By Sagar Shankaran, Founder of CallSphere
Call Detail Records are the cheapest, oldest, and most underused signal in voice AI. Joined to LLM traces and STT confidence they tell you why a campaign is converting at 8% instead of 12%. Here is the schema and the queries we ship.
Key takeaways
Every PBX has logged Call Detail Records since the 1980s. Modern AI voice teams treat CDR as a billing artifact and ignore it for analytics. That is a mistake. A CDR joined to LLM traces, STT confidence per turn, and CRM outcome is the highest-signal training data you have for outcome attribution.
Most platforms expose a CDR table with: call_sid, from, to, duration, status, price. That is enough to bill, not enough to optimize. Without joining to LLM trace, STT confidence, agent_id, campaign_id, and CRM outcome, you cannot answer "which agent variant converted best on inbound calls under 60 seconds from area code 415 last week."
The second mistake: not building the CDR cube as a star schema. Querying raw CDR per call across millions of records is slow. Pre-aggregating into hour x agent x campaign x outcome dimensions makes dashboards instant.
Persist a per-call enriched CDR row with: call_sid, tenant_id, agent_id, campaign_id, direction, duration, sip_final_code, mos_avg, ttft_p95, llm_tokens, stt_confidence_avg, intent_detected, outcome (booked/declined/voicemail/no_answer), revenue_attributed. Build hourly rollups by every dimension. Surface conversion rate, average handle time, and cost-per-outcome on the tenant dashboard.
flowchart LR
A[Twilio CDR] --> B[Enrichment service]
C[LLM traces] --> B
D[STT confidence] --> B
E[CRM outcome] --> B
B --> F[Enriched CDR table]
F --> G[Hourly rollup cube]
G --> H[Tenant dashboard]
G --> I[Outcome attribution model]
CallSphere persists enriched CDRs across all six verticals (Healthcare AI, Real Estate AI, Sales Calling AI, Salon AI, IT Helpdesk AI, After-Hours AI) into one of 115+ DB tables. Each of our 37 agents emits agent_id and the same campaign_id flows through Twilio CDR, LLM traces, and CRM webhook. Our admin dashboard shows hourly conversion per agent per campaign per direction. Starter ($149/mo) gets daily aggregates; Growth ($499/mo) ships the full cube with custom segments; Scale ($1499/mo) adds revenue attribution and predictive next-best-action. 14-day trial. Affiliates earn 22% on every plan.
Hear it before you finish reading
Talk to a live CallSphere AI voice agent in your browser — 60 seconds, no signup.
What is the minimum useful CDR? call_sid, direction, duration, sip_final_code, agent_id, outcome. Without outcome, you cannot do attribution.
How do I attribute revenue? Send a CRM webhook on deal-won with the call_sid (or campaign_id + phone) and join to CDR. Distribute revenue across calls in the path with first-touch or multi-touch model.
Should CDR live in OLTP or OLAP? Both. Hot 30 days in TimescaleDB for fast queries; cold archive in S3+Athena or BigQuery for year-over-year cohort analysis.
Does Twilio price CDR queries? Voice Insights has a free tier of summary records and paid Advanced Features for ten-second metrics. Standard CDR via StatusCallback is free.
How fast can I get attribution running? A two-table join (CDR x CRM outcome) and a daily Grafana panel is a one-day build. Star-schema rollups for fast slicing is a one-week build.
Start a 14-day trial, explore pricing for the full attribution cube on Growth, or book a demo. Healthcare on /industries/healthcare; partners earn 22% via the affiliate program.
CDR Analytics for AI Voice Agents in 2026 sits on top of a regional VPC and a cold-start problem you only see at 3am. If your voice stack lives in us-east-1 but your customer is calling from a Sydney mobile network, the round-trip time alone wrecks turn-taking. Multi-region routing, GPU residency, and warm pools become the difference between "natural" and "robotic" — and it's all infra, not the model.
Still reading? Stop comparing — try CallSphere live.
CallSphere ships complete AI voice agents per industry — 14 tools for healthcare, 10 agents for real estate, 4 specialists for salons. See how it actually handles a call before you book a demo.
The big fork is managed (OpenAI Realtime, ElevenLabs Conversational AI) versus self-hosted on GPUs you operate. Managed wins on cold-start, model freshness, and zero-ops; self-hosted wins on unit economics past a certain conversation volume and on data residency for regulated verticals. CallSphere runs hybrid: Realtime for live calls, self-hosted Whisper + a hosted LLM for async, both routed through a Go gateway that enforces per-tenant rate limits.
Latency budgets are non-negotiable on voice. End-to-end target is sub-800ms ASR-to-first-token and sub-1.4s first-audio-out; anything beyond that and turn-taking feels stilted. GPU residency in the same region as your TURN servers matters more than choosing a slightly bigger model.
Observability is the unglamorous backbone — every conversation produces logs, traces, sentiment scoring, and cost attribution piped to a per-tenant dashboard. HIPAA + SOC 2 aligned isolation keeps healthcare traffic separated from salon traffic at the storage layer, not just the API.
Why does cdr analytics for ai voice agents in 2026 matter for revenue, not just engineering? The IT Helpdesk product is built on ChromaDB for RAG over runbooks, Supabase for auth and storage, and 40+ data models covering tickets, assets, MSP clients, and escalation chains. For a topic like "CDR Analytics for AI Voice Agents in 2026", that means you're not starting from scratch — you're configuring an agent template that's already been hardened across thousands of conversations.
What are the most common mistakes teams make on day one? Day one is integration mapping (scheduler, CRM, messaging) and prompt tuning against your top 20 real call transcripts. Day two through five is shadow-mode running, where the agent transcribes and recommends but a human still answers, so you can compare side-by-side. Go-live is the moment your eval pass-rate clears your internal bar.
How does CallSphere's stack handle this differently than a generic chatbot? The honest answer: it scales until your tool catalog gets stale. The agent is only as good as the integrations it can actually call, so the operational discipline is keeping schemas, webhooks, and fallback paths green. The platform handles the rest — observability, retries, multi-region routing — without your team owning the GPU layer.
Want to see how this maps to your stack? Book a live walkthrough at calendly.com/sagar-callsphere/new-meeting, or try the vertical-specific demo at sales.callsphere.tech. 14-day trial, no credit card, pilot live in 3–5 business days.
Written by
Sagar Shankaran· Founder, CallSphere
Sagar Shankaran is the founder of CallSphere, where he builds production AI voice and chat agents deployed across healthcare, hospitality, real estate, and home services. He writes about agentic AI, LLM engineering, and shipping voice agents that handle real calls in production.
See how AI voice agents work for your industry. Live demo available -- no signup required.
A founder's guide to the female voice generator landscape: AI female voices, Japanese voices, robot voices, and how CallSphere ships 57+ voices live.
A VoIP telephone number is a phone number that routes calls over the internet instead of copper lines. Learn what a VoIP number is, how to get one, what it costs, and how to pair it with an AI voice agent in 2026.
How we built a fault-tolerant HVAC emergency triage and tech-dispatch platform on Kubernetes — three-tier CQRS, 11 micro-agents on the OpenAI Agents SDK + LangGraph, NATS JetStream, DTMF/SMS/WebSocket acceptance, circuit breakers, and an evaluation pipeline that catches regressions before they wake a tech at 3 AM.
MOS 4.3+ is the band where AI voice feels human. Drop below 3.6 and conversations break. Here is how to measure, improve, and alert on MOS in production AI voice using G.711, Opus, and the underlying packet loss / jitter / latency math.
Sentiment is not a single number per call - it is a curve. The shape (started positive, dropped at minute 4, recovered) tells you what your AI did wrong. Here is the per-utterance sentiment pipeline and the dashboards we ship by vertical.
How a Seattle SaaS team used the Vercel AI SDK 5 agent loop to build an in-product onboarding agent that converts trial users at measurably higher rates.
© 2026 CallSphere LLC. All rights reserved.