Rulebase

What do they actually do

Rulebase builds an AI coworker for financial services teams that layers onto existing contact‑center and back‑office systems. It reviews 100% of calls, chats, and emails to flag compliance risks and fraud signals, and automates routine operations like QA, dispute evidence collection, and case filing (site, Beam).

Live products include: Coworker/AutoQA for interaction review and scorecards, Beam for real‑time agent context/coaching with live fraud and deepfake detection, and disputes automation that assembles evidence and can file/track cases with networks/providers. Typical deployments integrate with CX/telephony tools (e.g., Zendesk, Intercom, Salesforce) so Rulebase can ingest interactions, check them against policies and regulatory libraries, surface issues, and route or automate follow‑ups; teams then use dashboards and alerts to prioritize and coach (site, Beam).

They sell to banks, fintechs, and BPOs; public materials reference fintech customers and enterprise financial services buyers. The company reports usage‑based pricing (per interaction/workflow) and “double‑digit” month‑over‑month revenue growth since YC, and highlights enterprise readiness (e.g., SOC 2/GDPR claims) (YC, TechCrunch, LinkedIn, site).

Who are their target customer(s)

• Contact‑center / CX manager at a bank or fintech: They can’t feasibly review every interaction, so QA misses coaching opportunities and late‑discovered SLA or policy breaches. They need full‑coverage review and real‑time surfacing of issues (site).
• Compliance or risk officer at a financial institution: Sampling and ad‑hoc reviews leave regulatory violations and audit evidence gaps; investigations are slow and inconsistent. They need built‑in regulatory checks and auditable workflows (site, TechCrunch).
• Back‑office / disputes operations manager: Chargeback/dispute handling requires manual evidence gathering and network‑specific filings, delaying resolution. They want automated evidence assembly and case creation/filing (site).
• Fraud operations or payments investigator: Fraud often happens during live calls (voice spoofing, social engineering), and teams lack timely signals to stop losses. They need real‑time fraud/deepfake detection and agent coaching (Beam, TechCrunch).
• BPO / outsourced contact‑center manager serving financial clients: They must enforce multiple clients’ policies and SLAs across many agents, causing inconsistent quality and costly escalations. They need centralized rules, QA, and routing tailored to financial services (site, YC).

How would they acquire their first 10, 50, and 100 customers

• First 10: Founder‑led, paid pilots with early fintechs and challenger banks that replace sampled QA with full‑coverage reviews and automate one disputes workflow; leverage warm intros and clear, one‑page success metrics to convert to references.
• First 50: Publish native integrations/marketplace listings for major CX/telephony stacks and recruit BPO partners for bundled offers; run targeted outbound to compliance/ops leaders with templated rule libraries and repeatable pilot playbooks.
• First 100: Build an enterprise sales and CS motion with SOC 2/SSO, standard contracts/SCAs, and multi‑site pilots; use reference customers and quantified ROI to win procurement at banks and large BPOs and deepen partnerships with payment/dispute networks.

What is the rough total addressable market

Top-down context:

Stacking adjacent markets—contact‑center software/analytics, fraud detection, RegTech, and chargeback/dispute automation—implies a gross TAM around $110B, acknowledging overlap between categories (CoherentMarketInsights, Grand View, Fortune Business Insights, ResearchIntelo).

Bottom-up calculation:

A BFSI‑focused SAM of roughly $20B–$45B is reasonable by taking 20–40% slices of each market to reflect finance‑relevant contact‑center, fraud, compliance, and chargeback spend and avoiding double‑counting (Grand View, VerifiedMarketResearch, Fortune Business Insights, ResearchIntelo).

Assumptions:

• BFSI accounts for ~20–40% of the relevant segments (largest vertical but not the entirety).
• Overlap between categories is material, so slices are applied to reduce double‑counting.
• Focus is on geographies/customers Rulebase can practically sell to in the near term (U.S./Europe banks, fintechs, BPOs).

Who are some of their notable competitors

• NICE: Enterprise contact‑center platform with interaction recording, voice biometrics, and fraud/compliance tooling; overlaps on real‑time fraud detection and automated compliance monitoring (source, source).
• Verint: Conversation intelligence and quality/compliance suite with auto‑scoring, QA workflows, and financial‑services compliance modules—direct overlap with 100% review, QA automation, and regulatory monitoring (source, source).
• CallMiner: Speech/text analytics with real‑time alerts, QA, and fraud detection; competes on AutoQA and fraud‑signal use cases for financial services (source, source).
• Observe.AI: Real‑time agent copilots and automated QA for regulated industries, including banking—maps closely to Beam and Coworker features (live coaching, compliance checks) with enterprise controls (source, source).
• Balto: Lightweight, real‑time agent guidance and QA inside agent desktops; strongest overlap with in‑call prompts, checklists, and coaching rather than end‑to‑end disputes or deep regulatory libraries (source, source).