and analysis. It's designed to let you experience the platform before committing. Beta access is available for a limited time only.

Yes. You can upgrade at any time and the change takes effect immediately. Downgrades apply at the next billing cycle. There are no lock-in contracts on any plan.

You'll be notified as you approach your limit. Once reached, you can upgrade your plan or wait for the next billing cycle to reset. We won't charge overage fees without your explicit consent.

Enterprise plans start from $799/month and are customised based on your team size, analysis volume, and integration requirements. This includes unlimited analyses, custom seats, everything in the Agency plan, plus a dedicated account manager. White-label, SSO, and audience segmentation features are coming soon. Contact us for a tailored quote.