Overview

Project

This project’s aim was to create an engaging experience for customers, allowing them to utilize a self-help option before they feel the need to call or chat with an agent. The goal is to empower users to find solutions independently while enhancing their overall satisfaction.

Role

Collaborate closely with the product owner and developers to enhance and streamline the process, making it significantly easier for customers to find the help they need on their own, rather than relying on calling or chatting with customer service agents.

 
 

Revenue

The Save to Invest initiative has the potential to save DIRECTV a substantial amount of $1 million during the very first year of this exciting launch.

Target Demographic

All DIRECTV customers.

 
 

Contact center guiding principles

  • As much as possible, replace instances of our phone number with a link to Contact Center

  • Driving traffic to Contact Center vs. showing phone number allows us an opportunity to better own the experience and reduce calls

  • On Contact Center, capturing customer intent first enables us to provide the best solutions and channels

 
 

Current DIRECTV “Contact Us” experience

Contact information

  • Presented at bottom of page

  • Not easily discoverable

  • No intent captured

 
 

COMPETITORS

I took a deep dive into analyzing competitors both within the industry and outside of it, exploring their strategies and offerings.

  1. Hulu

  2. Netflix

  3. Southwest

 

MVP VErsion

Learns from MVP

Contact Us gets about 6k sessions per day

  • 3k calls (50% of sessions)

  • 1.4k bot starts (23% of sessions) — 1.3k chats* (20% of sessions) *Assume 90% esclation rate

Separate Contact Us page doesn’t drive net new calls

Driving additional traffic on-site doesn’t appear to add net new calls

  • We added a “Contact Us” link to the Footer( (Services/Support) and our traffic doubled, but Center saw no noticeable increase in total calls

Experience improvements:

  • Presented on a separate page

  • SEO optimized

Outcome

  • New page ranks #1 for most “contact” related searches

  • Call volumes did not increase

How do we measure success?

<25% in call from Contact Us

  • Support lander currently drives- 14k calls per week, we want to see minimal increase in calls, but suggest setting the threshold at 17,500 (+25%)

Contact Us in organic search results

  • New page appears in google results in place, or in addition to, current contact options

  • This has been achieved; Contact Us ranks #1 for all related searches and has taken over the Featured Snippet position

 

Version 2

Experience improvements:

Results

  • 9% of session clicked on a self-help link

  • Agent Chats were reduced by 29% (v2 vs MVP) at current volumes, this represents an annualized savings of $288k

  1. Customer selects topic, self- help information is offered

  2. Customer can find link from every page

  3. Selected intent passed if customer chooses to chat

  4. Hide phone number from certain Google search results

Outcome

  • Traffic to page doubled while no noticable impact to total calls

  • Goal: Drive additional traffic

  • Goal: Reduce calls and chats

How do we measure success?

  • Decrease calls and chat by 15%

USER TESTING - Version 2 vs. Competitors

I collaborated with a UX Researcher to test Version 2 and gather feedback on Netflix and Hulu’s help centers.

8 moderated 1:1 interviews

Key Takeaways:

  1. People generally expect less support, so don't count on quick help when contacting the company.

  2. People want to quickly solve service problems by themselves. They enjoy having various options in a Help Center and several ways to reach out for help.

  3. Participants got frustrated when they had to repeat actions and didn't want to search for contact information.

  4. Participants from Google search and Support Lander had different frustrations because the latter didn't have to repeat actions.

Opportunities:

  1. Create Contact Us consistently throughout the site.

  2. Add a search field to the Contact Us page to offer people a secondary way to self-serve.

  3. Use a predictive search to lessen the cognitive load for people searching the support site.

  4. Consider presenting topics in a way that enables quick forward/backward movement and visibility with minimal interaction.

 

What did the data says from live A/B test?

Version 2 vs Version 3

The call and chat volume dropped. Call rate from 85.42 to 83.13 and chats from 23.32 to 10.65 which would be an annualized savings of over $600k.

 

version 3

MVP + v2

  • 50k sessions per week

  • 13k calls per week

  • 4k chats per week

Reduced Agent Chats by 29% for an annualized savings of $288k

What are the next steps?

  1. Guide customers to more specific self-help information based on their topic and sub topic

  2. Offer actions when possible

  3. AB test V3 vs MVP

  4. Continue to driive traffic to the page

How do we measure success?

  • Decrease call by 15%

  • Decrease chats by more than 29% (v2 results)

 

Current version up live Aug ‘24

 

Learnings

This project had quite a few ups and downs since this was the very first agile pod team that was assembled. Our goal was to produce work quickly and efficiently, getting it out the door in sprints while continuously building upon the valuable data learnings we gathered throughout the entire process. Utilizing components from the Figma Design System not only streamlined our design but also made it significantly easier for our developers to code these elements at a much faster pace as well.

Overall, this was one of the most enjoyable and engaging projects I have had the pleasure of working on at DIRECTV, as it truly went through the entire design process from start to finish.