Investing in Decagon

The pain of a bad customer service experience is universal. No one likes being placed on hold for hours, being transferred to a string of different agents, or emailing back-and-forth for days only to come to an unsatisfactory resolution. In fact, 85% of customers are so averse to bad customer service experiences that they are willing to go out of their way to do business with a company that has better service — a move that costs businesses an estimated $3.7 trillion every year.

The best companies know that an exceptional customer experience needs to be core to their business, and they understand that customer service is not a cost center, but a meaningful long-term revenue generator and retention driver. However, building a support function to continuously provide that experience is a challenge. Support in many industries can be highly seasonal, which makes it difficult to scale operations with demand. Employee attrition in the customer service industry is high, and training a new batch of employees can take months. And there’s a constant tension between providing first-class, on-demand support while also remaining cost-efficient.

We believe the paradigm for software has shifted. Before LLMs, the best way for companies to service their customers was by implementing workflow systems for human agents, where human agents still had to ultimately physically triage and resolve tickets. We talk a lot internally about how LLMs are able to tackle problems that were previously not addressable by software, and building always-on AI support agents that truly delight customers is a perfect example. That’s why we were so excited to invest in Decagon and to have led their seed round last year.

Decagon is an AI agent that automates customer support for a business. Their conversational AI platform fully integrates into a customer’s core business systems and operates as a full, 24/7 agent that can converse with a user over text, understand the context of the inquiry, and be able to fully automate resolution and actions for the customer. 

The customer experience feels seamless to the user, but the product is highly complex behind the scenes — leveraging a mix of third-party and fine-tuned models to reason through complex business logic, analyze themes, tag conversations, take automated actions, and more. Their AI-powered dashboard can also identify trends and consequently suggest changes to a customer’s knowledge base. This allows customer experience leaders using Decagon to become uniquely proactive in their support philosophy. 

The customer love is glowing as a result, and Decagon is one of the early examples of an LLM-native application in production at scale. In less than a year, they’ve been able to serve companies like Eventbrite, Rippling, Vanta, and ClassPass, who emphasize that the company has seamlessly resolved a significant portion of their existing support volume, has dramatically changed how the company operates, and in many cases, has caused their customer CSAT satisfaction scores to actually go up. This, in turn, allows their customers to plan for growth without having to increase their support headcount. 

Our investment in Decagon is particularly rewarding for us since Jesse Zhang and Ashwin Sreenivas, Decagon’s co-founders, met at an a16z retreat we hosted last year. Both are repeat founders — Jesse previously founded Lowkey, which a16z had invested in and which he successfully sold to Niantic in 2021, while Ashwin founded Helia, which was acquired by Scale in 2020. Both have high technical acumen (having won math and science competitions growing up), extreme customer empathy, and singular ambitions to build a generational company. Both also have exceptional work ethics and are in the office 6 days a week (which they do without drinking coffee!). 

We are thrilled to have been the first check into Decagon at inception and to have led the seed, and are excited to deepen our commitment with them in their Series A. If you’re a scaled company looking to build a first-class customer support experience, feel free to reach out to Decagon at sales@decagon.ai. Also, they’re hiring!