Service Design
18 Weeks // 2018

Cadence: Empowering isolated older adults to find and stay engaged in their community

My Role

Design: Voice Interface Design (Lead),  Brochure Visual Design (Lead), Video Production (Lead), Service Design (co-Lead), Interaction Design (co-Lead), 

Research: Literature Review (Lead), Report Visual Design (co-Lead), Usability Testing, Participant Interviews, Expert Interviews, Competitive Analysis, Ethnographic Research

Industry Mentors

Artefact Group

The Team

Duminda Aluthgamage
Kristina Rakestraw
Andrew Shiau
Alexandria Lee


Design Specification
Product Video
Public Presentation
Research Report

Social isolation and loneliness affect a significant and increasing number of older adults in the United States today.

Cadence is a service that helps seniors find groups and volunteer opportunities in their community, with the goal of creating meaningful and long-lasting social connections.

// Service Overview

Through interviewing seniors, senior community center leaders, and senior advocacy groups, we identified key insights that drove the design of Cadence:

Cadence Insights v.2

Cadence acts like a social maven who provides personalized suggestions based on questions about hobbies, tastes, experiences, location preferences, and skills. These choices build up an internal algorithm which helps Cadence continue to “learn” the types of activities that the user prefers.

Cadence uses simple conversational interactions accessed through a phone-based VUI or web browser GUI. The dual interaction ability stems from the desire to cater to varying levels of technology acceptance. Both interfaces are aimed specifically toward seniors through design choices based on extensive design research and usability testing.

Given the complexity of the problem space and the target demographic, we used a service design approach to create Cadence. Our approach considers the interlinking of new and existing components within the sociotechnical environment of older adults. Cadence is a holistic solution that handles discovery, registration, exploration, management, communication, transportation, and feedback. Each element of the service would be reviewed and optimized on an ongoing basis.

Outreach & Discovery

The Cadence Group Guide brochures are mailed to those above a certain age range within the district. They would also be available in various locations such as community centers, cafes and health providers. With a regularly updated list of groups, and instructions on how to register, browse, sign up, and host groups, the brochure provides a welcoming introduction into the Cadence experience.

Conversational & Personal

Through interaction design and dialogue/copy choices, the GUI and VUI provide a conversational experience which allows for learning, coordination, and collaboration from both a user and system perspective. Cadence gets to know the individual user through the registration process and by asking questions periodically after initial sign-up, in order to fine-tune suggestions.

Motivation & Feedback

Both the copy and the backend services that Cadence offers are designed to maximize motivation for seniors to start and stay engaged in groups. Cadence also sends reminders about upcoming events or invites users to new suggested groups through a phone call or email. Users can easily provide feedback and ask questions, through the GUI or by being directed to an agent via the VUI.

Accessibility and Comfort

Cadence helps with transport by providing bus routes, arranging carpools with other group members, or organizing ridesharing apps. She periodically asks users casual questions, such as their hometown or favorite places to travel, provided to the hosts to better facilitate groups. Hosts also have access to a digital playbook detailing group protocols such as for setting up a group, accessibility and safety.

// Background

Social Isolation: The Silent Killer

Social isolation and loneliness are considered grand challenges in American social work. Social isolation as an objective situation in which someone lacks personal relationships to fall back on in case of need. This contrasts with the subjective idea of loneliness; when a person’s actual number of relationships is less than expected, or the quality is less than desired. People can be both lonely and socially isolated, and this can be situational (circumstances cause isolation) or structural (ongoing isolation for many years).

Various US national surveys have indicated that rates of social isolation and loneliness in the “older adult” population (65+) are up to 40%. This is particularly troubling as effects are amplified within this age group, including increased stress and depression, high blood pressure, and ultimately, increased mortality rates.

The social ecosystem of older adults is complex, with several interrelated factors and opportunity areas. With this in mind, we set out to design a solution that would empower older adults to take the first step out of social isolation or avoid it all together.

// Research

“How might we alleviate social isolation and loneliness in older adults?”

Research Questions & Goals

  1. What are the barriers to social engagement that lead to social isolation?
  2. What do people who are not socially-isolated do to stay engaged?
  3. Identify elder technology use patterns.
  4. Make connections to our target population for future usability research

Research Methods

  1. Literature Review
  2. Competitive Analysis
  3. Expert interviews
  4. Target demographic activities (semi-structured interviews, guided storytelling)
  5. ‘Into the Wild’ (observational/ ethnographic studies)

Research Process

To better understand problem areas of aging and to help identify where a design intervention might be beneficial, we combined primary and secondary research methods over 9 weeks.

We conducted interviews with 8 subject matter experts across industry and academia, including gerontology, multigenerational social work, HCI research, behavioral health and product design.

We also interviewed older adults (60+) and senior communities around the Greater Seattle Area to understand how seniors currently stay engaged in the community and major barriers to social engagement. We tried to talk with people from all walks of life and socioeconomic statuses in order to understand the problem from multiple viewpoints.


Each member of our team independently coded all of our primary research, totaling more than 20 participant interviews. We then came together to synthesize this information into themes and insights. This was a challenging task conducted over several days, during which we became familiar with all the interviews, including those which we didn’t personally attend. The insights are listed below.


1. Despite preparation, many people are blindsided by sudden changes in life circumstances during retirement.

2. Without shared interests, being in close proximity to others doesn’t necessarily alleviate loneliness.

3. Socioeconomic factors, such as access to money or transportation, are barriers preventing many older adults from engaging in activities with others.

4. Societal attitudes towards aging can discourage older adults from engaging with others.

5. Despite there being existing outreach programs, it can still be difficult to locate isolated individuals in the community and encourage them to take the first step.

6. Usability issues and general attitudes towards technology limit older adults from adopting new technologies into their daily lives.

7. Technology is often a real and perceived barrier to meaningful social interactions for older adults.

“They were telling Alexis or whatever to play certain music… and I don’t know if I want to go there… I’m perfectly happy to go over there and turn it on.”
“I would get to know other people [at mealtimes] but I got tired… I sat with too many people who aren’t cognitively able”

// Design Principles

1. Make technology visible only to those who want to see it

Given the variance in technology acceptance among our target users, the solution should not force technology onto those who are not accustomed to it. Our focus is not to change current behaviors around technology, but to leverage older adults who are willing and able to embrace new technology in order to reach those who cannot.

2. Be replicable with adjustments in different contexts

We will start by focusing on Puyallup and Seattle, since they are the first two AARP Age-Friendly Communities in Washington state, however we assume that more cities will follow this trend. Our proposed solution should be scalable and benefit other cities.

3. Employ accessibility standards for seniors

Comply with accessibility guidelines to ensure that our primary users are able to fully utilize the technology. Secondary research identifies that basic accessibility issues are a major barrier to older adults adopting regular technology use. We found that this sentiment to be echoed in our interviews with our target population who cited that the following prevented them from using technology: small font sizes; difficulty with keyboards due to limited finger dexterity; screens too small to understand iconography; difficulty with ongoing software updates.

4. Protect users from age-specific harms

Acknowledge that target users are vulnerable to age-specific harms related to technology and protect them through design decisions that deliberately mitigate against them. Particular sensitivities of the older adult population include:

  • Loss of dignity or being patronized by caretakers and support networks
  • Higher susceptibility to being scammed (eg. tax scams, phishing scams, impulse buying, email scams, etc.)
  • Limited understanding of privacy, data retention and online security.

// Ideation & Prototype Testing


Using our research insights, we went through two rounds of ideation, starting with 30 concepts each (120 in total). These explored a huge variety of ideas, such as IoT devices catered to seniors, tools to provide outreach for community centers, and mobile apps to connect seniors to volunteer opportunities.

We narrowed and refined our ideas down to six, creating storyboards to further illustrate and understand the interactions. These ideas were discussed with our experts and participants and formed the basis of the Cadence service.


Prototypes were created and tested for both the GUI (clickable prototype) and VUI (Wizard of Oz, Investigative Rehearsals). During the first round, interactions within the GUI were designed based on paradigms similar to existing products such as Meetup. For the VUI, we experimented with dialogues that were meant to replicate the GUI experience. We found many elements of these designs to be confusing for participants, leading us to reconsider our approach.

Based on our first round of testing, new prototypes were created to afford a more conversational experience. For the GUI, users were guided through a clickable wizard allowing them to answer questions and receive suggestions. The VUI dialogue was improved to feel more like a conversation with a real person, whilst resolving issues around too many search results, barge-in and improving prompts.

// Service Design

We presented the final Cadence service through a comprehensive design specification which includes:

  • Service design documentation (service blueprint, customer journey map, ecosystem map)
  • System architecture
  • VUI specification and sample dialogues
  • GUI Hero Flows and interaction logic
  • Visual artefacts (Brochure, Playbook)
  • Visual style guide
Download Spec

Ecosystem Map

I created the ecosystem map represent how Cadence fits into the social life of older adults. Nodes within this map were based on secondary research into gerontology, expert interviews and discussions with participants

Customer Journey

I created the customer journey diagram to better understand the possible customer experience throughout the use of Cadence. This would be an iterative diagram, refined on an ongoing basis using stakeholder feedback, in order to improve the end-to-end service.

Voice Interface

I led the design of the VUI, including creating all dialogues, flow charts and specifying the system.  In its ideal implementation, users just need to call Cadence and converse in their natural language, being guided through registration, discover and signup in a conversational manner. I specified goals of the conversation in line with Gordon Pask’s process of conversation, considered VUI design challenges, Prompts, Inputs, Technology, Algorithms, Errors, as well as Ethics & Privacy.

Through carefully designed interactions and dialogue, the VUI aims to reduce the load on users’ working memory, semantic memory and procedural memory. Whilst essentially an IVR system, it attempts to act as a conversational service. It includes technology components intended to improve its ability to understand users, deal with errors or unexpected responses and ultimately provide a satisfying user experience.

Interaction Flow – Registration

Sample Dialogue – Registration


I co-led the overall interaction design for Cadence, as the VUI was designed to closely mirror the GUI. I assisted with branding and design choices for the GUI, which was crafted by Kristina.

// Takeaways


Through research and design, Cadence challenged many assumptions that we had about older adults. Meeting and working with seniors from different walks of life, and researching the causes and effects of social isolation and loneliness on this population was an eye-opening experience that will stay with us.

Working in a team possessing different skills and values proved both rewarding and challenging. I was able to build skills across Service Design and Voice Interface design and continue to develop my visual design, research and prototyping competencies.

Ultimately, whilst we don’t see Cadence as the one-step solution to solving loneliness and isolation in seniors, we hope that it will inspire more HCI design work in the gerontechnology space.

Where to next?

Cadence was an ambitious project and in order to take it to the next phase, further work needs to be done across all areas of the process. As we took an iterative, service design approach, ongoing testing of interactions (VUI, GUI) as well as feedback from users and stakeholders would be required.

The voice system itself, whilst comprehensive, would require further investigation of technology requirements, and some technological advancements to be truly useful for our target demographic.

We would also need to work with experts to further develop and refine the dialogues to be appropriate for our target demographic.