Case Study: Product Design & Strategy

Puck
Line
Pro

Designing an AI-powered hockey betting dashboard that turns predictive intelligence into a winning edge.

The Product

PuckLine Pro is an AI-powered sports betting platform purpose-built for NHL fans and bettors. It combines real-time fixture data, machine-learned odds analysis, and a clean dashboard experience to give users a measurable edge over traditional betting platforms.

The platform distills complex statistical models into simple, actionable picks, surfacing the right bet at the right time, powered by a proprietary AI engine called BTRZedge.

Client
PuckLine Pro
Role
Product Design Lead
Domain
Sports Betting AI
Platform
Web App (Desktop)
Industry
iGaming / FinTech
Year
2020

What Was Built

Scope

  • End-to-end dashboard UI/UX design for a betting AI web application
  • AI-powered Top Picks feed with real-time odds display (Home / Away / AI Pick)
  • Individual game bet view with player spotlights and main/side bet categories
  • Bet Slip component: live stake management and return estimation
  • Schedule & Table widget with filterable calendar view and conference standings
  • AI Report module with Ask-AI prompt input and bubble chart visualization
  • My Report section with active bet donut chart and tag-based segmentation
  • Next Fixtures ticker with team matchup logos
  • User balance, notification, and navigation system

Tools

FigmaFigJamFramerReactTypeScriptTailwind CSSOpenAI APIRechartsNode.jsPostgreSQLSportsRadar APIVercelLottieNotionLinear

14 Weeks

Phase 01
Discovery & Research
Competitive audit of existing betting platforms (DraftKings, Bet365, Betway). User interviews with casual and semi-pro NHL bettors. Identification of friction points: information overload, lack of AI guidance, poor mobile bet-slip UX.
Weeks 1–2
Phase 02
Information Architecture & Wireframes
Mapped user journeys from fixture discovery to bet submission. Defined dashboard layout hierarchy (AI picks surface first), calendar and standings persist. Low-fidelity wireframes in FigJam validated with stakeholders.
Weeks 3–4
Phase 03
Visual Design System
Established dark-first design language: deep carbon backgrounds, neon-green AI accents, amber for caution states. Defined component library: odds pills, team badges, calendar cells, bet-slip cards, chart styles.
Weeks 5–7
Phase 04
AI Feature Design
Designed the BTRZedge AI flow: how picks surface, how confidence scores display, how the Ask-AI input works. Collaborated with ML team to define output schemas and edge-case states (low-confidence, live game, suspended match).
Weeks 8–10
Phase 05
Prototyping, Testing & Handoff
Interactive Figma prototype tested with 18 participants. Iterated on bet-slip clarity, AI pick labeling, and calendar interaction. Full developer handoff with annotated specs, component tokens, and motion guidelines.
Weeks 11–14

The Objective

Make AI betting intelligence accessible, not intimidating.

The core goal was to design a betting platform where AI recommendations feel natural, trustworthy, and immediately actionable. The product needed to serve two user archetypes: the casual NHL fan placing a first bet, and the experienced bettor who wants deeper data on demand. Both should find value within 60 seconds of opening the dashboard, without onboarding, without friction, and without information overload.

What We Were Up Against

01
Trust in AI Picks
Users were deeply skeptical of AI-generated betting advice. Early testing showed that unexplained AI picks increased anxiety rather than confidence. Establishing credibility without overwhelming users with statistical jargon was a core design tension.
02
Information Density vs. Clarity
A hockey betting dashboard must surface fixtures, odds, standings, AI picks, bet slips, and calendar data simultaneously. The challenge was achieving this density without the interface feeling chaotic or requiring users to hunt for key actions.
03
Real-Time Data Freshness
Odds shift constantly. Players get injured. Lines move. The UI needed to communicate data recency clearly, distinguishing between live AI recommendations and pre-game analysis, without alarming users or triggering unnecessary refreshes.

How We Solved It

01

Hierarchy-First Dashboard Layout

The main dashboard was structured around a clear visual hierarchy: the Top Picks feed dominates the primary content area with Home, Away, and AI Pick odds always visible. The AI Pick column uses a distinct color treatment to signal provenance without requiring a label read. Peripheral data (Schedule, Bet Slip) occupies the right rail, accessible but never competing.

02

BTRZedge AI Branding & Transparency Layer

To address trust, the AI was given a distinct identity, BTRZedge Verified by AI, displayed prominently on the individual game screen alongside real player imagery. This humanized the recommendations. An "Ask AI" prompt input in the My Report panel further positioned AI as a responsive advisor, not an opaque oracle.

03

Progressive Disclosure on Game Pages

The individual bet page (Dallas Stars vs Arizona Coyotes) used a progressive disclosure model: hero images establish emotional context, player spotlights add depth, and Main Bets / Players / Game tabs allow users to go as deep as they choose. Bets are scannable at a glance but expandable for detail.

04

Persistent Live Bet Slip

The Bet Slip panel was fixed in the right rail across both dashboard states, always showing current stakes, multipliers, and estimated returns. This persistent context prevents the common error of losing track of active bets. Color-coded bet rows and inline ×close actions keep management effortless.

05

Schedule × Table Widget with Tag Filtering

The calendar component was designed with a dual-mode toggle (Schedule / Table) and persisted tag filters, allowing users to quickly segment upcoming fixtures by team or competition. Highlighted date cells use a subtle green glow to draw attention to event-rich days without demanding it.

Delivered Screens
Dashboard: Home View
Balance$2,886
Vegas Golden Knights vs Arizona CoyotesDallas Stars vs Nashville Predators
TOP PICKS
Nashville Predators
1.254.30AI
Edmonton Oilers
1.254.302.28
Winnipeg Jets
8.252.571.27
OCTOBER
891011121314171819202122232425
BET SLIP
Arizona Coyotes ×3.8 → $80.5
MY REPORT: 29 Total Active
29
Tag ItemTag Item
AI REPORT
Ask AI
2
Title
14
Title
4
Title
Dashboard: Bet / Game View
← Back to Main Page
Dallas Stars
Arizona Coyotes
BTRZedge Verified by AIReal-Time Betting Advice →
A. Radulov
M. Heiskanen
J. Benn
M. Hanzal
Main BetsPlayersGame
Arizona Coyotes: To Win×3.8
Arizona Coyotes: To Draw×1.6
San Jose Sharks×5.6

The Outcome

+68%
Increase in bet slip completion rate vs. competitive baseline
41s
Average time to first placed bet in usability testing (down from 3m+)
4.7
Trust score for AI picks (out of 5) after redesign, up from 2.9
92%
Task success rate across 5 core user journeys in final prototype testing

"The design finally made me feel like the AI was on my side, like a betting partner, not a slot machine."

Beta User, NHL Bettor, Toronto