Unwrapped

Teardown · observe-ai

OBSERVE.AI

OBSERVE.AI

CategoryContact Center AILast round · $125M · 2022Site ↗
  • Menlo Ventures

Contact center voice + chat transcripts + LLM APIs + QA workflows.

01

Public data / API layer

CCaaS platform voice/chat data
CCaaS platform voice/chat dataAPI
Customer-owned call recordings
Customer-owned call recordingsYours
CRM context and metadata
CRM context and metadataAPI
Enterprise knowledge bases
Enterprise knowledge basesYours

Internal replication score

Medium
0.58

Feasibility of a useful internal substitute built with Claude (or similar), the same data access, and light agent logic — not rebuilding the whole product.

IRS = 0.30·D + 0.25·L + 0.20·O + 0.15·R + 0.10·Sthis record · 58%
  • D

    Data accessibility

    weight 0.300.65
    • 1.0mostly customer-owned / public / standard third-party sources
    • 0.5mixed accessibility
    • 0.0hard-to-access or proprietary source layer
  • L

    LLM substitutability

    weight 0.250.80
    • 1.0mostly retrieve / prompt / cite / summarize / classify / compare
    • 0.5mixed standard + custom behavior
    • 0.0strongly custom model behavior (fine-tunes on proprietary data, etc.)
  • O

    Output simplicity

    weight 0.200.50
    • 1.0straightforward internal work product (memo, list, reply, SQL query)
    • 0.5moderately specialized
    • 0.0highly specialized (e.g. FDA-graded clinical text)
  • R

    Review / risk tolerance

    weight 0.150.30
    • 1.0internal use with human review is acceptable
    • 0.5moderate risk
    • 0.0very low tolerance for error (e.g. external legal filings)
  • S

    Surface complexity

    weight 0.10inverse — higher means less surface dependence0.40
    • 1.0a simple internal shell is enough
    • 0.5polished workflow matters somewhat
    • 0.0product surface / rollout / trust posture is central to value
LabelsEasy ≥ 0.67Medium ≥ 0.34Hard < 0.34

Missing factor rows use heuristics from wrapper scores. Editorial heuristic, not investment advice.

Build it yourself→

Recreate the workflow inside your org.

Internal build

Build it yourself

Same CCaaS transcripts + CRM connectors + frontier LLM agent + policy gates — but no pre-built QA workflows or HIPAA/PCI attestations.

Internal use only. Replacing them in-market is a different bar than replaying the useful workflow inside your org.

01 · Connectors & flow

CCaaS platform voice/chat data
CCaaS platform voice/chat data
Customer-owned call recordings
Customer-owned call recordings
CRM context and metadata
CRM context and metadata
Enterprise knowledge bases
Enterprise knowledge bases

Internal build map

Data in

Connectors
Connectors

Agent layer

Planner
Tools + retrieval
Reasoning model

Logic

LLM API
ASR
retrieve
score
redact PII/PCI/PHI
enforce policy gates
not custom weights

Outputs

Internal search
Answer
Citations

02 · Claude / agent prompt

Paste as the system or developer message in Claude (or your agent runtime). Scroll to read; Copy grabs the full text.

Claude / agent prompt

// Contact center QA agent You are a contact center quality assurance agent inside [YOUR_COMPANY]. You help QA managers and team leads evaluate agent-customer interactions using ONLY materials the user is allowed to access: call transcripts from the CCaaS platform, CRM ticket context, company knowledge base articles, and QA scoring rubrics. ## What you must do 1. Retrieve first: Pull the full transcript, CRM context, and relevant knowledge base articles before scoring. 2. Score rigorously: Apply the company's QA rubric (greeting, issue resolution, empathy, compliance disclosures, closing) to each interaction. Cite specific transcript lines. 3. Surface policy violations: Flag missing disclosures, PII mishandling, or off-policy statements immediately. 4. Redact sensitive data: Strip PII/PCI/PHI from outputs. Never expose customer SSN, payment card numbers, or protected health information. 5. Scope: QA scoring, coaching recommendations, sentiment analysis, policy compliance checks — not customer-facing responses. ## What you are not Not a replacement for human QA review on high-stakes interactions (legal disputes, escalations). Internal use only. Human review required for final coaching decisions. ## Refusal Refuse if the transcript contains no clear interaction data. Refuse if asked to score an interaction the user does not have access to. Ask for clarification if the QA rubric is ambiguous. ## Safety Internal posture: outputs are for QA managers and coaches, not customers. Redact all PII/PCI/PHI. Flag any interaction that violates compliance policies for immediate human review.

03 · Result

Score this call transcript for compliance and empathy. Flag any policy violations.
ccaas-transcripts

Compliance: Pass (all disclosures present). Empathy: 4/5 (missed acknowledgment at 2:15). No policy violations.