Avoca

The retrieved chunks do not specify Avoca’s per-user monthly cost or a minimum number of seats required to sign up. A blog post states that answering services for contractors generally cost $150–$500/month, but it does not confirm whether this is Avoca’s subscription pricing or provide a per-user breakdown. None of the retrieved chunks mention a seat minimum.

Add a dedicated pricing page or section that clearly lists per-user monthly costs and any minimum seat requirements.

Where we looked: /blog/how-much-does-an-answering-service-cost-a-guide-for-businesses

Avoca is not currently SOC 2 Type II certified. According to its security documentation, Avoca's own SOC 2 Type II certification is in progress with an estimated completion date of December 15, 2025, while its SOC 2 Type I report is expected in the near term. Its infrastructure providers are SOC 2 Type II certified. Avoca states that it maintains compliance with HIPAA (if applicable), along with GDPR and CCPA/CPRA. The retrieved site content does not mention PCI-DSS compliance.

Add explicit PCI-DSS compliance status to the security documentation and clarify what 'HIPAA (if applicable)' means for customers.

Where we looked: /custom-integration/security-program-overview.md, /custom-integration/security-program-overview.md, /custom-integration/security-governance.md, /custom-integration/security-data-protection.md

Avoca publishes a Target SLA of 99.9% uptime annually (≤8.76 hours downtime/year) in its public documentation and maintains a public status page. However, the retrieved chunks do not state a guaranteed monthly availability percentage; the status page only lists historical performance figures (e.g., 99.54%–99.95% for Apr–Jun 2026).

Add a dedicated SLA page or section that explicitly states the guaranteed monthly availability percentage, or clarifies how the 99.9% annual target translates to monthly commitments.

Where we looked: /custom-integration/security-monitoring.md, /custom-integration/technical-considerations.md, status.avoca.ai, status.avoca.ai

Avoca's API rate limits are **1,000 requests per minute per API key** and **10,000 requests per hour per workspace**. For outbound messages, teams can set **daily send caps** so total outbound volume stays within agreed thresholds. The chunks do not specify a separate daily cap for outbound AI voice calls beyond these general API rate limits.

Add explicit documentation on whether outbound AI voice calls have a daily cap or are only limited by the general API rate limits. Include default daily send cap values for messages if they exist.

Where we looked: /api-reference/introduction.md, /faq/messaging-throttling.md, /faq/messaging-throttling.md

Avoca's infrastructure is AWS-based with primary AWS US infrastructure, and they state they defer to organizational data residency requirements for healthcare/customer data (chunk [1]). They also note multi-region capabilities are leveraged where available from infrastructure providers (chunk [14]). However, the chunks do not explicitly specify the exact geographic regions where call recordings and customer data are physically stored, nor do they explicitly confirm whether US-only data residency can be enforced for call recordings specifically.

Add a dedicated data residency or security page that explicitly lists the AWS regions where call recordings and customer data are stored (e.g., us-east-1) and clarifies whether customers can contractually enforce US-only data residency.

Where we looked: /custom-integration/technical-architecture.md, /custom-integration/security-monitoring.md

Avoca writes conversation transcripts, summaries, tags, and job fields back to ServiceTitan job and customer records, and the Web Chat documentation references `live_chat.*` webhooks/API. However, the retrieved chunks do not describe user-configurable outbound webhooks for pushing conversation data into internal data warehouses or BI tools. The webhook guides instead cover inbound Speed-to-Lead triggers for initiating AI calls from external forms.

Add a dedicated integrations or API reference page documenting outbound event webhooks, including available event types (e.g., conversation.completed), payload schemas, and how to configure custom endpoints for data warehouse or BI ingestion.

Where we looked: /faq/service-titan-sync.md, /config/web-chat.md, /products/speed-to-lead.md, /products/speed-to-lead.md, /products/speed-to-lead.md

The retrieved chunks do not specify Avoca’s disaster recovery RPO or RTO, nor do they state how frequently platform backups are tested. Chunks [1] and [19] include a pre-production checklist item that simply says “Disaster recovery tested.” Chunk [2] describes a “Monitoring, Logging, and Resilience” page as covering “Operational telemetry, backup planning, and reliability benchmarks for the Avoca platform,” but chunk [3]’s retrieved content under that page only details audit-log retention (e.g., “AWS Infrastructure: 1 month retention,” “Datadog Application Monitoring: 15 days retention”) and omits any backup-planning specifics.

Publish specific RPO and RTO targets (e.g., in hours or minutes) and state the frequency of backup/restoration testing (e.g., monthly, quarterly) in the Monitoring, Logging, and Resilience or Security documentation.

Where we looked: /custom-integration/technical-considerations.md, /custom-integration/technical-considerations.md, /llms.txt, /custom-integration/security-monitoring.md

The retrieved chunks do not specify a confidence threshold at which Avoca's AI automatically escalates a call or chat to a human agent. However, chunk [4] indicates that escalation logic is configurable per workflow/campaign: users can "Define which situations should escalate the conversation to a human agent" within a workflow, and preset workflows include recommended escalation logic. The only confidence threshold mentioned in the chunks is a "high confidence threshold" for auto call reasons reclassification (chunk [2]), which is a separate feature from live escalation.

Document the specific confidence threshold (if any) that triggers escalation to a human agent for calls and chats, and clarify whether that threshold can be adjusted per campaign or workflow.

Where we looked: /end-to-end-campaign-creation-and-best-practices.md, /How-to-set-up-auto-reclassification-2cef2b56d4d58020b670cfcab000e492

Avoca’s integration documentation lists a total estimated implementation timeline of 9–12 weeks from discovery to go-live, broken into five phases: Discovery & Design (2 weeks), Development & Unit Testing (3–4 weeks), Integration Testing (2–3 weeks), UAT & Refinement (1–2 weeks), and Production Deployment (1 week). The site does not specify how many hours of onboarding support are included, though it notes that onboarding includes an assigned CSM, testing calls, a go-live call, bi-weekly sync meetings, and a shared Slack channel.

Add a clear statement of the typical implementation timeline for standard (non-custom) deployments and specify the number of included onboarding support hours (e.g., 'up to X hours of CSM-led onboarding') on the getting-started or pricing page.

Where we looked: /custom-integration/testing-support.md, /Tips-for-a-successful-Go-Live-2cef2b56d4d58001ad12d64589bdbfaa

Avoca supports exporting call data from the dashboard as CSV files. From the Coach Dashboard, the exported CSV includes call ID, date, score, agent name, customer name, call type, call reason, direction, campaign, booking status, tags, summary, rubric name, and individual rubric item results. From the KPI dashboard, the CSV export includes transfer destination names along with filtered call data. The retrieved chunks do not mention any file formats besides CSV, and they do not describe a separate export for conversation logs or transcript-level data.

Add documentation clarifying whether conversation logs or transcripts can be exported, what format(s) they use, and whether CSV is the only supported export format across all dashboard modules.

Where we looked: /Navigating-the-Coach-Dashboard-2cef2b56d4d580c5b3fed201f3345655.md, /kpi-dashboard.md

Avoca supports multiple teams under a single enterprise account and implements role-based permissions (Admin, Member, Agent) with team-based data isolation, but the retrieved chunks do not state a maximum number of teams or business locations allowed per account, nor do they explicitly confirm whether role-based permissions can be scoped to an individual business location rather than a team.

Add a clear statement on the maximum number of teams or business locations supported per enterprise account, and clarify whether permissions can be scoped to individual locations versus teams.

Where we looked: /Inviting-your-Managers-and-CSRs-2cef2b56d4d580749983f884d5a976e2, /custom-integration/security-access-controls.md, /Settings-Adding-Members-to-the-Avoca-Dashboard-2f0f2b56d4d580b2a4c4e115a0af92bb.md, /how-to-set-up-sso.md