> For the complete documentation index, see [llms.txt](https://docs.raiaai.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.raiaai.com/use-cases/case-study-us-eye.md).

# Case Study: US Eye

**US Eye slashes caller wait-time with an AI “Agentic Workforce” powered by raia**

***

**Snapshot**

| **Metric**                                          | **Before raia**                                       | **After raia (AI Call-Assist Agent)**                                 |
| --------------------------------------------------- | ----------------------------------------------------- | --------------------------------------------------------------------- |
| Average time to find an answer during a call        | **4 min 30 sec** on-hold while staff searched manuals | **< 55 sec** (-80 %) — answer surfaced instantly in the agent console |
| First-call resolution rate                          | 62 %                                                  | **89 %**                                                              |
| Caller abandonment                                  | 11 %                                                  | **4 %**                                                               |
| Locations, doctor rosters & insurance plans indexed | Manual lists for 60+ clinics                          | **Dynamic vector store** refreshed nightly                            |
| Equivalent human workload offset                    | —                                                     | Capacity equal to **≈ 5 FTEs**                                        |

> “With raia, every support rep has a virtual expert at their fingertips. Calls are shorter, answers are accurate, and patients feel the difference.”\
> \&#xNAN;**– VP Patient Services, US Eye**

<figure><img src="/files/HkYBBTibRDEXTbJWFhnh" alt="" width="375"><figcaption></figcaption></figure>

***

#### 1 | Company & Challenge

**US Eye** is a physician-led network of ophthalmology and optometry practices with **60 + locations across Florida, the Carolinas, and Virginia**. Rapid growth left the support centre juggling:

* Varying clinic hours, specialties, and doctor schedules
* Differing pre-op and post-op procedures by location
* Dozens of accepted insurance plans that change each quarter

Agents spent minutes rifling through PDFs or transferring callers—driving long hold times and caller frustration.

<figure><img src="/files/VbpjKLXloEXzl5dt0LYR" alt=""><figcaption></figcaption></figure>

***

#### 2 | Why US Eye chose raia

* **Bulk document ingestion** — 8 000 + policy PDFs, doctor bios, insurance matrices, and EMR FAQs were drag-and-dropped into Launch Pad; auto-vectorised for semantic search.
* **Real-time console integration** — the AI sits in a side panel of US Eye’s Five9 softphone, surfacing answers as agents type or speak keywords.
* **Nightly sync pipeline** — any update in the corporate SharePoint refreshes the knowledge base before the next business day.
* **CoPilot feedback loop** — during a 30-day shadow phase, supervisors thumbed up/down AI suggestions, pushing precision above 96 % before going live.

***

#### 3 | Solution Design

| **AI Agent**                 | **Primary Role for Support Reps**                                                                                      | **Key Integrations**           |
| ---------------------------- | ---------------------------------------------------------------------------------------------------------------------- | ------------------------------ |
| Call-Assist Agent            | Listen (text stream) to caller’s question, search vector KB, display top answer & citation instantly; log interaction. | Five9 CTI, SharePoint doc feed |
| Insurance-Lookup Micro-agent | Autocomplete payer name, verify plan acceptance by clinic, display copay notes.                                        | EMR API, nightly payer feed    |
| Procedure-Prep Micro-agent   | Retrieve location-specific pre-op & post-op instructions, text/email to patient on demand.                             | Twilio SMS, SendGrid email     |

***

#### 4 | Implementation Timeline

1. **Weeks 1–2** – Collect & convert policies, schedules, insurance lists to clean Markdown; bulk upload to raia Launch Pad.
2. **Weeks 3–4** – Shadow mode in CoPilot across two call-pods; gather feedback, tune prompts.
3. **Week 5** – Go-live for all agents; nightly SharePoint sync turned on.
4. **Month 3** – SMS / email integration added for instant patient hand-offs.

***

#### 5 | Early Outcomes (first 90 days)

* **Call answer-search time cut 80 %** (4 m 30 s → < 55 s).
* **First-call resolution up 27 percentage points** (62 % → 89 %).
* **Caller abandonment nearly halved** (11 % → 4 %).
* AI capacity equals **≈ 5 full-time agents**, allowing US Eye to absorb seasonal peaks without overtime.

***

#### 6 | Next Steps

* Expand the knowledge base to include **LASIK financing FAQs** and **clinical trial enrollment**.
* Deploy Spanish language pack for bilingual service lines.
* Add analytics dashboard to surface most-asked questions and drive policy updates.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.raiaai.com/use-cases/case-study-us-eye.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.