Forethought is the most advanced generative AI agent for customer support, trained on your unique data set. The company’s AI-driven solutions integrate seamlessly to resolve issues instantly, predict and prioritize tickets, and assist agents with relevant knowledge. Pulse 2.0 interviewed Forethought co-founder and president Deon Nicholas to learn more about the company.
Deon Nicholas’ Background
What is Deon Nicholas’ background? Nicholas said:
“I’ve been passionate about AI and its possibilities for over 15 years. My journey started at the University of Waterloo and the Alberta Machine Intelligence Institute, a leading AI lab in Canada. I was very interested in AI’s capability to answer questions and solve problems. Having worked in customer service, I saw firsthand how AI could transform the industry, which led me to co-found Forethought with Sami Ghoche in 2018. Since then, Forethought has raised over $90M. Customers like Upwork, Cotopaxi and Airtable use our generative AI platform to automate customer support and serve as a copilot for human agents.”
“Our mission is to unlock human potential through AI, starting with transforming customer service. We fundamentally believe that customer experience will be the most consequential use case for generative AI.”
Core Products
What are the company’s core products and features? Nicholas explained:
“Forethought is the only gen AI for customer support that can autonomously solve customer inquiries. Our AI trains on our customers’ data so we can deliver consistent and precise answers. While most AI for CX platforms rely on decision trees, our agentic platform does more than simply regurgitate answers from FAQs and help desk articles–it can take action, like hitting APIs or resetting passwords, to resolve customer issues. With patent-pending Autoflows, our agentic reasoning engine, CX professionals can use natural language prompts to train AI on specific goals or policies–no coding or technical ability required. This saves agents time and energy, allowing them to focus on priority interactions while improving customer satisfaction.”
“For more complex support tickets, our Triage product can help identify those and route them to a human based on historical data and intent. Over time, Forethought collects data on customer support tickets to help companies improve their products and processes, create more helpful customer resources, and improve satisfaction overall.”
Challenges Faced
What challenges have Nicholas and the team faced in building the company? Nicholas acknowledged:
“One of the biggest challenges in AI is hallucinations – when models give confident but totally wrong answers. To tackle this, we developed an internal evaluator model that checks every interaction for confidence, relevance, and user engagement. It ensures our AI only shares information when it’s sure of the answer.”
“Another challenge has been standing out in a crowded market. Many vendors make similar claims about AI in CX, but not all AI is created equal. So, we began centering our go-to-market strategy around running proof-of-value free trials. We’re confident in our tech, and we want our customers to see for themselves how well it works compared to the rest of the competitors out there. We’ve achieved a 75 to 80% conversion rate on these free trials, meaning when customers try the product, they stick around.”
Evolution Of Forethought’s Technology
How has the company’s technology evolved since launching? Nicholas noted:
“We’ve been laser-focused on making our AI more advanced and creating an exceptional experience for our customers’ customers at scale. From the start, our product was agent-focused, meaning an agent always reviewed answers before they went out. We eventually launched our end-user-facing agent (Solve) during the pandemic, which saw rapid adoption.”
“We were some of the first to start leveraging GPT-1 and GPT-2 in 2019 and among the first to launch a Retrieval Augment Generation (RAG) solution in 2023 after the advent of ChatGPT. A year ago, we launched Autoflows, a fully agentic solution allowing CX teams to specify their desired issue resolution outcomes in natural language instead of relying on old-school decision trees. Autoflows trains on a company’s support tickets to provide more accurate, contextual responses over time. And it’s working: companies who use AI trained on their own data achieved the highest deflection rates (meaning more tickets were resolved without a live agent needing to intervene) and average customer satisfaction scores of 90%.”
“Additionally, we’ve built integrations with leading helpdesks like Zendesk and Salesforce, enabling enterprises to achieve a complete AI solution for CX.”
Significant Milestones
What have been some of the company’s most significant milestones? Nicholas cited:
“Publicly launching at TechCrunch Disrupt in 2018 was a major early milestone for Forethought. Leading up to the competition, we knew it was critical to showcase how we were delivering value to customers, so we dedicated ourselves entirely to bringing on at least five customers whom we could tout to the judges. My co-founder Sami and I also spent excessive hours in a room preparing for any questions we would receive on stage. We raised $9M in our Series A round shortly after!”
“In 2021, we raised our $65 million Series C round, which allowed us to accelerate our growth and invest in R&D. In 2023, we launched Autoflows, eliminating the need for CX teams to build automation workflows manually. In 2024, we saw exponential growth, and our net-new ARR per quarter grew 400% year-over-year.”
Customer Success Stories
When asking Nicholas about customer success stories, he highlighted:
“At Fetch, which is the no. 1 rewards app in the US, Forethought’s AI helped achieve a 3.9x ROI while fully resolving over 250,000 inquiries in just the first 90 days. YAZIO, a leading nutrition app, leveraged Forethought Solve to deflect 80% of tickets within six months, saving the equivalent of three full-time agents while maintaining an impressive 4.0 CSAT score. For Forma, Forethought’s AI helped their team achieve a 99% accurate triage rate, ensure that customer inquiries were routed correctly, significantly reduce response times, and enhance their ability to scale globally.”
Funding/Revenue
When asking Nicholas about the company’s funding and revenue details, he revealed:
“Forethought has raised over $90 million, and in 2024, our net-new ARR per quarter grew by 400%.”
Differentiation From The Competition
What differentiates the company from its competition? Nicholas affirmed:
“We’re the only AI for CX platform using truly agentic AI. Forethought agents can solve issues end-to-end. Unlike traditional bots that simply provide information, our AI takes meaningful action – whether resetting passwords or troubleshooting complex issues – autonomously and seamlessly. Our patent-pending Autoflows technology eliminates the need for complex decision trees, achieving a resolution rate of 70-80%, surpassing the 10-20% typical of RAG-based solutions. We focus on delivering a human-like, fluid customer experience supported by our internal evaluator model that ensures every interaction is confident, relevant, and positively received.”
Future Company Goals
What are some of the company’s future company goals? Nicholas concluded:
“We’re continuing to build on the most advanced enterprise-gen AI platform for customer support. In the near future, we plan to introduce multi-modal capabilities, like voice and video, to provide a more human-like support experience. Long term, we envision a world where AI is more seamlessly built into every experience, handling actions without prompting. Today, we interact with AI by prompting it, but in the future, AI could be more proactive and evolve outside of the chat interface.”
