From Information Overload to Focused Insights: A Traffic Dashboard Redesign

STWD Challenge - Dashboards that Deliver

SWDchallenge

2025

Shiny

A dashboard redesign case study transforming a cluttered multi-tab interface into a focused single-page solution for transport planners. Shows how strategic design choices can dramatically improve user experience and analytical clarity.

Author

Steven Ponce

Published

September 3, 2025

The Challenge

This month’s Storytelling with Data challenge focused on “Dashboards that Deliver” - revisiting existing dashboards to make them more focused and user-centric. I chose to redesign my own traffic flow dashboard using the TidyTuesday (2024 wk 49) National Highways Traffic Flow dataset.

The Problem: Information Overload

Original Dashboard - The “Before”

Figure 1: Baseline dashboard with multiple tabs, cluttered interface, and unclear focus

Key Issues Identified:

Multiple tabs scattered related insights across different views
Too many controls (sensor picker, date range, day type, hour slider, plotly toggle, KPI toggle)
Unclear audience - tried to serve everyone, served no one well
No visual hierarchy - everything competed for attention
Analysis paralysis - users didn’t know where to start or what questions to ask

The original dashboard suffered from a common problem: attempting to show everything at once rather than guiding users toward the most important insights.

Audience & Analytical Needs

Target Users: Transport Planners & Local Policy Staff

Primary Questions They Need Answered:

How is daily volume trending this month? (For capacity planning)
What’s the weekday vs weekend hourly profile? (For staffing and maintenance scheduling)
Which sensor shows consistently different patterns? (For identifying problem areas)

Secondary Needs:

Quick overview metrics at a glance
Ability to focus on specific date ranges
Simple comparison between weekday/weekend patterns
Minimal cognitive load - they’re busy people

The Solution: Focused Clarity

Redesigned Dashboard - The “After”

Figure 2: Clean, focused dashboard with three KPIs and two strategic charts

Key Design Decisions

1. Single Page Layout

Eliminated tabs to keep related insights together
Created clear visual hierarchy with KPIs at top

2. Streamlined Controls (3 vs 8)

Date range selector (essential for time-series analysis)
Sensor selection (allows focus on specific locations)
Weekday/Weekend toggle (key operational insight)

3. Strategic KPIs

Avg daily volume: 28,906 (capacity planning)
Median speed: 44 mph (performance monitoring)
% large vehicles: 3.8% (infrastructure wear planning)

4. Two Focused Charts - Daily traffic volume with trend line (answers trending question) - Weekday vs Weekend profile (answers operational scheduling question)

Interactive Versions

Baseline App (Before): https://0l6jpd-steven-ponce.shinyapps.io/01_before_app/
Makeover App (After): https://0l6jpd-steven-ponce.shinyapps.io/02_after_app/

Navigation: Use the simple sidebar controls to filter by date range, select specific sensors, or toggle between weekday and weekend patterns.

Design Impact & Results

What Changed

Reduced cognitive load from 8 controls to 3
Eliminated navigation from 5 tabs to 1 page
Focused insights from scattered metrics to strategic KPIs
Clear visual hierarchy guiding users from summary to detail

Questions Now Easily Answered

✅ “How’s traffic trending?” - Immediately visible in main chart with trend line
✅ “When are peak hours?” - Clear hourly profile shows morning/evening peaks
✅ “Weekend vs weekday differences?” - Toggle comparison built into interface
✅ “Which location needs attention?” - Sensor comparison readily available

What Users Gain

Faster insights - key metrics visible within seconds
Actionable intelligence - focused on decisions they actually need to make
Reduced training time - intuitive interface needs minimal explanation
Mobile-friendly - simplified layout works on smaller screens

Technical Implementation

The redesign was built using R Shiny with the National Highways Traffic Flow dataset from TidyTuesday 2024 Week 49, covering A64 road sensors during May 2021.

Data Source Details

Source: WebTRIS Traffic Flow API via TidyTuesday (2024 wk 49)
Coverage: 4 sensors on A64 road, May 2021
Granularity: Hourly vehicle counts, speeds, and size classifications
Volume: ~30K daily average across all sensors

Key Technical Choices

Single reactive dataset instead of multiple data processing streams
Conditional rendering based on user selections rather than separate tabs
Optimized calculations focusing on the three key metrics that matter most
Simplified interactivity - removed unnecessary plotly complexity

Reflection & Lessons Learned

What Worked Well

Less is more - Removing features improved usability dramatically
User-centered design - Focusing on specific audience transformed the experience
Visual hierarchy - Clear information architecture guides user attention effectively

What I’d Do Differently Next Time

User testing - Would validate assumptions with actual transport planners
Progressive disclosure - Could add “deep dive” option for power users
Contextual help - Brief explanations of what “good” looks like for each metric

Key Takeaway

The most powerful dashboards don’t show everything possible - they show everything necessary for specific users to make specific decisions.

Session Info

Expand for Session Info

R version 4.4.1 (2024-06-14 ucrt)
Platform: x86_64-w64-mingw32/x64
Running under: Windows 11 x64 (build 22631)

Matrix products: default


locale:
[1] LC_COLLATE=English_United States.utf8 
[2] LC_CTYPE=English_United States.utf8   
[3] LC_MONETARY=English_United States.utf8
[4] LC_NUMERIC=C                          
[5] LC_TIME=English_United States.utf8    

time zone: America/New_York
tzcode source: internal

attached base packages:
[1] stats     graphics  grDevices datasets  utils     methods   base     

loaded via a namespace (and not attached):
 [1] htmlwidgets_1.6.4 compiler_4.4.0    fastmap_1.2.0     cli_3.6.4        
 [5] htmltools_0.5.8.1 tools_4.4.0       rstudioapi_0.17.1 yaml_2.3.10      
 [9] rmarkdown_2.29    knitr_1.49        jsonlite_1.8.9    xfun_0.49        
[13] digest_0.6.37     rlang_1.1.6       renv_1.0.3        evaluate_1.0.1

Code Repository

Expand for Code Access

QMD File: swd_2025_09.qmd
Baseline App: GitHub Repository
Makeover App: GitHub Repository
Full Repository: Personal Website Repo

References

Expand for References

Challenge Details: - STWD Monthly Challenge

Data Sources: - TidyTuesday 2024 Week 49 - National Highways WebTRIS API

--- title: "From Information Overload to Focused Insights: A Traffic Dashboard Redesign" subtitle: "STWD Challenge - Dashboards that Deliver" description: "A dashboard redesign case study transforming a cluttered multi-tab interface into a focused single-page solution for transport planners. Shows how strategic design choices can dramatically improve user experience and analytical clarity." author: "Steven Ponce" date: "2025-09-03" categories: ["SWDchallenge", "2025", "Shiny"] tags: [ "storytelling-with-data", "dashboard-design", "user-experience", "traffic-analytics", "transport-planning", "data-visualization", "shiny-apps", "before-after", "design-principles", "focused-analytics", "tidytuesday", "national-highways", "a64-traffic" ] image: "thumbnails/swd_2025_09.png" format: html: toc: true toc-depth: 4 code-link: true code-fold: true code-tools: true code-summary: "Show code" editor_options: chunk_output_type: inline execute: freeze: true cache: true error: false message: false warning: false eval: true --- ## The Challenge This month's Storytelling with Data challenge focused on **"Dashboards that Deliver"** - revisiting existing dashboards to make them more focused and user-centric. I chose to redesign my own traffic flow dashboard using the TidyTuesday (2024 wk 49) National Highways Traffic Flow dataset. ## The Problem: Information Overload ### Original Dashboard - The "Before" ![Baseline dashboard with multiple tabs, cluttered interface, and unclear focus](https://raw.githubusercontent.com/poncest/SWDchallenge/main/2025/09_Sep/img/01_before_app.PNG){#fig-before} **Key Issues Identified:** - **Multiple tabs** scattered related insights across different views - **Too many controls** (sensor picker, date range, day type, hour slider, plotly toggle, KPI toggle) - **Unclear audience** - tried to serve everyone, served no one well - **No visual hierarchy** - everything competed for attention - **Analysis paralysis** - users didn't know where to start or what questions to ask The original dashboard suffered from a common problem: attempting to show *everything* at once rather than guiding users toward the most important insights. ## Audience & Analytical Needs ### Target Users: Transport Planners & Local Policy Staff **Primary Questions They Need Answered:** 1. **How is daily volume trending this month?** (For capacity planning) 2. **What's the weekday vs weekend hourly profile?** (For staffing and maintenance scheduling)\ 3. **Which sensor shows consistently different patterns?** (For identifying problem areas) **Secondary Needs:** - Quick overview metrics at a glance - Ability to focus on specific date ranges - Simple comparison between weekday/weekend patterns - Minimal cognitive load - they're busy people ## The Solution: Focused Clarity ### Redesigned Dashboard - The "After" ![Clean, focused dashboard with three KPIs and two strategic charts](https://raw.githubusercontent.com/poncest/SWDchallenge/main/2025/09_Sep/img/02_after_app.PNG){#fig-after} ### Key Design Decisions **1. Single Page Layout** - Eliminated tabs to keep related insights together - Created clear visual hierarchy with KPIs at top **2. Streamlined Controls (3 vs 8)** - Date range selector (essential for time-series analysis) - Sensor selection (allows focus on specific locations) - Weekday/Weekend toggle (key operational insight) **3. Strategic KPIs** - **Avg daily volume:** 28,906 (capacity planning) - **Median speed:** 44 mph (performance monitoring) - **% large vehicles:** 3.8% (infrastructure wear planning) **4. Two Focused Charts** - **Daily traffic volume** with trend line (answers trending question) - **Weekday vs Weekend profile** (answers operational scheduling question) ## Interactive Versions - **Baseline App (Before):** https://0l6jpd-steven-ponce.shinyapps.io/01_before_app/ - **Makeover App (After):** https://0l6jpd-steven-ponce.shinyapps.io/02_after_app/ *Navigation: Use the simple sidebar controls to filter by date range, select specific sensors, or toggle between weekday and weekend patterns.* ## Design Impact & Results ### What Changed - **Reduced cognitive load** from 8 controls to 3 - **Eliminated navigation** from 5 tabs to 1 page - **Focused insights** from scattered metrics to strategic KPIs - **Clear visual hierarchy** guiding users from summary to detail ### Questions Now Easily Answered ✅ **"How's traffic trending?"** - Immediately visible in main chart with trend line\ ✅ **"When are peak hours?"** - Clear hourly profile shows morning/evening peaks\ ✅ **"Weekend vs weekday differences?"** - Toggle comparison built into interface\ ✅ **"Which location needs attention?"** - Sensor comparison readily available ### What Users Gain - **Faster insights** - key metrics visible within seconds - **Actionable intelligence** - focused on decisions they actually need to make - **Reduced training time** - intuitive interface needs minimal explanation - **Mobile-friendly** - simplified layout works on smaller screens ## Technical Implementation The redesign was built using R Shiny with the National Highways Traffic Flow dataset from TidyTuesday 2024 Week 49, covering A64 road sensors during May 2021. ### Data Source Details - **Source:** WebTRIS Traffic Flow API via TidyTuesday (2024 wk 49) - **Coverage:** 4 sensors on A64 road, May 2021 - **Granularity:** Hourly vehicle counts, speeds, and size classifications - **Volume:** \~30K daily average across all sensors ### Key Technical Choices - **Single reactive dataset** instead of multiple data processing streams - **Conditional rendering** based on user selections rather than separate tabs - **Optimized calculations** focusing on the three key metrics that matter most - **Simplified interactivity** - removed unnecessary plotly complexity ## Reflection & Lessons Learned ### What Worked Well - **Less is more** - Removing features improved usability dramatically - **User-centered design** - Focusing on specific audience transformed the experience - **Visual hierarchy** - Clear information architecture guides user attention effectively ### What I'd Do Differently Next Time - **User testing** - Would validate assumptions with actual transport planners - **Progressive disclosure** - Could add "deep dive" option for power users - **Contextual help** - Brief explanations of what "good" looks like for each metric ### Key Takeaway The most powerful dashboards don't show everything possible - they show everything *necessary* for specific users to make specific decisions. ------------------------------------------------------------------------ ## Session Info ::: {.callout-tip collapse="true"} ##### Expand for Session Info ```{r, echo = FALSE} sessionInfo() ``` ::: ## Code Repository ::: {.callout-tip collapse="true"} ##### Expand for Code Access - **QMD File:** [swd_2025_09.qmd](https://github.com/poncest/personal-website/tree/master/data_visualizations/SWD%20Challenge/2025/swd_2025_09.qmd) - **Baseline App:** [GitHub Repository](https://github.com/poncest/SWDchallenge/tree/main/2025/09_Sep/01_before_app) - **Makeover App:** [GitHub Repository](https://github.com/poncest/SWDchallenge/tree/main/2025/09_Sep/02_after_app) - **Full Repository:** [Personal Website Repo](https://github.com/poncest/personal-website/) ::: ## References ::: {.callout-tip collapse="true"} ##### Expand for References **Challenge Details:** - [STWD Monthly Challenge](https://community.storytellingwithdata.com/challenges/sep-2025-dashboards-that-deliver) **Data Sources:** - [TidyTuesday 2024 Week 49](https://github.com/rfordatascience/tidytuesday/tree/main/data/2024/2024-12-03) - [National Highways WebTRIS API](https://webtris.nationalhighways.co.uk/api/swagger/ui/index) :::