Encore AI
by Encore AI
Revenue-generating AI agents that sell, collect and convert like your best reps
Encore AI is an agentic customer-interaction platform for banks, lenders, insurers and healthcare providers that mines an organisation's own call, chat, email and CRM history to learn what its top-performing reps actually do, then deploys those behaviours as AI agents across voice, chat, IVR and form-fill. Unlike deflection-focused contact-centre automation, it is built to convert leads, complete applications and recover balances rather than to close tickets cheaply.
Encore AI is an agentic customer-interaction platform for regulated, revenue-critical industries — banking, lending, insurance and healthcare — built around a patented technique the company calls Interaction Mining. Rather than scripting agents from a knowledge base, the platform ingests an organisation's own historical calls, chats, emails, IVR transcripts and CRM records, statistically isolates the actions its highest-performing human representatives take before a successful outcome, and compiles those behaviours into reusable playbooks. An agent is then assembled as a bundle of playbooks and deployed in one of two modes: Wingman, a real-time assistant that coaches a live rep mid-conversation, or Autopilot, in which the agent handles the interaction end to end across voice, chat, email and live form-fill in multiple languages. The company positions this explicitly against conventional deflection-oriented contact-centre automation: the stated objective is to qualify leads, close applications, recover overdue balances and drive upsell and cross-sell, not to resolve tickets at the lowest cost. Founded in 2022 in Israel as Insait IO by Dr Dvir Ginzburg, a former Microsoft researcher with a doctorate in geometric deep learning, the company rebranded to Encore AI and raised a $30 million Series A on 29 July 2026 led by Team8, Planven and The Garage, following a $3.5 million seed in May 2025. It reports more than 40 enterprise customers across Israel, Australia, Europe and the United States and more than 5x ARR growth since that seed round, with roughly 50 staff split between Israel, New York and Australia. Several customers, including Harel and Bank Leumi, invested in the Series A.
The VP or head of contact-centre / digital sales at a bank, lender or insurer that already has years of recorded customer conversations and is measured on conversion and collections, not on deflection rate.
Turns the tactics of your best-performing reps into agents that run the same plays across every channel and shift, instead of buying a generic chatbot and writing scripts for it.
At a Glance
- Category
- Marketing & Sales
- Pricing
- Contact for pricing, Subscription
- Target Market
- CIOs, CTOs, VP Customer Experience, Heads of Digital Sales, Contact Centre Operations Leaders
- Deployment
- Cloud-only
- Founded
- 2022
- Headquarters
- New York, United States
- Team Size
- 11-50
- Customers
- 40+ enterprise customers
Key Features
- ✓Interaction Mining
Patented analysis of historical calls, chats, emails and CRM data that isolates which specific rep actions preceded successful outcomes, and encodes them as agent playbooks.
- ✓Autopilot mode
Fully autonomous agents that handle a customer conversation end to end across voice, chat, email and live form-fill without a human in the loop.
- ✓Wingman mode
Real-time in-call coaching that surfaces the next best action to a human rep, for workflows where full autonomy is not permitted.
- ✓Multi-channel and multilingual coverage
The same agent playbooks run across voice, chat, IVR, email and web form-fill in multiple languages, so behaviour stays consistent per channel.
- ✓Funnel gap and revenue opportunity discovery
Analyses interaction and operational data to surface where revenue leaks out of the funnel before any agent is deployed.
- ✓Weeks-to-live deployment
Agents are bootstrapped by ingesting existing calls, chats, emails and CRM records rather than through a months-long scripting and configuration project.
Capabilities
Use Cases
- •Lead conversion for lenders
Inbound mortgage and personal-loan enquiries are qualified and progressed to application by an agent running the plays that the top-converting human reps use.
- •Balance recovery and collections
Agents run compliant outbound collections conversations at scale, escalating to a human only where regulation or customer hardship requires it.
- •Insurance policy upsell and cross-sell
Existing policyholders are engaged on renewal or service calls with offers that historical interaction data shows actually convert for that segment.
- •Application completion assistance
Live form-fill agents walk a customer through a partially completed application in-session, cutting the abandonment that drives most digital funnel loss.
- •Rep performance uplift
Wingman surfaces top-performer tactics to newer agents mid-conversation, narrowing the gap between best and median rep outcomes.
Ideal For
Best For
- ✓Converting inbound loan, mortgage and insurance leads that currently drop out of the funnel outside business hours
- ✓Automated collections and balance recovery conversations in regulated lending environments
- ✓Real-time rep coaching (Wingman mode) where compliance rules prevent full autonomy
- ✓Cross-sell and upsell campaigns run against an existing bank or insurer customer base
- ✓Multilingual voice and IVR containment for financial-services contact centres
Not Ideal For
- ✗Organisations without a substantial archive of recorded calls, chats and CRM outcomes — Interaction Mining has nothing to learn from, so the core differentiator does not apply
- ✗Teams wanting self-serve evaluation or transparent list pricing; there is no free tier, no published rate card and deployment is enterprise-direct
- ✗Buyers outside financial services, insurance and healthcare, where the reference base and compliance tuning are thinnest
- ✗Deflection-first support organisations whose mandate is cost-per-contact reduction rather than revenue per interaction
Deployment
Market & Ratings
40+ enterprise customers
Market Analysis
Pros
- ✓Bootstrapping agents from an organisation's own interaction archive is a genuinely different starting point from prompt-and-knowledge-base rivals, and it means the agent inherits domain and compliance nuance the buyer already has
- ✓Unusually strong customer validation signal: multiple banks and insurers were customers first and chose to invest in the Series A
- ✓Wingman/Autopilot split gives regulated buyers a path to value without committing to full autonomy on day one
Cons
- ✗No independent review presence at all — nothing on G2, Capterra, TrustRadius or PeerSpot, and no Hacker News or Reddit discussion — so every capability and outcome claim traces back to the vendor or to its funding announcement
- ✗The headline proof point ('10x ROI within months') is a single unnamed lending client with no published case study, methodology or baseline
- ✗Several named customers are also investors, which weakens the independence of any reference call a buyer arranges through the vendor
- ✗CMSWire's coverage cautions that near-term measurable gains are more likely operational (routing accuracy, reduced handle time) than direct revenue, and that revenue-side results depend heavily on how tightly the agents integrate with existing sales and conversion processes
- ✗Security and compliance certifications are not published — no trust page, no named SOC 2 or ISO 27001 attestation — which is a notable gap for a vendor selling into banking and insurance
- ✗Concentrated in financial services and insurance; buyers in other verticals are early adopters with a thin reference base, and CRM incumbents (Salesforce, HubSpot, SAP, Zoho) are shipping overlapping agentic features
Pricing
Enterprise
Contact for pricing
- ✓Interaction Mining across existing call, chat, email and CRM archives
- ✓Autopilot and Wingman agent modes
- ✓Voice, chat, IVR, email and form-fill channels
- ✓Multilingual deployment
- ✓Regulated-industry deployment support
No list pricing is published anywhere — not on the website, not in the Series A coverage. Encore AI sells enterprise-direct with a scoped implementation, and every public metric it offers is an outcome claim (a 10x ROI at one unnamed lending client) rather than a rate, so cost has to be established in a commercial conversation. There is no free tier and no self-serve trial, which makes cheap technical evaluation impossible; budget for a paid pilot on a single funnel.
Security & Compliance
Sources
This page was written from 5 sources, 4 on domains other than gainencore.ai.
Stay Ahead of the Curve
Weekly enterprise AI insights for technology leaders. No spam, no vendor pitches—unsubscribe anytime.
Subscribe