Integrating ChatGPT Into Carenet AI

The release of ChatGPT by OpenAI in November 2022 has created quite a stir. Why?

Imagine a person who has read every single book ever written, every article, every video ... and can recite all of it in seconds.

That’s ChatGPT.

If you’re a techie, ChatGPT (Generative Pre-trained Transformer) is a form of “generative AI” which is built on top of OpenAI's GPT-3 family of large language models and is fine-tuned with both supervised and reinforcement learning techniques. The models are applied to the entire index of internet content. For the non-techie, that means in addition to answering questions, the AI can create long-form content – whether text, images, computer code, books, business material.

It's currently free to use, so try it here.

ChatGPT gained a million users just a week after launching in November 2022, a milestone that took Facebook 10 months, and some project it to reach a billion users by the end of 2023. While there are healthy ethical debates about having AI generating recommendations for patients or healthcare research, there’s no debate that ChatGPT can optimize MANY mundane tasks across a multitude of industries, including:

• Physicians summarizing medical reports
• Automating repetitive writing
• Writing website content
• Developing and documenting computer code
• Providing answers to questions
• Writing product brochures or marketing information

ChatGPT in Carenet AI

Carenet AI delivers actionable intelligence for commercial teams targeting the healthcare market, processing overwhelming amounts of data about healthcare facilities, physicians, and their practices. The platform utilizes a custom ICP (Ideal Customer Profile) model to build key signals that are most important to our users.

By using this model to analyze large datasets of skilled nursing facilities, physicians, care providers and other healthcare entities, Carenet AI can accurately 'stack rank' these entities according to their match to the ICP. A Carenet AI customer gains access to all the data points attached to these entities, and the top signals that indicate the best match to the ICP are highlighted for easy reference. This allows the companies to easily use the signals as talking points in phone calls or even in cold emails, making the data more actionable and streamlining in targeting efforts.

While Carenet AI's ICP model and highlighted data points make it easier to target the ideal customer audience, creating nuanced communication that takes into account the unique score and signal profile of each skilled nursing facility or physician can still be time-consuming. That's where ChatGPT comes in.

With Carenet AI's alpha integration of ChatGPT, creating customized email content for every target customer is now easier than ever. Instead of relying on templated emails, Carenet AI customers can take advantage of ChatGPT's ability to generate human-like text responses based on a given prompt, providing suggested custom email outreach content for each target customer.

Carenet AI passes the following information to ChatGPT as part of the messaging prompt to generate customized email content:

• The type of content desired (e.g. a short email sales pitch)
• The name, title, and organization of the person being contacted (e.g. Frank Smith, Director of Nursing at Sleepy Meadows Nursing Home)
• Key signals that match the unique ICP model for that person or organization (e.g. recent F880 health deficiency violations related to infection prevention, lower than average nursing staff)
• Key selling points of the product or offering

With this information, ChatGPT returns a well-crafted email message tailored to the specific situation. Here’s what the signals and the recommended email message can look like:

In the above example the healthcare facility had strong signals of recent issues with UTIs and with insufficient nursing staff. The email copy is generated purely by ChatGPT, as it weaved in these signals and then highlighted the product benefits and how that ties to the signals mentioned above.

Sometimes a health facility could be an ideal target because of certain shortcomings or problems they have faced in the past, and they thus need better solutions to manage their issues. Targeting only these facilities can significantly limit the addressable market. But taking the opposite approach and looking for facilities that have a ‘perfect track record’ in managing certain issues can also provide good targets for products. But contacting them naturally calls for different messaging. The below example shows how the ChatGPT generates a message for a healthcare facility that never had a violation regarding infection prevention.

The benefits of the product offered are the same, but in this case the email mentions how it can be used to maintain the high standards in the facility. And this can be done easily at scale across hundreds and thousands of facilities and data points.

Carenet AI's alpha integration with ChatGPT is now available to all customers. Experience it with a free trial.