All Field Guides

From Confusion to Clarity: How Empathy Project Nailed Search Accuracy with Guru

Empathy used Guru’s Knowledge Agents to separate U.S. and Canadian content, giving their care team faster, more accurate answers.

Reading time
4
 minutes

Company Background

Empathy supports families navigating loss by providing practical and emotional guidance through a combination of human care and technology. Their platform helps simplify complex processes like funeral planning, probate, and benefits while also offering emotional support. By partnering with major companies like Prudential and MetLife, Empathy helps families save an average of 189 hours and $3,007 during one of the most challenging times in their lives.

Empathy’s services are designed to remove administrative burdens so families can focus on healing. Their platform provides a personalized Care Plan, live support, and a resource library that addresses both emotional and logistical needs.

The Challenge

When Empathy expanded its services to Canada, they faced a major operational challenge: how to accurately separate US-specific and Canada-specific content.

Estate administration laws and processes vary significantly between the US and Canada, so the team needed a way to direct questions to the right set of information without risking confusion or misinformation.They wanted to avoid having Care Managers search in Guru’s general chatbot, which pulled from both US and Canadian sources and could have potentially led to incorrect answers and longer resolution times.

“We were seeing cases where someone would ask about estate fees in Manitoba and the bot would pull up the Ontario answer. That would have been confusing for our care managers.” - Oliva Estes-Downes, Sr. Research Associate

The Method

Empathy leveraged Guru’s Knowledge Agents to create a streamlined, country-specific support structure that made it easier for their care team to access the right information quickly and confidently.

1️⃣ Built separate Knowledge Agents

Empathy created two distinct Knowledge Agents—one for the US and one for Canada—ensuring that each bot pulled answers only from the correct regional knowledge base.

“Having separate bots for each market helped our support staff feel more confident that they were getting accurate answers specific to the right country.”


__wf_reserved_inherit

2️⃣ Pinned Agents to Custom Pages

They linked each Knowledge Agent to a dedicated US or Canada resource page in Guru, so care managers could quickly navigate to the correct bot based on the user’s location.

“Linking agents to specific pages has been a big thing for us. It makes it so easy for our care team to find the right answers.”

__wf_reserved_inherit

3️⃣ Used Country Flags for Clarity

To further reduce confusion, they added a visual cue—a Canadian 🇨🇦 or US flag 🇺🇸—next to each content card. This signaled to care managers that they were searching in the correct place.

4️⃣ Set Up AI Prompts to Avoid Generalization

Estate administration laws vary by province in Canada and by state in the US. Olivia and her team set up AI prompts to ensure the agents wouldn’t generalize across regions. If a user asked about estate fees in Ontario, the Canada bot would pull the Ontario-specific answer, not a general Canadian one.

“We set up prompting so that the agent would be extra sensitive to province-specific and state-specific information. That’s worked brilliantly so far.”

5️⃣ Integrated with Guru’s Chrome Extension

Empathy embedded the Knowledge Agents directly into the Guru Chrome extension, allowing care managers to quickly access the right information without switching tabs.

Olivia also made adjustments based on early challenges. After realizing that AI sometimes pulled the wrong answer due to multiple provinces listed on a single card, she worked with Guru’s team to restructure the content. She used headers instead of tables and adjusted the layout to make it easier for the AI to recognize province-specific information.

“The way it was structured initially made it hard for the AI to tell which answer was for which province. We switched to headers and that’s really helped.”

The Outcomes

Empathy’s use of Guru’s Knowledge Agents delivered major improvements in both efficiency and confidence for the care team:

  • 🔍 Improved Searchability - Care managers can now find answers quickly and accurately without sifting through irrelevant content. Olivia noted that they “rarely now get a question where no direct answers yield.”

  • ⬇️ Reduction in Irrelevant Information - The agents’ regional separation means the team is no longer served incorrect or conflicting information.

“We’ve significantly reduced the amount of irrelevant information that shows up in searches. That’s huge for us.”

  • ✅ Increased Trust in Guru - By delivering consistent, accurate answers, Guru’s Knowledge Agents have helped the care team trust the platform more deeply. Olivia shared that this improvement in trust means the care team feels more confident while providing support to users in real time.

“This has been great for improving our support staff’s trust with Guru, and that’s essential because we need them to feel totally confident in the answers they’re getting.”

Scalable for Future Expansion

By setting up a clear process for separating knowledge by region, Empathy has built a framework that can scale as they expand into new markets.  “This was our first iteration, but it’s worked so well that we’ll keep building on it as we expand.” 

Key Stats

Customer Testimonials

Key Takeaways

Guru Capabilities Leveraged

No items found.
Published on 
March 28, 2025

Further reading