Flow Control Designer (Cisco Webex)
Overview
Webex Flow Control Designer -- is a visual scripting tool that will enable the configuration and management of any scripted/automated experiences impacting the Webex platform. The main goal of this project is to conduct usability testing on the developer environment of the -Webex Flow Designer and gather user’s detailed feedback on global settings, main functions, icons, and overall usage. This is one main project I have worked on while interned at Cisco this past summer.
TEAM
Researcher(me), UX designers, Product Managers, Engineers
Duration:
May 2020 - Aug 2020 (12 weeks)
Research Questions
Creating a flow:
-
Are users able to build and understand a simple, commonly used flow?
-
If errors are thrown during the validation, are they able to figure out which part is broken?
-
What difficulties do they encounter? Are they able to fix these issues on their own?
-
What's their understanding of each activity in the library?
Interface and icon
-
Do these icons make sense to you? Is the function of these tools clear by its icons?
-
Are there any changes you would like to make?
Customized variable
-
Do you know what can be customized and what cannot be customized?
Settings:
-
What’s their understanding of the global settings?
-
What are the default settings?
Process
Many departments in Cisco's Contact Center Team are involved in the scripting tool building process and each plays an important role:
​
Customer Service Manager
​
- CSMs work 1:1 with partners, sometimes with admins to troubleshoot
-
Probably know the Flow Control tool and show partners how to use it
-
Probably use the Flow Control tool when provisioning new customers for initial validation
Professional Service Team
- A dedicated team that gets paid to code custom flows for customers
- Probably use the Webex Contact Center portal
-
The process is: 1. customer submits word doc request form, 2. Professional Services codes script based on request and bills for hours, 3. PS team may work directly with the customer to get access to their portal
Pre-Sales Architects
-
Coordinate with Sales, Business Development, and Account Manager to pursue sales cycles.
-
Coordinate with cross-functional teams to understand customers’ needs during the pre-sale process and demonstrate how Cisco tools can support their business goals.
​
​
Thus, it is vital to meet early and align everyone's expectations. We paid close attention to collaboration at each step — working closely with the core stakeholders, and engaging the extended stakeholders at the right time. We wanted to take all stakeholders on the journey with us so that there would be no big reveal and surprises at the end of it.
Project Timeline
Research & Insights
What I did
Usability Testing ( 7 sessions)
Synthesis
For the good part of this project, the data analysis was conducted in various forms.
-
Coding data and creating themes
-
End of day debriefs after testing
-
Brain-writing and flow charts
-
Video watching and insights analysis
Snapshot of Data Log
What I Learnt
Overall, this tool satisfies customers' needs overall. Most participants, especially Technical Solution Architects understand how to use this tool to build call flow. As shown in the graph below, the overall ease of use rating is 4.05. However, the success rate is not high on some specific tasks. For example, the success rate for the activity " Add Variable"on the global panel and "Set variable" local activity is 0. "Add HTTP" is hard for most participants because this activity is relatively new for them.
Graph of Ease of Use rating and success rate
Priorization Strategy
Since the scope of the project is big and limited resources, the prioritization of research findings is vital. To utilize all stakeholders' knowledge to prioritize the proposed initiatives, I faciliated a long meeting with more than 6 stakeholders involved in the flow control designer building process to discuss my pririzitation of each block and differnet workload for research findings.
​
The research findings that have high impacts and high cost are our top priority.
​
Next Steps
When the time came to hand the project back to engineer team, I wanted to make sure they feel comfortable with the strategy, the recommendations, and all they need to implement. I presented the final deliverables to different teams and combine the readout deck with other researchers.
​
Reflections
Research Impact is the Key. Presenting Readout is not the end
During my one-on-one meetings with my manager, one of the most important lessons that I learned is how to make the research influenial. After presenting, it is not the time to celebrate and relax. Instead, creating a spreadsheet and keeping track of the implementation of different research findings is the key. Thus, after wrapping up the 1st project, I documented all the important notes, meeting recording to the group Confluence, and created a spreadsheet to follow the project. Even after presenting, I still schedule meetings to discuss the findings when engineers have questions. I firmly believe that, it is not how many research finds you have synthesized, it is all about how much you push the project forward and create a better experience for users and your colleagues.
​
When you and your teammate think and communicate, try new methods, and find what works for both of you
Internship onboardig is hard. It is ever harder for a remote internship. My main team is a group of 5 people in different timezones in the United States. Starting a project without face to face communication is challenging. I am a visual person. When I try to make sense of things, any form of structure (e.g., a table, a tree, a diagram, a story) brings me comfort in the chaotic sense-making process. Thus, my remote internship onboarding became into a miro board session.I created lots of graphs, whiteboard, roadmap, brain-writing with many sticky notes to help me understand the product line, and communicate with my team members. This is also a good way to double check I understand my teammates completely. In this case, the communication between me and my team members became more efficient.
​
​
​
Be aware of the potential bias while analyzing data
While analyzing data, I initially prioritized the findings that were mentioned several times by our participants and deprioritize those with low frequencies. In classical statistics, this makes sense because the average is a way to measure data trends. However, during the weekly meeting with my colleagues, they brought up the Bayesian theory. In this theory, a single case can have a high impact. I learned to be aware of the data bias and rethink the big picture
Stories are a natural and powerful way to connect people. Telling a good story connects users, designers and researchers.
Jeff Bezos said that in his experience, when anecdotes and data diverge, he believes the anecdotes. In this living breathing world, we are not all perfectly rational beings. It’s stories that bring tears to our eyes, not numbers. It is the story that bridges me as a researcher to users.
Collecting stories - Creative toolkits, even the very simple ones, can help research participants open up and tell their stories. During the background research, ask participants to think back to a specific past experience and draw you a timeline and some storyboards.
​
Telling stories - Researchers love to bring up anecdotes to support the findings. Not because anecdotes prove anything, but because the audience can empathize with a real person not a statistic. Some of them will gasp, smile, and say “Oh, yes, I get it. That reminds me of …” I used to be very self-conscious when I brought up the same story again and again. I worried that my audience would notice the repetition and get bored. But there’s magic in repetition — it’s how stories stick with people and become contagious.
Note: I cannot disclose more about this product given the NDA policy. Please email me at dingyu.hu@sjsu.edu to discuss more details.
More works
