Omni Experience / Service Design

Troubleshooting

Role

Experience Designer (Network)

Start

June 01, 2020

Launch

Ongoing

About

Tap it once, fix it all nice.

The Team

A junior designer and I worked on this project. I worked on the Network Troubleshooting part, and the junior designer worked on the Device Troubleshooting part. I collaborated closely with different key stakeholders for a consistent experience.

The Time

The project timeline was aggressive. I planned 80% of my time for user research, workshops, experience audit, service blueprint creation, system paradigm, concept exploration, and usability testing. I spent only two days out of less than four weeks on high fidelity mockup creation.

The Deliverables

Aside from high fidelity mockups, I delivered a user research report, several system paradigms, five different service blueprints, three workshop briefs, one concept testing result, and three usability testing results.

01.-

Introduction

Life is too short to...troubleshoot

On a regular Saturday afternoon, I opened my Google Meet for a scheduled virtual coffee session with my best friends, Macy, to hear about her travel in Taiwan last year. Five minutes past the scheduled time, she texted:"Got caught up on something, 10 more min, sorry".

She was upset and agitated when she joined the meeting: "I can't wait to 'kiss goodbye' to Verizon, I cannot believe this is how my Saturday started. I had this simple st*pid call dropping issue, I troubleshot on Verizon website, I went through this lengthy, tedious and unhelpful process, then I had to call Verizon, and they asked me the same questions I answered on the website. They even asked what kind of issue I had, can you believe that? What's worse, they tried to troubleshoot with iPhone instructions, when I'm using Samsung! I spent two hours on this and I'm so done with Verizon!" Then she saw my embarrassed face and remembered : "Wait a minute. You’re working at Verizon!"

I did not talk much about her travel that day. She is just one of millions of Verizon customers who experience issues. Our digital troubleshooting experience is unnecessarily lengthy and repetitive. It makes already frustrated users feel defeated and it diverts issues to call centers, which is a burden for our representatives and a high operational cost for Verizon.

This was a product that benefitted no one. I needed to change this.

02.-

Challenge

What is it, and where does it go?

A sad fact: Millions of customers come to us for support across channels every year. Very few issues are resolved without talking to a representative.

I wanted to design a predictive (Verizon will know the customers and their devices) and proactive (Verizon will know their issues) omnichannel (Seamless, no interruption even if they switched to a different channel) experience. It should empower customers like Macy to be able to solve their issues quickly and on their own. It also enables Verizon to document the user's journey in a more intelligent and organized approach.

Macy promised she would stay a Verizon customer while giving me this opportunity to make her happy.

03.-

Understand

Understand, then resolve it.

To further understand the problem, I always start with users, then phone representatives, and then complete a round table experience review. As always, user first!

I partnered with an internal research team to conduct eight user interviews in our research lab, where I can record and observe participants through cameras. They were asked to try out our existing troubleshooting experience with different scenarios, and scrutinize the language, their reactions, and emotions.

I also did six phone interviews with our representatives in call centers to learn about their frustrations.

Then the whole design team did an experience audit on our existing troubleshooting flow to understand it from a designer's point of view.

I contacted the Verizon analytics team for analytical data to reference, which set up the baseline for future metrics. I also had to understand the constraints and limitations of the technology team.

I used Miro Board to synthesize post-it notes from research and identified seventeen main themes via Affinity Diagramming. They were color-coded for easy recognition and I evaluated and prioritized all themes based on frequency, efficiency, and other factors.

I mapped out end-to-end service blueprints based on existing experience to further analyze user's pain points and our opportunities during different stages. Spreadsheets and Miro were our best friends!

04.-

Quantifiable Success Metrics

Define success, in a quantifiable way.

As designers, although I always tend to describe our success as some subjective descriptive wording like "easy", "happy" or "satisfied", I also cannot ignore the tremendous value objective data can bring to us and the company. After several rounds of hot discussions with Analytics, Intelligence, and Marketing teams, I pushed our limitations and fearlessly defined our success metrics as follows:

26%

Troubleshooting Call-in Deflection

Reduce over 26% of the Call In Rate about Troubleshooting related topic.

4+

Net Promoter Score Increase

Increase NPS by 4 points in Verizon Service and Support space

33%

Call In Time Reduction

From leaders directly managed me, designer peers collaborated with me, key stakeholders partnered with me.

12%

Customer Retention Rate Increase

While I am working in Verizon, with hundreds of small projects to fix here and there across Verizon Fios and Wireless.

Let's start chasing goals!

05.-

All inclusive activities

Persona + Pain Point + Use Case + Channel => How Might We

I always believe design is an all-inclusive activity that involves not only designers ourselves but also stakeholders across different teams that share a piece of it. Due to the pandemic situation, I hosted a remote three-day workshop to illustrate personas, generate How Might Wes, design concepts, and present them to the whole task force. There were 20 delegates from different teams involved, divided into four groups, on five channels (Website, App, Retail Store, Interactive Voice Response, and Chatbot).

Modular Persona is a persona creation technique I use in Verizon that focuses on key identities, behaviors and attributes that present a wide range of Verizon customers. It was designed more so to be a card game that everyone, designer or not, can jump in and play.

All hands on deck, I meticulously created three personas, Johnny, a tech-savvy millennial Android user, Margaret, a Gen-X Apple user, and Jennifer, a Gen-Z who has a busy life with little patience, to cover the majority of our customer base.

I further generated different How Might Wes based on pain points from research themes (customers, representatives, designer experience audit) and personas I created, with typical scenarios assigned.

The format of How Might We is "How Might We #verb# #Persona#'s #Pain Point# when #Use Case#". For example, one of the main themes I generated was "repetition", and corresponding HMWs from the research were "How might I ensure that #users# aren't asked/told to repeat the steps they previously attempted on Verizon website?" and "How might we recognize the steps #customers# have taken on their own and pick up where they left off on My Verizon App?"

06.-

Concepts & Service Blueprints

From How Might We to concepts, sailing with ideal Service Blueprints & System Paradigms

I created our new Service Blueprints on User Journey Maps, both a Reactive approach and a Proactive approach, as well as System Paradigms, to illustrate our ideal user experience.

I conceptualized over forty different ideas and trimmed them down to fifteen for further validation through testings based on How Might We. I categorized them into three groups, Proactiveness & Transparency, Future Retail Store Experiences & Support, Third-Party Resources & Continuity, asking participants to rate each concept on a scale of one (dislike) to five (like).

At the same time, I asked the whole workshop task force to evaluate each concept based on Operational Difficulty and Technical Difficulty to calculate the mean value, also on a scale of one (easy) to five (hard).

I arranged each concept on a Cost/Impact Chart, and fit them into our ideal experience.

07.-

Concepts

Some interesting concepts

I want to touch on two concepts that were very bold and insightful, even they were not included in our Minimal Viable Experience: One of them was partnering with Google to allow users to fill in Verizon store visit reservations so our retail store representatives can get prepared. Since our research showed customers usually went for the third-party (like Google Search) when they had issues with Verizon device or service, our customer experience actually starts way before they reach out to us, and thus how to design this pre-Verizon experience became critical than ever before.

The other is the Continuity Code that enables users to resume where they left off on any channel without additional authentication! Perfect for channel hoppers!

08.-

Initial Design

Initial Design

Due to the aggressive timeline, I prioritized our efforts on digital channels for Minimal Viable Experience: Website and App. I based our design on the current live / feed card design that Verizon commonly employs on both website and app to keep the familiarity that users have been experiencing. I introduced "Proactive Messaging and Predictive Tile".

Proactive Messaging is a serial of messages that were sent by Verizon via a user's preferred way of contact, based on a user's profile, when issues were detected by Verizon. Verizon is capable of detecting some issues and take preventative actions. For example, if I detected that a user has dropped a few calls, and the user has not turned on Wi-Fi Calling, this user will get a proactive text message to recommend they should turn on Wi-Fi Calling. For Andriod users, all the transactions can even be nicely done via a text message or email.

Predictive Tile is a dynamic tile that presents predictive troubleshooting information and offers automatic or manual actions after diagnostics were triggered by certain criteria (to avoid system overload) I defined in our system mapping, like search history and areal alert information.

09.-

Usability Testing

To be happy, or not to be happy.

I conducted two rounds of usability testing and received great feedback from seven participants with minor complaints.

For Proactive Messaging, like expected, all seven participants were quite happy that Verizon was being proactive and showed care about customers. They referred to this as a self-contained, direct, efficient, and automated troubleshooting experience within a single text chain. Its copy provided the problem and a solution. One participant, however, did express some privacy concerns as Verizon knows too much about her device and settings. This opened a valuable educational moment at which I need to emphasize the explicit confirmation from customers before Verizon makes any changes. Three participants questioned the validity of the SMS as they cannot easily tell if this message was really from Verizon, other than some sort of SMS spoofing.

For Predictive Tile, it was welcome by all participants as well. However, five of them did miss the tile at first glance, which means I needed to make some changes to address this.

10.-

Final design, deconstructed isometric

Final Design, explained.

With what I found, I made a few changes to Proactive Experience. I talked with the Technology Team to understand how I can convince our users that our messages are not spams or scams, which is a valid security concern. Fortunately, I do have solutions on both iOS and Android. iOS Business Chat allows us to show verified Verizon Business Chat Account directly over Messages, and Android has interactive messages by turning on the chat feature, which has similar functionality. In this way, users can totally trust our messaging!

I also made a few improvements backed by our usability testing on Reactive Experience. The purpose of these improvements was to attract more attention to users and signify actions users can take to resolve the issue. All the feed card designs were managed by another team so it took us some time to convince them. I had to make some compromises but I did it.

Check prototype

11.-

Impact

Impact, on emotions.

Our design was quite a success in the managed trial period, the projected data speaks for itself:

31%

Troubleshooting Call-in Deflection

5% higher than expected.

4+

Net Promoter Score Increase

Same as expected.

35%

Call In Time Reduction

2% more reduction than expected.

15%

Customer Retention Rate Increase

3% higher than expected.

Our design also took good care of customer emotions when in usually stressful troubleshooting scenarios, easing down the anxiety and stress I observed during the research period. When I were watching the participants' reactions and emotions when they were trying out our new design, they did not seem confused or lost.

From the designer's perspective, I did a fantastic job as not only did I resolve the users' issues, but also I offer users a peace of mind and error-proof one-directional path.

It will definitely positively impact my opinion of Verizon if they're proactive about issues I might be having.

Verizon Customer

Research Participant

It didn't take me to another app or to the website. I didn't have to navigate anywhere else. It was just really easy.

Verizon Customer

Research Participant

It will definitely positively impact my opinion of Verizon if they're proactive about issues I might be having.

Verizon Customer

Research Participant

Really beyond my expectation!

Verizon Customer

Research Participant

It didn't take me to another app or to the website. I didn't have to navigate anywhere else. It was just really easy.

Verizon Customer

Research Participant

12.-

Reflections

I did not stop here.

If I had more time, I would have done more solid research on how phone representatives interact with customers and work on workforce facing product to help phone representative handle difficult troubleshooting issues. As of now, I did not have time to do this, but eventually, it will come to us as an alienable part of the omnichannel experience.

I learned three key things from this project:

Plan ahead. As I had such an aggressive timeline, planning ahead allowed us to use time wisely and efficiently because I literally had no time to waste. At the very beginning, I did not start design activities: I spend two days to just plan each design phase and involve key teams. For example, I contacted the Research Team to confirm and book in advance research and usability testing reservations, I also reached out to the Analytics Team to get the data I may need ready at a later stage. I did not wait until I need it, then realized I did not have it. This saved us quite some time at the later stage. I even contacted other Design Teams for some touchpoints I may cross path for our design to have them walk through their designed experience.

Proactive outreach. Never hesitate to ask for other team's help: if you do not tell them, they cannot give you what you want.

Work harder. Deadline is the deadline, for huge efforts that involve many different teams, I cannot let the Design Team be the bottleneck, especially for a project that was highly exposed to leadership.

13.-

Macy

Macy is happy.

"So, you mean I will just receive a message, then reply 'Yes' to turn on Wi-Fi Calling, and problem solved", Macy waved her phone like a kid with excitement on her face, "It is that simple?!"

"Yes, it is that simple, and even our grandmas can resolve issues on their own now!", I was chowing hard with food in my mouth, keeping nodding my head.

"Wow!" Macy was surprised and took out her credit card, "I gotta buy you dinner this time, it sounds fantastic and I cannot wait! This time on me!"

"Waiter, please, I want to add another dish!" I raised my hand, did not even hesitate.

yangdaosen#gmail