Seedlabs

Attention-Driven Data Audit Tool

A visualization interface that uses eye-tracking to identify overlooked data points during manual reviews. The tool provides subtle visual cues to signal unexamined areas while monitoring cognitive load to prevent analyst fatigue.

Computer ScienceData Visualization and Analytics
Cognitive-Adaptive Visual AnalyticsJunior Compliance Officers at mid-sized fintechs performing manual KYC/AML reviews who currently rely on static spreadsheets and face high error rates due to oversight.
attention-driven-data-740d.seedlabs.tech

Live prototype, embedded from its own deployment — interact right here.

Open full app
Explainer video — the idea and its research foundation. Scripted and voiced by Claude Sonnet 5 and Piper · 2 Aug 2026.

Concept

This tool is designed to ensure comprehensive review of complex datasets by integrating gaze-tracking into the auditing workflow. The system tracks which specific data points have been viewed and for how long, using this data to reactively modify the visual representation—such as adjusting the opacity or color of elements via a CSS overlay or a Canvas-based heatmap—to signal which areas remain unexamined.

Technical Implementation (MVP)

To ensure feasibility for a first version, the tool will utilize consumer-grade webcams paired with JavaScript libraries like WebGazer.js. This allows for deployment without specialized hardware, using a browser-based implementation where the gaze coordinates are mapped to the DOM elements of the data table. Reactive cues are implemented as a semi-transparent CSS overlay that fades as the user's gaze dwells on a specific cell, providing a non-intrusive 'checklist' effect.

Evidence-Based Refinement

Research indicates that eye-tracking is a valuable instrument for quantifying behavioral patterns and cognitive differences in how users process information [3]. The integration of dual-coding (combining visual diagrams with text) can further reduce extraneous cognitive load, allowing users to split attention efficiently between different data formats [2]. Additionally, pupillometry can be used to infer cognitive load, enabling the tool to dynamically adjust the intensity of 'unseen' alerts based on the user's current mental effort [1].

Constraints and Caveats

There is a critical trade-off regarding visual complexity; overly complex visualizations can exceed a user's cognitive capacity, hindering decision-making and negatively affecting working memory [1]. To avoid 'over-signaling,' the tool's feedback loop must remain subtle. If the reactive highlighting becomes too visually noisy, it may create the very cognitive overload it intends to solve. Therefore, the system is bounded by a 'subtlety threshold' to ensure cues do not distract from the primary auditing task.

Written by Gemma 4 · 29 Jul 2026 · revised 2× with new research (latest 9 Sept 2026, Gemma 4)

AI assessment

Backed by 6 papers72

A high-risk but focused tool that attempts to solve a real compliance pain point using a specific technical wedge, though it relies on fragile consumer-grade eye-tracking accuracy.

Evidence strength
4/5
The idea is strongly supported by Paper [6], which explicitly proposes 'Attention-Aware Visualizations' to highlight unseen data, and is supplemented by papers on cognitive load and pupillometry.
Market pull
3/5
While KYC/AML oversight is a genuine pain point, the named beneficiaries (PwC, SEC) are more likely to buy systemic AI automation than a gaze-tracking UI overlay for junior staff.
Novelty & moat
3/5
Applying AAV to compliance is a clever niche, but the underlying technology (WebGazer.js) is open-source and the concept of a 'checklist' is common.
Feasibility
2/5
Webcam-based eye-tracking is notoriously imprecise for cell-level granularity in spreadsheets, making a reliable MVP difficult without specialized hardware.
Wedge clarity
5/5
The focus on Junior Compliance Officers performing manual KYC reviews is a sharp, well-defined entry point.
Simplicity / focus
5/5
The product is a single, focused utility—a visualization overlay—rather than a broad platform.

Scored by Gemma 4 · 9 Sept 2026. Rated against a fixed rubric (evidence, market, novelty, feasibility, wedge, simplicity). A prior estimate to compare ideas before real-world signal arrives.

Persona discussion

AI personas trained on real people's expertise debate this idea as it evolves.

View the discussion →

Act on this idea

Ideas only matter if someone runs with them. Your message goes straight to the founder's inbox — nothing is stored on our servers.

Business analysis

The analysis reveals a strong technological and economic fit due to the high cost of compliance errors, but highlights significant legal and social hurdles regarding biometric privacy and employee surveillance. The success of the tool depends on balancing the 'subtlety threshold' to avoid cognitive overload while navigating strict data protection laws.

Political2

Economic3

Social3

Technological3

Environmental2

Legal3

The tool operates in the highly regulated KYC/AML compliance space where legal requirements and data privacy laws are decisive factors. · Generated 2026-09-09 by Gemma 4AI-generatedFull PESTEL Analysis →

Who benefits

  • PwCcompany

    Auditors reviewing massive datasets for anomalies can use this to ensure no outlier or data point was accidentally skipped during a visual scan.

  • Deloittecompany

    Risk management consultants can use the tool to validate that their analysts have thoroughly examined all risk indicators in a dashboard.

  • Securities and Exchange Commission (SEC)organization

    Regulators analyzing complex trading patterns can ensure a rigorous, documented review of all visual evidence in a case.

Research it builds on

  1. Measuring Cognitive Load using Eye Tracking Technology in Visual Computing
    Johannes Zagermann, Ulrike Pfeil, Harald Reiterer · 2016 · 222 citations
    All ideas from this paper →
  2. Designing for Confidence: The Impact of Visualizing Artificial Intelligence Decisions
    Alexander J. Karran, Théophile Demazure, Antoine Hudon et al. · 2022 · 54 citations
    All ideas from this paper →
  3. Role of diagrams in problem solving: An evaluation of eye-tracking parameters as a measure of visual attention
    Ana Sušac, Andreja Bubić, Maja Planinić et al. · 2019 · 47 citations
    All ideas from this paper →
  4. Using Eye Tracking to Identify Cognitive Differences
    George E. Raptis, Christos Fidas, Nikolaos Avouris · 2016 · 31 citations
    All ideas from this paper →
  5. Users’ Cognitive Load: A Key Aspect to Successfully Communicate Visual Climate Information
    Luz Calvo, Isadora Christel, Marta Terrado et al. · 2021 · 30 citations
    All ideas from this paper →
  6. Attention-Aware Visualization: Tracking and Responding to User Perception Over Time
    Arvind Srinivasan, Johannes Ellemose, Peter Butcher et al. · 2024 · 9 citations
    All ideas from this paper →
Leave feedback
feasibility