Universal Help Module

A machine-learning-powered help application to provide self-service resources, guided assistance, and human support in one unified experience. Over time, it evolved into a localized, modular platform embedded across all customer touch points.

ROLE:  LEAD DESIGNER
AREAS:  STRATEGY, RESEARCH, DESIGN

TIME LINE: 2019 - 2021

Background

Autodesk is a leader in 3D design, engineering, and entertainment software. In 2016, the company transitioned to a SaaS model, shifting customer purchases from resellers to direct online sales. As a result, customers began expecting faster, more direct support.

Customer Problem

We were hearing from customers that finding help was very frustrating.

  • No consistent way to get help

  • No direct paths to help

  • Scattered options

  • Inefficient agent to agent transfer

“Provide a support number or a live chat feature”

"Extremely confusing website - getting help was impossible"

"Why do I have to jump through hoops and spend hours trying to find someone to help me”

Internal Issues

  • Varying help strategies across products and geos led to confusing pathways, exposing customers to internal complexities

  • Complicated routing rules resulted in Support and Sales teams wasting time solving similar issues and re-routing cases

Goal

  • Get the customer to the right answer at first touch

  • Provide in-context and customer-specific help options where the customer needs it

  • Reduce post-sales human contact by 13% while maintaining customer satisfaction

Teamwork

This project was the first of its kind where many different organizations worked together to align and implement this product over the span of 2 years.

I collaborated with 50+ Business Stakeholders, Sales Specialists, Customer Service Agents, Product Managers, Engineers, QA, Data Scientists, Taxonomists, Data Analysts, Program Managers, Scrum Masters, Content Strategists, Researchers, and Visual Designers.

Design Principles

I developed key principles grounded in research insights from the learn and help space. These principles served as a foundation for making informed decisions and navigating trade-offs during roadmap planning and feature prioritization.

MVP

The very first application was deployed on 5 product pages in the US website, delivering guided support for issues with buying, installing and troubleshooting Autodesk products.

The experience was divided into 3 parts:

  1. Issue classification: Users navigated pre-determined decision trees; in some branches they formulated their own questions and we experimented with machine learning to classify and route issues

  2. Self-help solutions: curated links to articles, short answers, dynamic links if question was captured, and Watson-based workflows

  3. Human help: variety of options for customers to choose from to connect with an agent, such as phone, chat, case, etc

Issue Classification

Users narrowed down their issue through 2-4 levels of topic selection. I worked with data science and content strategy teams to design several versions of this tree based on case data. Then I partnered with our lead researcher to test labeling and presentation with customers to ensure that topic selection was meaningful and efficient.

🔆 Customers prefer typing in their question so we introduced machine learning enhanced search in a few branches where users could type in their question to get more specific answers.

Self-Help Solutions

For branches with topic selection, 2-6 curated knowledge articles or a short answer were presented in the help tool. I worked extensively with content strategy to identify which solutions would be shown and provided guidance on styling and layout of the short answers.

In branches where users entered a question, the recommendations were dynamic and depended on how well the users described their question. I worked with the search platform team to understand their technology enough to offer guidance on structuring better queries and helped fine tune how the results are displayed. We also piloted a self-service workflow where a virtual agent guided them to get a download.

🔆 Clear navigation ensured that users know where they are and get to where they want to go.

There’s always a way out

I included ways for customers to exit the decision tree or virtual agent workflow so the user can change course, if needed, and doesn’t feel trapped. At any point in the flow, the user can jump back up to the home screen, track back to previous steps or use an escape hatch to talk to a human agent.

🔆 Customers always had an escape hatch to exit the flow to get to a human.

Human Help

While self-service is what users primarily seek, there are always urgent situations or complex issues where they prefer connecting with a human.

By asking the user to sign in we are able to look up their account and serve up personalized contact options based on their subscription type or product. We recognize if users are already signed and do not ask again, and make sure that we utilize their account information for their benefit only.

In the places where they typed in a question, we use machine learning to classify their question and route them to the right support team.

Variety of Human Contact Options

Human help options are all offered in a compact way all in one place. We offer you ways to chat with a human agent, schedule a call at a time that’s convenient to you, give you a phone number depending on your subscription type, or allow you to submit your question via email. And pretty soon, you will also see a call back option so you don’t have to wait in line over the phone and for those purchasing questions, we will also be providing chat with a sales expert.

Under the Hood

Top Pain Points

  • Evaluated a year’s worth of case data from various touch points to establish high level issue types and topics.

  • Led workshops with stakeholders to align on top pain points to prioritize for e-commerce and post-purchase support.

Complex Business Rules

  • Helped stakeholders document a matrix of business rules based on customer’s subscription type and user role. Engineers then coded these rules into the help application so that customers got the contact options they are entitled to.

Agent Ecosystem

  • Sales and support agents are skilled based on topics across the customer lifecycle, the product a user has and the language they speak. I facilitated workshops with agents to try and help simplify the skill categories to reduce overlap and make it easier to guide customers to the right agent.

User Research

I formulated research goals and worked closely with the Research team in evaluating various aspects of the experience.

We used Treejack and First click studies in testing the decision tree of topics. A key insight was that the initial selection is extremely important so having fewer topics upfront and displaying a short description helps them choose the right path.

Results & Reflection

  • While engagement with the ““?” icon is consistent month over month, it is overall lower than expected < 1% dotcom visitors)

  • 75% of customers who engage with the app click on at least one article or contact an agent

  • Visits to dotcom that had interaction with the help module resulted in 5% higher cart additions and almost 2% higher number of orders placed

“I really like how this is laid out… live chat, schedule a call, create a case….because sometimes you have to go through menu after menu after menu and it just drives you crazy to do it. This is actually very nice.”

— Customer with a Standard Subscription

Expansion of Platform: Broader - Deeper - Smarter

After the first 6 months, the modular, extensible Universal Help platform was expanded to cover more use cases in more languages across multiple web properties.

  • Localization: Universal Help was deployed on all global dotcom sites (>1000 pages) and adapted to 30+ countries with geo-specific content. The UI was translated to the local language but in some cases the articles themselves were in one of 10 languages. Visitors saw article titles in their language but often times the articles themselves were in English and the agents they were connected to did not speak their language. So I designed appropriate ways to help transition users from local language to English, in addition to help identify country-specific content needs.

  • Account Management Portal: The help app was embedded in the customer’s Account recognized the authenticated customer right away and guided to help with their downloads, installation and user management activities. It was also embedded within Fusion product to offer customers a quick way to chat with an agent.

  • Contact Support: Full page website version of the application helped customers to find the quickest path to post-purchase and contact Autodesk Support

  • In-Product Support: Embedded inside Fusion that decreased resolution time by 20%; open rate in ACAD was 1% but in Fusion it was 41% in canvas

  • Pop-up nudge: Drove awareness around benefits of using the new module - increased engagement by 15%

Account Management Portal

Contact Support website

Fusion in-product