Chatbots have largely been a disappointing development up to now in the world of artificial intelligence applications, with too many mistakes made when the AI fails to address a user’s actual intent. But in the world of enterprise apps, there remain some interesting, adjacent opportunities: for example, where AI may not be used to underpin a whole conversation, but could at least help get it started.
This week, one of the bigger companies working in this area, Conversica, is announcing a growth round of $31 million — to help it continue its own conversation with customers, so to speak. It will be using the funding to expand to new markets and add more features to its platform.
The Series C round is being led by existing investor Providence Strategic Growth Capital Partners L.L.C. (an affiliate of Providence Equity Partners), which also led a $34 million round in the company two years ago. Existing investors Toba Capital and Kennet Partners and new investors CIBC and Savano Capital also participated.
The company is not disclosing its valuation but off the record I was told by a source that it’s now around $300 million, after the company passed an annual run rate of $35 million earlier this year, doubling its ARR over the last two years. The Silicon Valley startup has raised $87 million raised in total over the last five years after bootstrapping for the first three of its existence, and fending off a number of acquisition approaches from some high-profile suitors.
Conversica’s CEO Alex Terry tells me that the basic idea of what Conversica does is to be the first point of engagement between a potential lead and the company in text-based interactions over email or SMS messaging, before handing off the relationship to a human.
“Our AI sales assistant engages in human conversations with marketing and sales leads to engage and qualify then, and then hand them over the most qualified leads to humans to close the deal,” he said.
What that means, in effect, is that its AI is trained to get basic contact details about a person and his/her company, and some more about the kind of interest that lead might have in contacting a company such as when a he/she is looking to make a purchase and when would be a good time to speak further. All of that is then updated in a company’s CRM system — saving the Conversica customer the time and money it would have had to spend to get that information manually.
But he is clear about one thing: “We’ve never looked at ourselves as a chatbot,” Terry said, distancing his company from much of the disenchantment we’ve seen with AI-based messaging ‘bots’ that provide automatic responses to users’ questions, which have largely failed to deliver on their promise. “Part of the trick is that we have tried to fix specific problems for specific customers. A chatbot that looks for keywords and parrots back is a frustrating experience. We’ve focused on a full problem and solving that.”
Conversica is not the only one that is building AI for this particular use case, though. There is Drift, which earlier this year raised $60 million; X.ai, another venture-backed AI startup that started by building virtual assistants to help people schedule meetings with each other but has been commandeered also to engage sales leads; Exceed.AI, which has gone so far as to buy Google ads against searches for Conversica, as one measure of how closely the two products resemble each other; and many others.
Terry believes that Conversica has a clear edge over these, not just because of the revenues it is making and their growth, but because of the size of its customer base — it works with some 1,000 large enterprises including AT&T, Box, Chrysler, Snowflake Computing, CenturyLink, the Sacramento Kings, Oracle, and SAP — and that has afforded it a huge trove of data to “teach” its system to work better.
“We’ve been doing this for eight years, and have reached over 70 million people, working out to 25 percent of all adults in the US,” said Terry. “We’ve trained our AI on over 400 million messages and there is a strong component where the AI gets smarter over time.”
As for what might come next, while today’s focus is on “entry-level” sales engagement and essentially replacing entry-level sales assistants, in the future you might expect that to extend deeper into the conversation as the system gets more sophisticated.
The company recently secured a U.S. patent for an AI-powered system that automatically carries on conversations, and Terry tells me that it’s already been testing a deeper set of “talking points” for its AI, including answering frequently asked questions, and looking into how to also use third-party data about the potential customer and the sales people to enhance how it interacts and matches up customers with salespeople. (That would put it into closer competition, or perhaps partnership, with Afiniti, a startup we wrote about last week.)
There is also the fact that today, Conversica’s natural language engine is focused around written words, and over time it expects to also add a voice component to the mix.