Join top executives in San Francisco on July 11-12, to hear how leaders are integrating and optimizing AI investments for success. Learn More

The process of writing marketing and sales materials has long been a human-driven endeavor, but it doesn’t have to be that way.

Startup Regie, which was founded in 2020, has been building out an AI-powered content generator for sales and marketing. The Regie technology makes use of the GPT-3 natural language model from OpenAI, alongside a business user-friendly interface to help automatically build engaging content and process workflows for sales and marketing campaigns. The first private release of the Regie technology was made available in December 2021, and now, after months of beta user testing, the service is ready for broader consumption with general availability announced on June 7.

“GPT-3 came out in June of 2020, and my thesis was that there had to be something that we can put it to use for in the enterprise,” Srinath Sridhar, CEO and cofounder, told VentureBeat.

The idea for using GPT-3 to help enable the sales process came from a pain point that Sridhar’s cofounder Matt Millen had experienced. Millen had previously led revenue at where he experienced first-hand the need for performant content at scale for modern revenue teams. The promise of Regie is to help solve that challenge in the fastest way possible with an AI-powered system that can automatically generate content with a few clicks.


Transform 2023

Join us in San Francisco on July 11-12, where top executives will share how they have integrated and optimized AI investments for success and avoided common pitfalls.


Register Now

How Regie AI empowers resource-constrained organizations

The challenge that Millen had at Outreach wasn’t a unique one. Sander Bien, CEO of sales and marketing vendor Thought Horizon, had a similar issue.

Bien told VentureBeat that his firm is a growing startup, and for the past few years it has looked for outsourced services to help it with sales prospecting. The outsourcing vendors that Bien contracted with didn’t quite work out as well as he had hoped.

“The idea at Regie was that AI could create these same high-value emails and touches at scale for a fraction of the price,” Bien said. “Largely, this is what we have seen working with Regie and we are quite pleased.”

Bien explained that Regie is fully integrated with his company’s sales processes. On the software side, Regie creates the emails and campaigns and makes them available to Bien and his staff to read, edit and approve before they are uploaded into ZoomInfo Engage, which is what his company uses as a sales enablement stack.

“Quite simply, Regie has made it possible for our small company to compete with our much larger competition in digital and omnichannel outreach,” Bien said.

Blake Vernon, chief revenue officer at B2B lead generation firm Sapper Consulting, also relies on Regie to build content for the firm’s sales messages. Vernon said that having salespeople write their own content seemed like a great idea at first, but realized over time why an approach like Regie is a better way to go.

“Our SDRs [sales development representatives] were writing based on subjective opinions on what they thought sounded good based on how they would buy,” Vernon told VentureBeat. “The issue was that SDRs aren’t our buyers, so we often missed the mark.”

Regie uses AI to build the sales messages and are generated based on content and buyer intent.

“By taking the subjectivity out of writing, our emails are now based on our buyer personas, and written in a fraction of the time,” Vernon said.

How Regie uses AI to automatically generate effective sales content

Sridhar explained that in December 2021, OpenAI began to allow users to create custom models for GPT-3. It’s a capability that Regie has taken advantage of, building a custom model with GPT-3 to meet the requirements of sales and marketing.

“What we’re doing behind the scenes is starting out with GPT-3, we’re building a new custom model that can generate pain points, using a lot of label data for business,” he said.

Going a step further, Regie has also trained its data model to better incorporate diversity and inclusion. Sridhar commented that typically GPT-3 and other large models are trained on a large volume of public data that might not always have the right sensitivity for diversity and inclusion data. To solve that issue, Regie has built its own database of inclusivity language, which is now part of the company’s service as well.

“So we make sure that no matter what GPT-3 puts out, our language is really good,” he said.

VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.