Driven
UX
Engagement
Purpose Design
..at The Intersection of AI & Climate
Creativity + Technology + Mission
“Drawing upon my creative experience, love of tech and a mission to leave this world a better place for my son, I created Purpose Design.”
- Sarah Berchild
In 2023, I set out to work with companies that wanted to make an impact.
This includes those working with A.I., improving our climate through sustainability and regenerative practices, or reducing carbon emissions.
Client:PathFactory
AI-Infused Insights with Human Experiences
Industry
Technology
B2B SaaS
Generative AI
Services
User Research
User Experience
Interface Design
Prototype
Product Testing
Duration
24 months
This mid-market SaaS Account-Based Marketing company, based in Canada has incorporated an LLM into its service offerings. The redesign leverages generative AI for media development and predictive AI to create concierge-like digital experiences.
Multi-pronged Challenge:
Market Differentiation: What offerings will customers care about (and pay for) while also making life easier for the customers they already have?
Talent Gap: What does it take to launch a customer-focused product if your team has never had a UX role to shape a human-centered strategy that drives adoption and retention?
Goal:
I was hired to research, design and prototype a human-centric approach known as the "concierge experience" to an AI-infused product. Additionally, establish trusted UX frameworks and processes to show the value of customer-centric design approaches.
“You’ve got to start with the customer experience and work backwards to the technology. You can’t start with the technology and try to figure out where you’re going to sell it.”
- Steve Jobs
Solution:
1. Introduce a natural and concierge-like AI experience:
This new concept, coined the "concierge experience" was our north star in establishing an updated product that simplified work process with having "done for you" campaign creation as well as key account and lead analytics all assisted by AI.
We developed comprehensive prototypes and guidelines on etiquette, cultural norms, and conversational styles that trained their LLM to interact naturally with customers. This framework provided a warm, welcoming first impression all the way through a return visit, ensuring customers feel understood and valued.
2. AI-assisted micro-site creation:
In 2023, the client aimed to broaden their market and integrate LLM to build microsite experiences, connecting with other agent actions. This new agent experience empowered marketing teams to develop campaigns based on insights on customer habits, engagement, and intent signals. It features "done for you" campaign creation and provides key account and lead analytics, all supported by AI.
3. Robust Analytics and Reporting:
Analytic dashboards and infamous for providing data but little to no insight on what actions to take. Our approach was to take it one step further, to allow our analytics to "connect the dots" for our customers, providing them insights to make impactful action.
This mid-market SaaS Account-Based Marketing company, based in Canada has incorporated an LLM into its service offerings. The redesign leverages generative AI for media development and predictive AI to create concierge-like digital experiences.
“The key is to set realistic customer expectations, and then not to just meet them, but to exceed them.”
-Richard Branson
Through competitive analysis and interviews with both internal customer service teams and customers, we gained insights into what users find most valuable for completing their everyday tasks and achieving their goals.
Methodology : Research
Defining Success : KPIs
With a solid data foundation and clear understanding of customer needs, we set out to make exceptional, concierge-like service our guiding star. However, we needed to ground our success in measurable outcomes.
Demonstrate added value to the user
GOALS
Provide valuable, actionable content
Gain trust, gather insights
Loyalty. Distinguishing frequent users from occasional ones
Personalized interactions means more revenue.
MEASURES
Content Consumption/User Engagement: Depth and breadth of content engagement. (pre & post personalization)
Experience Engagement: % of time customers engage with, create, and share experiences
Conversion/Attractiveness: % Rate at which users voluntarily share personal information on 1st visit.
Return Visitor Rate: Frequency & Recency of Site Visits
Pipeline Attribution: Revenue impact measurable to personalized agent interactions
Considerations
We acknowledge project success doesn't depend solely on customers.
Many projects get derailed by internal processes, team dynamics, and factors beyond our control.
Such as:
-
Project Ambiguity
-
Platform Complexity
-
Data Sources / Constraints
-
Customer Variation
-
Team Bandwidth
Mapping the Journey
Pinpointing moments of delight
As stewards of our brand and builders of clarity, we designed AI-native experiences that make complex systems feel simple, human, and intuitive. We began by focusing on the end-to-end customer journey, identifying the moments with the greatest potential for delight, then mapped out that experience.
Creating Personality
"Hi Sandy, welcome back! Glad to see you again.
Would you like to start where we left off or can I show you around?”
Learning from real human connection
With our AI Guiding principles, I set forth defining a personality and set of guidelines to establish a human-like concierge experience based on some of the best books and resources about etiquette, case studies of companies with best customer service, art of conversation, manners and ethics. Taking the best of those learnings, created a guideline and prototype for the LLM to have a learned behavior applied to customers
Just like a real butler or concierge, we aim to make customers feel cared for throughout their entire journey.
“...Is there anything else I can help with today?”
Based on our KPIs, our aim was to provide valuable, actionable content, and a genuine human approach via progressive personalized interactions because that is what builds stronger relationships and ultimately more stable revenue.
Guiding Principles for AI
-
Progressive Trust Building via Value-First Personalization
-
Surface Multi-Level Contextual Memory
-
Transparent, Consensual, User Control and Ethically Minded
-
Agent as Trusted Partner, Not Just a Tool
Ideation
With a solid data foundation and clear understanding of customer needs, the design team and I set out to make exceptional, concierge-like service our guiding star. However, we needed to ground our success in measurable outcomes.
AI Adaptive Conversation
Driven by our AI Guidelines, I gathered insights from sources that outline cultural norms, etiquette, manners, examples from renowned companies celebrated for their outstanding customer service. Additionally, I explored definition of the five stages of conversation. We then integrated Marketing Archetypes and Sales Funnel stages to create a cohesive conversation, offering guidelines the LLM on what to say, when to say it.
All of this research informed how to respond based on four criteria: Sales Funnel, Archetype, 5 Conversational Stages, and Use for exemplary customer service. This comprehensive information formed a structured framework.
Attribution: Sarah Berchild
Actionable Campaign Analytics
We have data, but it’s only valuable if it offers insights. We've examined ways to present data and provided marketers with tips for utilizing this information effectively.
Attribution: Maria Motta, Ayobami, Balogun
Our objective was to enable marketers to effortlessly create campaigns featuring relevant articles and media tailored to their customers' insights, allowing them to quickly develop a microsite experience without any coding knowledge.
"Done for You" Campaign Creation
Attribution: Ayobami, Balogun
Prototypes
Progressive AI Conversation and "Done for You" Campaign
Conversational ease of use is just as important as interface flow. We set out to showcase a display that helps customers accomplish their work by surfacing progressive disclosure—a design principle that reveals information gradually, showing only what's needed at each step to avoid overwhelming users while maintaining access to deeper functionality.
Mobile View
Conversational ease of use is just as important as interface flow. We set out to showcase a display that helps customers accomplish their work by surfacing progressive disclosure—a design principle that reveals information gradually, showing only what's needed at each step to avoid overwhelming users while maintaining access to deeper functionality.
Protyping with Artificial Intelligence
Along the way, it became clear that the complexity of the conversations were too robust for the static, yet clickable prototypes. So I began building light-weight prototypes that began to incorporate logic.
In this case, to build out the "guest" vs. "known" flows - and begun testing areas of the journey were we can begin to introduce value-added offers. Building in Vercel (V0) also gave great flexibility as the prototype was flexible, responsive in size and could introduce logic in ways that far exceeded the previous Figma prototypes.
Applied Design
With established style guides, the team began applying visual design to the screens to ensure they adhered to the primary client brand. However, it became clear that this could be resold and therefore alternative styles and visual designs were also created with the capability to be white-labeled to serve their customers.
Flexible Whitelabel Branding
