Analytics Overview

Problem

The existing Analytics experience was data-rich but insight-poor.
📈 No centralised view that told a coherent performance story
📈 High cognitive load too many charts, filters, and technical terms
📈 Unclear value for leadership teams struggled to answer basic questions like “Are we improving?” or “Where are the biggest risks?”

Outcome

📈 Shifted product and leadership conversations from “Which charts should we show?” to “What decisions must this data enable?”
📈 Established a repeatable analytics model that now guides how new data modules are structured and evaluated.
📈 Prioritized insights based on the decision-making needs of security stakeholders, rather than raw data volume or chart count. ‍

My Contribution

I led the research and reframed Analytics around decision-making, not data presentation.

Discovery opportunity

Identify which metrics are most valuable for AppSec users to measure performance, and how they prefer to explore data when answering questions or proving impact.

Figure 1: Initial production design

Methods

Run workshop sessions with customers
•  Top-of-mind metrics exercise ( Left)  → uncover which numbers users instinctively track or report first.
•  Wireframe Design review ( right)  → tested early explorations with internal stakeholders and select customers.

Figure 2: Design methods diagram

From Patterns to a New Analytics Model (JTBD)

Through interviews, workflow mapping, and real dashboard reviews, I identified five core Jobs-to-be-Done that Analytics needed to support:
1. Strategic evaluation
2. Investment justification
3. Investigation & prioritization
4. Monitoring & reporting
5. Collaboration

Using these insights, I created Jobs-to-Be-Done models for our two primary personas:

Figure 3: Strategic persona JTBD

Figure 4: Operational persona JTBD

Analytics model

Zooming out from the JTBD, a universal analytics flow emerged:

📈 Question - Each begins with an intent or a prompt.
• Strategic persona: “Are we improving overall?”
• Operational persona: “Where are the risks?”
‍
📈 Explore Data - They navigate metrics or drill deeper into details.
‍
📈 Summarize & Interpret - They review trends, patterns, or comparisons to form an understanding of the situation.
‍
📈  Act on Insights - They take action based on findings.
• Strategic persona: Shares or presents summaries to leadership.
• Operational persona: Prioritizes issues, assigns tasks, or communicates progress.

Figure 5: Analytic model storyboard

Initial Sketch

I sketched multiple flows exploring how AI could simplify the experience:
📈 Ask a question → AI insights
📈 Search field → AI insights
📈 Hybrid: predefined dashboards +  AI insights
📈 Hybrid: filters +  AI insights

Figure 6: Initial sketch

Interaction Flow

The early interaction model mapped the user journey:
📈 Explore data → drill into relevant report
📈 Ask a question → result → refine → drill further
📈 Select a saved view → interpret → act
This visualized the shift toward a question-first analytics experience.

Figure 7: Interaction flow (early draft)

Data Visualization Principles

To ensure clarity and consistency, I defined four principles now used across Snyk’s analytics visuals:

📈  Highlight what matters — Reduce noise; emphasize trends
📈  Support exploration — Layered drill-down paths
📈  Respect hierarchy — Clear reading order and scannability
📈  Standardize visuals — Consistent chart types and interactions

Figure 8: Experimenting with different visualization approaches to balance clarity and actionability.