Customer support · Production

Trust & Will achieves 68% automated resolution with Will-E AI agent built on Ada

The problem

Trust & Will's scripted chatbot could not handle the nuanced, emotionally sensitive estate planning questions their customers brought, producing looping experiences and lacking the empathy required, while a lean team faced growing demand.

First attempt

The previous scripted chatbot caused looping experiences, lacked empathy for estate planning conversations, and could not resolve even simple issues like applying promotional codes, forcing escalation to human agents.

Workflow diagram · grounded in source
1
Customer submits support inquiry
Trigger
Customers contact Trust & Will's support team about nuanced estate planning situations.
▾ source quote
“the nuanced situations that our members contact our support team about every day”
2
Will-E generates grounded AI response
Ai action
Will-E uses generative AI and pulls information only from company content to provide safe, accurate, and relevant responses.
▾ source quote
“Will-E leverages generative AI and the best artificial intelligence models to provide scalable, automated resolutions that combine efficiency with the quality of human-like interactions. To ensure that generated responses are always safe, accurate, and relevant, the AI agent is designed …”
3
Route to resolution or escalation
Routing
Routine inquiries are resolved autonomously by Will-E while high-stakes, emotionally charged cases are routed to the human team.
▾ source quote
“By covering routine inquiries, Will-E empowered the team to focus on high-stakes, emotionally charged cases”
4
Human team handles escalated cases
Human review
The support team delivers white-glove service on cases Will-E cannot resolve.
▾ source quote
“We've gone from firefighting to delivering white-glove service. Will-E doesn't just handle inquiries, it gives us the bandwidth to be there for our customers when they need us most”
5
Coaching and continuous improvement
Feedback loop
The Trust & Will support team uses Ada's coaching tools to continuously improve Will-E's responses and close knowledge gaps.
▾ source quote
“Using ada's advanced coaching tools to continuously improve is performance. The Trust & Will support team treated Will-E like a new team member, dedicating significant time to its training and continuous improvement.”
Reported outcome

Will-E automatically resolves 68% of conversations with a 75% CSAT on AI interactions, and within less than a year has maintained a steady 70% automated resolution rate, enabling the human support team to shift from firefighting to white-glove service on high-stakes cases.

Reported metrics
Steady automated resolution rate70%
Reported stack
Will-Eada
◆ Does this fit your context?

Compare to your context

Tell us your scale, team, and constraints. We'll show what changes at your size, what fails at your scale, and whether this case is a fit, needs adaptation, or won't scale to you. Free demo, no signup.

Compare to your context →
~30 seconds · free
Source
https://www.ada.cx/case-study/trust-and-will
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Will-E automatically resolves 68% of conversations with a 75% CSAT on AI interactions, and within less than a year has maintained a steady 70% automated resolution rate, enabling the human support team to shift from f…

What tools did this team use?

Will-E, ada.

What results were reported?

Steady automated resolution rate: 70% (source-reported, not independently verified).

What failed first in this deployment?

The previous scripted chatbot caused looping experiences, lacked empathy for estate planning conversations, and could not resolve even simple issues like applying promotional codes, forcing escalation to human agents.

How is this customer support AI workflow structured?

Customer submits support inquiry → Will-E generates grounded AI response → Route to resolution or escalation → Human team handles escalated cases → Coaching and continuous improvement.

Related customer support cases

WHAT TO DO WITH THIS

Now compare it to your context

This case is one data point. Whether its pattern fits you depends on your volumes, your stack, and your exception load — that comparison is the step no case study can do for you.

Browse the Customer support cases →