Console Dashboard - Insights

mockuuups-mockup-of-user-staring-at-his-imac-when-working (2).jpeg

Navigation

Insights

CONSOLE DASHBOARD - INSIGHTS

Redesigned the navigation architecture to streamline access to customer and product insights. Conducted comprehensive audits to revamp data visualizations, filtering, and layouts, to ensure consistency, improve usability, and accommodate insights from multiple channel sources. All Anewgo customers now utilize the 3.0 Insights dashboard.

ROLE

After working on the high-level unified dashboard navigation, I collaborated with the PM and sales managers to understand mental models, map out the new navigation, and audit the data visualization. Later on, I collaborated with QA to pinpoint UX/UI issues and developers to aid with the accuracy of implementation.

CHALLENGE

Despite having valuable customer and product data, users struggle with navigation and relevant insights. The 2.0 dashboard also focused on a single channel which won't align with the upcoming release of the 3.0 Anewgo multi-channel ecosystem

The wealth of lead and product analytics we provide are typically underutilized by home builders due to the cluttered UI, disorganized navigation, and a lack of awareness about available features.

Additionally, the original My Home App 2.0 which is a visualization and design web app catered to each builder serves as the source of the Insights dashboard. However, potential buyers can engage with Anewgo products through various entry points, including the Sales Center App at physical sales centers, Anewgo design tools embedded into builder websites, or Anewgo-supported home builder websites."

APPROACH & IMPACT

A multi-channel analytics dashboard that mirrors the homebuyer journey with a consistent and approachable UI that aids in learnability

There were two potential causes of the discontent in the effectiveness and usability of the Insights Dashboard. Either the UI did not offer proper visual feedback to extract insights from the data analytics or the navigation was confusing and overwhelming which led to flow abandonment. Both were identified as the root causes. So I collaborated with our sales managers to understand key metrics and mental models that served as the foundation of the new architecture. I then conducted UI audits of both the global layout and specific data visualization panels to resolve UI issues, enhance UX, and introduce multi-channel compatibility for the 3.0 launch. As a result, the outdated 2.0 dashboard has been retired, and hundreds of home builders have successfully transitioned to the improved 3.0 version.

DISCOVER

BACKGROUND

What was the Insights Dashboard 2.0?

The original Insights dashboard contained four specific areas:

Lead Analytics - lead database, buyer preferences and interactions
App Analytics - community, elevation, lots, and floor plan analytics
Marketing Automation - email campaigns and email templates
Buy Now - homesite reservation and down payment feature

Who are the main users of the Insights dashboard?

The dashboard is used by a myriad of teams both from our external customers and in-house employees.

Sales Teams - track and manage leads
Marketing Teams - campaign performance, customer engagement
Design Teams - floor plan, exterior/interior popularity
Product Teams - assess market demand
Customer Success Teams - aid and educate Anewgo customers

DEFINE

INFORMATION ARCHITECTURE

Overhauled both the high-level and sub-level navigation greatly reducing the need for long-scrolling pages

In the first iteration, I consolidated the four modules into two. The Buy Now feature was discontinued and the marketing module can be integrated further into the Leads Insights, leaving Leads and Analytics as the primary modules. Despite this consolidation, we are still left with lengthy and overwhelming pages due to the lack of specificity of the different data panels.

In the next iteration, I worked with sales managers to identify key metrics valued by customers and found that their expectations clashed with the product-centric architecture of the current dashboard. So to further fit it into their mental models, I sub-divided the pages into page tabs organized by key stages of the homebuyer journey: communities, plans, elevations and floor plans, exteriors and interiors. Furthermore, we renamed the high level categories into “Customers” and “Products” which aligns more accurately to the analytics data

JOIST Research_Brainstorm Workspace (13).jpg

JOIST Research_Brainstorm Workspace (15).jpg

DESIGN

DATA ANALYTICS UI/UX AUDIT

Focused on ensuring consistency in the global analytics UI while refining the specific UX of individual data tables

When conducting the audit, I used a 4-point scale to determine the severity of an issue.

1 - minor UI problem (subjective, aesthetics)
2 - major UI problem (contrast, consistency, may affect UX)
3 - minor UX problem
4 - major UX problem (affects core user flow/goal)

For each issue, I also offered some recommendations to resolve the issue and included notes based on stakeholder input. I identified UI inconsistencies, such as misaligned tables, inconsistent legends, and varying icon styles. Additionally, UX issues emerged, including barely noticeable active states in charts and a general lack of helper text for guidance

In the original UI, layouts for the data panels had inconsistent widths depending on the data being shown, which resulted in a cluttered-looking UI. As a result, users struggle to distinguish between important data points or miss out on key insights due to the illogical layout. The main users for the analytics dashboard largely vary from sales, marketing, home agents, and researchers so wasting time to interpret the data can be a frustrating experience for everyone.

Additionally, the inefficient use of whitespace also contributed to unnecessarily long web pages.

DASHBOARD LAYOUT

How might we optimize the data panel layout layout to reduce visual clutter and ensure a more digestible interface?

In the original UI, date filtering was restricted to selecting time ranges, which limited its usability. As this is a frequently used feature, allowing users to quickly select common or specific timestamps is a more efficient method, especially in smaller breakpoints. Additionally, the overall filtering functionality and UI lack organization. The elements are inconsistent in style and scattered with no logical grouping to guide the user.

In terms of breadth of analytics. 2.0 only receives data from the builder-specific My Home App, where majority of our traffic comes from. So in 3.0, the primary requirement is to integrate and display analytics from the other Anewgo channels and touchpoints.

Based on conversations from sales managers, their customers often complain about the tedious process of filtering to specific date ranges. To address this, I enhanced the date range filtering by adding commonly used time periods such as "Last 7 Days," "Last 14 Days," and "Last 6 Months." This improvement is especially beneficial for mobile users, as it reduces the number of clicks required. In terms of placement, I intentionally separated the date filter due to its frequent use to ensure better visibility and accessibility.

I decided to separate the date filter from the other filter options since the remaining filters pertain to the home's location or specifications. To streamline the UI, I grouped all non-date filters under a "Location Filters" button and data-specific ones such as "Material" or "Color" were included into the data panel itself. The location filters button opens in a drawer panel—an element commonly used across the other sub-dashboards.

This approach also considers future scalability. As the insights dashboard evolves into an analytics hub for the entire Anewgo ecosystem, it will accommodate more specific filters without cluttering the interface.

combined_screenshots_with_drawing (96).png

DATE AND FILTERING

How might we help users quickly and easily filter the relevant data they need while still allowing for customization?

Originally called the Leads Analytics, it contained potential homebuyer statistics and details such as sign-in frequency, rating, name, email, etc. It was divided into two parts:

Leads Table - This page contained three tables for new, returning, and active leads. These were all in one page which was overwhelming as each table had different column categories. If there were a lot of data, the active leads, which is arguably the most important table is buried to the bottom of the page.
User Traffic - All analytics under the sun about leads were thrown into this bin and so little value is seen by some users as they don’t typically even visit this page or can’t really tell what they’re looking for.

I organized the "New," "Returning," and "Active" Leads tables into three distinct tabs, aligning each with its corresponding user actions. This approach improves clarity and makes the tables easier to scan. Additionally, I consolidated all location filters into a single button, keeping the name search bar as the primary filtering tool.

To address user feedback requesting more actionable data beyond static screenshots, I added an "Export CSV" button, to allow users to download the data in an accessible format.

For the lead profile I wanted to understand which data agents find valuable when evaluating a lead. I collaborated with sales managers and asked them which stats or markers they commonly use to gauge a person's interest in a home. Based on their insights, I implemented the following design changes:

Removed the pop-up that appears when clicking on a lead as it just duplicates info already in the lead profile.
Reorganized the profile layout to make contact information and engagement analytics easily scannable.
Created dedicated tabs at the top of the page for critical data markers such as favorited homes and selected design options.
Added a streamlined method for assigning agents to leads directly within the profile page.

Marketing has also been bundled under Customers instead of having its own separate page. This creates a more natural flow when agents want to follow-up on a lead or create a specific marketing strategy as they can send emails or suggest plans directly from that lead's 's profile.

CUSTOMER ANALYTICS

How might we streamline the presentation of lead data to make the most relevant information more accessible and actionable?

Initially named “App Analytics”, these pages enumerated all analytics relating to the community, plan, elevation, floor plan, exterior, and interior selections. It was also divided into two parts:

Analytics Overview - condensed page with 15+ data visualization tables and charts on specific product analytics such as performance metrics, selections, brochure downloads, and more.
Reports: Allows users to narrow down analytics based on communities, sales centers, or plans.

Here my main focus was to evaluate whether the current layout and organization allowed them to efficiently access the data they need.

Through interviews with dashboard users and sales managers, I identified two prevalent mental models for approaching the analytics dashboard:

Top-down: Starting at the community or plan level and drilling down to specific exterior or interior selections.
Bottom-up: Beginning with exterior, interior, or floor plan options and analyzing their performance at a broader scale.

To accommodate these patterns, I reorganized the navigation to align with the typical homebuyer journey by dedicating a page for each specific step in the homebuyer journey from a community level down to the selection level. This enables users to more intuitively and efficiently navigate to the specific data they need.

In the original UI, the layout for this page was cluttered and overwhelming. Multiple iterations were done to make the info here digestible. The large thumbnail and limited space also made using this functionality difficult. After several iterations of redesigning the card layout, I realized that sticking with a card-based design might not be the best solution. Instead, I transitioned to a table format with specific filtering options.

Furthermore to aid users focus on that mattered most to them, I introduced a filter option to specify which metrics to include in the table. By default, it shows all metrics but can be modified through the drop-down to only show the most relevant or commonly used metrics, making the dashboard more tailored and efficient.
Overall Learnability

combined_screenshots_with_drawing (97).png

PRODUCT ANALYTICS

How might we present the data in a way that helps builder users ascertain key home metrics, enabling them to make informed business decisions easily?

CONCLUSION

NEXT STEPS

Test multi-channel capabilities and explore AI-driven methods in gathering quick insights

Evaluate KPIs such as task completion rates and time on task to evaluate effectiveness of the navigation
Continue to identify and revamp the data graphs and charts to account for multi-channel sources.
Gather real-world data on how users utilize the multi-channel data and how this affects their business decisions.
I want to integrate quick insights which explicitly summarizes data for people who don’t have time to quickly synthesize all the numerical numerical information. Exploring AI-driven solutions could help with this function

mockuuups-imac-mockup-on-a-table-with-green-and-pink-background.jpeg

mockuuups-tablet-mockup-on-a-wooden-table-with-iphone-11-at-the-side (1).jpeg

KEY TAKEAWAYS

Context is key!

Don't assume users know what word means. Analytics dashboards that use extensive terminology require substantial and purposeful use of helper elements such as tooltips and instructions to provide context and properly define data.
More data ≠ more insightful. Mirroring user mental models and how they view the homebuyer journey into the overall navigation significantly helps establish the overall flow.

Initially, the data visualization in the Insights Dashboard provided no context. Certain terms such as 'Views', 'Clicks', and 'Popularity' were not defined. Calculated metrics like 'Engagement Rate' lacked explanation of how they were derived. As a result, users were not able to derive any meaning insights from the data.

"Info hovers everywhere!" has become the common sentiment among dashboard users. Thus, for every data panel, information hovers are added to provide clear explanations of what each data panel represents and define any terms that require clarification. While well-designed charts and tables can often speak for themselves, additional instructions have been added to reduce user overwhelm, particularly for panels with addtional user actions, such as the Leads tables and Performance panels.

OVERALL LEARNABILITY

How might we improve the learnability of the Insights Dashboard to ensure users can easily understand the data, including what each chart, table, and metric represents?

CONSOLE DASHBOARD - INSIGHTS

ROLE

CHALLENGE

Despite having valuable customer and product data, users struggle with navigation and relevant insights. The 2.0 dashboard also focused on a single channel which won't align with the upcoming release of the 3.0 Anewgo multi-channel ecosystem

APPROACH & IMPACT

A multi-channel analytics dashboard that mirrors the homebuyer journey with a consistent and approachable UI that aids in learnability

DISCOVER

BACKGROUND

What was the Insights Dashboard 2.0?

The original Insights dashboard contained four specific areas:

Lead Analytics - lead database, buyer preferences and interactions

App Analytics - community, elevation, lots, and floor plan analytics

Marketing Automation - email campaigns and email templates

Buy Now - homesite reservation and down payment feature

Who are the main users of the Insights dashboard?

The dashboard is used by a myriad of teams both from our external customers and in-house employees.

Sales Teams - track and manage leads

Marketing Teams - campaign performance, customer engagement

Design Teams - floor plan, exterior/interior popularity

Product Teams - assess market demand

Customer Success Teams - aid and educate Anewgo customers

DEFINE

INFORMATION ARCHITECTURE

Overhauled both the high-level and sub-level navigation greatly reducing the need for long-scrolling pages

DESIGN

DATA ANALYTICS UI/UX AUDIT

Focused on ensuring consistency in the global analytics UI while refining the specific UX of individual data tables

When conducting the audit, I used a 4-point scale to determine the severity of an issue.

1 - minor UI problem (subjective, aesthetics)

2 - major UI problem (contrast, consistency, may affect UX)

3 - minor UX problem

4 - major UX problem (affects core user flow/goal)

DASHBOARD LAYOUT

How might we optimize the data panel layout layout to reduce visual clutter and ensure a more digestible interface?

DATE AND FILTERING

How might we help users quickly and easily filter the relevant data they need while still allowing for customization?

Originally called the Leads Analytics, it contained potential homebuyer statistics and details such as sign-in frequency, rating, name, email, etc. It was divided into two parts:

User Traffic - All analytics under the sun about leads were thrown into this bin and so little value is seen by some users as they don’t typically even visit this page or can’t really tell what they’re looking for.

CUSTOMER ANALYTICS

How might we streamline the presentation of lead data to make the most relevant information more accessible and actionable?

Initially named “App Analytics”, these pages enumerated all analytics relating to the community, plan, elevation, floor plan, exterior, and interior selections. It was also divided into two parts:

Analytics Overview - condensed page with 15+ data visualization tables and charts on specific product analytics such as performance metrics, selections, brochure downloads, and more.

Reports: Allows users to narrow down analytics based on communities, sales centers, or plans.

Here my main focus was to evaluate whether the current layout and organization allowed them to efficiently access the data they need.

PRODUCT ANALYTICS

How might we present the data in a way that helps builder users ascertain key home metrics, enabling them to make informed business decisions easily?

CONCLUSION

NEXT STEPS

Test multi-channel capabilities and explore AI-driven methods in gathering quick insights

KEY TAKEAWAYS

Context is key!

OVERALL LEARNABILITY

How might we improve the learnability of the Insights Dashboard to ensure users can easily understand the data, including what each chart, table, and metric represents?