Encore AI raises $30M to turn customer conversations into AI sales agents

Updated July 29, 2026 6:24 pm

Encore AI has raised $30 million in fresh funding to expand its platform for building AI voice agents that learn from real customer conversations and can support or replace human sales and service staff. The Series A, announced Wednesday, was led by Team8 and gives the startup more room to push deeper into financial services, where it already counts many of its early customers.

The company is betting that the most effective AI agents will not come from generic prompts alone, but from studying how top-performing employees actually talk to customers. Encore says it mines calls, emails, text messages and CRM records to identify the tactics, language and decision points that move deals forward or resolve support issues.

That approach has helped the startup win more than 40 enterprise customers globally, most of them financial institutions, while it scales a product that can either work alongside staff or operate independently. The new capital will be used to grow sales in the U.S. and broaden deployments inside large banks and insurers.

What Encore AI does and why investors backed it

Encore AI is building software that turns a company’s own customer interactions into a training set for AI agents. Rather than relying on a generic model to guess how to handle a prospect or a support issue, the platform looks for patterns in successful conversations and uses those patterns to shape agent behavior.

That pitch is attractive to enterprise buyers because it tries to connect automation with actual business outcomes. In sales and customer success, tiny differences in wording, timing and objection handling can determine whether a conversation advances or stalls. Encore argues that those details are often already hidden inside a company’s historical records.

By surfacing them, the company says it can help organizations do three things at once:

• deploy AI agents that mirror what works in the business today;
• assist employees with suggested responses and tactics in real time; and
• spot weaknesses in current customer-facing workflows.

The result is not just a chatbot, Encore says, but a system that learns from what humans have already proven successful.

How interaction mining works

Encore calls its method “interaction mining,” a process that goes beyond simply storing recordings or transcripts. The platform ingests call recordings, emails, text messages and CRM data, then organizes those exchanges into stages of the customer journey.

From there, it tries to determine which parts of a conversation helped move a sale or service case forward and which ones caused friction. The goal is to build a practical playbook from actual behavior rather than from abstract best practices.

Why the company says historical data matters

Encore’s view is that the best training material for an AI sales or service agent is not a large library of generic scripts, but the organization’s own record of what has worked with real customers. That record may show, for example, that one relationship manager is especially effective at opening a conversation, while another is better at closing or handling objections.

According to founder and chief executive Dvir Ginzburg, the platform aims to capture those differences and combine them into a single system. He said the company’s agents are built from multiple successful playbooks, including the kind of light humor, anecdotes and concrete examples that human relationship managers use to keep conversations moving.

Ginzburg said the agents are designed from the playbooks that have already worked in customer conversations, so the system can incorporate the language, examples and tone that drive outcomes.

That philosophy is central to Encore’s product positioning. The company is not just transcribing customer touchpoints; it is trying to convert them into operational intelligence that can be reused across teams.

Who founded Encore AI?

Encore AI was founded in 2022 by Dvir Ginzburg, who originally launched the company under the name Insait IO. The startup first focused on recommendation software for financial advisers and relationship managers before shifting into a broader platform for AI agents.

The rebrand to Encore AI reflects that evolution. What began as decision-support software for a narrow set of professionals has become a system intended to learn from a much wider set of interactions across sales and customer support.

Ginzburg’s background and the company’s customer base both help explain the startup’s direction. Financial services, with its high-value client relationships and carefully structured workflows, is a natural fit for software that tries to preserve and replicate the best human techniques at scale.

How Encore uses customer data to train agents

Encore’s platform relies on a company’s own communication history to build and refine agents. The software connects to customer relationship management systems and pulls in the conversational data surrounding a deal or support case.

That data can include voice calls, written correspondence and customer records. Once the platform breaks the process into stages, it can analyze where a specific interaction succeeded, where it failed and which employee behaviors correlated with better outcomes.

This matters because customer-facing work is rarely uniform. Different employees can be stronger at different moments in the process, and not every successful interaction follows the same script. Encore says its system is built to identify those differences and turn them into learnable patterns.

The company also says its software can help organizations diagnose weak spots in their existing processes. If a certain stage regularly stalls, or if a team loses momentum after a particular kind of objection, Encore’s analytics are meant to expose that problem.

What the agents can actually do

Encore says its agents can talk to customers directly by voice or text. They can also function as copilots for human staff, suggesting next-best responses, tactics or follow-up steps during live conversations.

That dual-use design gives the startup flexibility. Some customers may want full automation for routine interactions, while others may prefer to keep people in the loop and use the agent as an assistant.

For enterprises, that distinction is important. Many organizations are still cautious about handing sensitive customer communications over to autonomous systems. A setup that begins with human oversight and grows toward more independence can lower the adoption barrier.

How big is Encore AI’s business today?

Encore says it already has more than 40 enterprise customers around the world, with most of them in financial services. That base suggests the company has found a product-market fit in an industry where client relationships are valuable and highly regulated.

Ginzburg also said annual recurring revenue has increased by more than fivefold since the startup’s seed round, which closed less than 18 months ago. He did not disclose the company’s current revenue figure or its valuation after the Series A.

While Encore is still early in its market, the growth figures point to steady demand from large organizations looking for AI tools that are more closely tied to measurable business interactions.

Why the startup’s moat may be under pressure

Encore’s strategy gives it a clear angle in a crowded AI market, but it also comes with a challenge: major software vendors may eventually offer similar capabilities. CRM giants such as Salesforce, SAP, Zoho and HubSpot all have access to customer data and could add AI features that tap into their existing platforms.

That possibility is one reason investors and customers are paying attention now. Encore is trying to establish itself before larger incumbents move more aggressively into the same use case.

Still, Ginzburg argues that having data is not the same as redesigning a product around that data. In his view, legacy CRM companies would need to make conversational history the center of their systems, not just another field in a database.

He said the largest competitors do not yet treat conversation history as the core fuel for their AI systems, and that using it that way would require them to change both their implementation stack and their technology stack.

That is the company’s central defense: not exclusive access to records, but a workflow built from the ground up to interpret them in a specific way.

Who joined the round?

Team8 led the financing, while Planven, Lukatz and Garage also participated. Some banks and insurers joined as well, including a few financial institutions that had already used Encore’s product before choosing to invest.

That detail is notable. It suggests the startup is not merely selling into its target market, but in some cases turning early users into backers. For enterprise AI companies, that kind of customer-investor overlap can be a powerful signal of confidence.

It also reinforces the company’s focus on financial services, where decision-makers are often especially motivated by workflow efficiency, compliance and customer retention.

What the Series A money will be used for

Encore plans to use the new funding to expand its U.S. sales operation and to deploy the platform with more large financial institutions. That likely means more enterprise pilots, more integrations and longer sales cycles as the company targets regulated clients with complex procurement processes.

For a startup in this category, distribution is often as important as product quality. Enterprise buyers want proof that an AI system can handle sensitive conversations reliably, fit into existing systems and produce measurable returns. The new capital gives Encore more runway to make that case.

Why financial institutions matter so much

Financial institutions are especially attractive customers for AI agents because they handle large volumes of repetitive but high-stakes customer interactions. They also tend to have rich archives of recorded conversations, detailed CRM histories and strict performance metrics.

That combination makes them ideal for a product built around analyzing what already works. It also means successful deployments could lead to broader adoption across other industries later on, even if finance remains the startup’s anchor market for now.

How Encore fits into the wider AI agents race

Encore’s raise comes as enterprises continue to look for AI systems that do more than answer questions. Companies increasingly want tools that can execute tasks, follow workflows and learn from proprietary information. In that environment, AI agents have emerged as one of the most competitive categories in the industry.

What differentiates Encore is its emphasis on customer interaction history. Instead of asking an agent to reason from scratch, the startup tries to ground the system in an organization’s own evidence of successful human behavior.

That could give it an edge with buyers who care about practical results more than novelty. But it also raises the standard for performance. A system trained on real customer conversations must not only sound competent; it must consistently improve outcomes in live business settings.

In other words, the bar is higher than a typical chatbot. The value proposition depends on whether the AI can actually help close deals, resolve issues faster or reveal previously unseen problems in the sales and service funnel.

Timeline of Encore AI’s evolution

Encore’s path from recommendation software to AI agent platform shows how quickly enterprise AI startups are evolving as the market matures.

• 2022: The company launches as Insait IO, focused on recommendation tools for advisers and relationship managers.
• Seed stage: The startup begins collecting enterprise traction and refining its approach to customer-facing workflows.
• Less than 18 months before the Series A: Encore says its recurring revenue has grown more than fivefold since the seed round.
• 2026 Series A: Team8 leads a $30 million financing round as the company rebrands and scales its AI agents.

Why this funding round matters beyond one startup

Encore’s financing is a useful signpost for where enterprise AI is heading. The market is moving away from general-purpose demos and toward systems built around company-specific data, workflows and measurable outcomes.

That shift is especially visible in customer support and sales, where organizations already have long histories of recorded interactions and clear benchmarks for success. If Encore can prove that historical conversations can reliably train better agents, it could help validate a broader model for enterprise AI adoption.

The next test is whether the startup can scale without losing the specificity that makes it useful. As it expands across the U.S. and deeper into banking and insurance, Encore will need to show that its interaction-mining approach works not just in pilots, but across large, complex organizations.

For now, the $30 million Series A gives Encore the resources to make that argument at a larger scale. It also places the startup among the more closely watched players in the fast-moving race to build AI agents that do more than talk — they learn from how businesses already win.