# Parloa — AI Agents & Assistants

Parloa by Parloa GmbH holds a panel rating of 2.5/5 from 6 AI judges (category median 3/5), as of 2026-10-28. Provenance: EU-Made; sovereignty: not determined.

- Vendor: Parloa GmbH (https://www.parloa.com)
- Category: [AI Agents & Assistants](https://whats-best.ai/software/ai-agents.md)
- Provenance (editorial classification): EU-Made
- Canonical page: https://whats-best.ai/software/ai-agents/parloa
- Alternatives: https://whats-best.ai/software/ai-agents/parloa/alternatives ([markdown](https://whats-best.ai/software/ai-agents/parloa/alternatives.md))
- Data as of: 2026-10-29 · evaluated 2026-10-28 · oldest source captured 2026-10-26

**AI-generated.** Every score, rationale and verdict here is written by a panel of AI judges from public evidence — an analysis, not a hands-on test. Every fact quotes the vendor page it was read from. No score can be bought. How scores are made: https://whats-best.ai/methodology

## At a glance

- Vendor: Parloa GmbH
- Headquarters: Not stated on the vendor’s pages as of 26 Oct 2026
- Data hosting: Not stated on the vendor’s pages as of 26 Oct 2026
- Free plan: Not stated on the vendor’s pages as of 26 Oct 2026

## The panel's verdict

Parloa GmbH, Berlin-registered, sells an AI agent platform for customer service on phone and chat. Agent capabilities is strongest, scores clustering at 8: no-code agents built from SOPs, caller authentication before account actions, simulation and A/B testing before live traffic, per-conversation drift detection, and context-carrying warm handoffs. Data use and training is the weakest finding at 0 to 2 — we found no public statement on whether conversations train a model, on retention, or on a data processing agreement. Admin and access sits at 4: central agent management, column-level access policies and audited queries are documented, while single sign-on and provisioning are not. Integrations and knowledge scores span 4 to 6 — lower ones weighing undocumented sync behaviour, source citations and open protocols, higher ones crediting connectors into CCaaS, CRMs and telephony. Sovereignty scores 5-6: Azure with a Germany-hosted stamp, Zones keeping transcripts, PII and analytics inside the chosen region, but models hosted by OpenAI or Azure and no subprocessor list or DPA captured. The captured pricing page shows no figures; booking a demo is the route.

**Speaks for it:**
- Agents are built without code from natural-language briefings or existing SOPs, with blueprint updates propagating fleet-wide
- Voice agents must authenticate the caller before acting on an account
- Performance Lab simulates dozens of failure scenarios and A/B tests versions before live callers, with per-conversation tracing and drift detection after launch
- Zones deployments keep transcripts, personally identifiable information, analytics and the services that process them inside the chosen regional boundary, on Azure with a Germany-hosted stamp
- Published certifications include ISO 27001:2022, ISO 17442:2020, SOC 2 Type I & II, PCI DSS, HIPAA, GDPR and DORA

**Held against it:**
- Data use and training scores run 0 to 2, with no public statement on whether conversations train a model, on retention periods, or on a data processing agreement
- We found no public information on single sign-on, SCIM provisioning, admin roles, or exportable audit logs for staff
- Models are hosted by OpenAI or Azure, non-European providers, and no subprocessor list with locations or data processing agreement was captured
- We found no public information on a documented public API, webhooks, MCP support, connector sync frequency, or how grounded answers cite their sources
- We found no public information on configurable human approval before consequential agent actions

**Best for:**
- You run a contact centre and want voice and chat agents that authenticate callers before account actions and hand off with full context
- You need pre-launch simulation and A/B testing plus per-conversation drift monitoring before agents touch live callers
- Your legal team wants a provable regional boundary for transcripts, PII and analytics, with a Germany-hosted Azure stamp
- You want to connect existing CCaaS platforms, CRMs and telephony while keeping current contracts in place

**Avoid if:**
- Your procurement rules out non-European model providers or requires a committed EU inference path — the platform names OpenAI or Azure as its model hosts
- Your data protection review requires a published subprocessor register with locations and a data processing agreement before shortlisting — ask the vendor: the public pages we read do not show it
- You need published prices to forecast cost before engaging sales — ask the vendor: the public pages we read do not show it

## Scores by judge and criterion

Scores out of 5, each judge scoring independently against the published rubric (v1). "n/r" = not rated: the vendor publishes no price, or the judge found no public information on that criterion.

| Criterion | The Head of IT | The Data Protection Officer | The Automation Lead | The Customer-Service Director | The Works Council Member | The Skeptic |
| --- | --- | --- | --- | --- | --- | --- |
| Data use, training & retention | 1/5 | 0.5/5 | 1/5 | 0.5/5 | 1/5 | n/r |
| Models & processing location | 2.5/5 | 3/5 | 3/5 | 3/5 | 3/5 | 2.5/5 |
| Admin, identity & access controls | 2/5 | 2/5 | 2/5 | 2/5 | 2/5 | 2/5 |
| Integrations, connectors & knowledge | 2.5/5 | 2/5 | 3/5 | 3/5 | 2.5/5 | 3/5 |
| Agent capabilities & guardrails | 4/5 | 4/5 | 4/5 | 4/5 | 4/5 | 4/5 |
| European sovereignty | 2.5/5 | 3/5 | 3/5 | 3/5 | 3/5 | 3/5 |
| Pricing transparency | n/r | n/r | n/r | n/r | n/r | n/r |
| **Weighted verdict** | **2.5/5** | **2.5/5** | **3/5** | **2.5/5** | **2.5/5** | **3/5** |

### The judges

- **The Head of IT** (v1.0): Has to roll an AI assistant out to eight hundred staff without a second shadow-IT wave. Wants SSO and SCIM, groups that decide who gets which model, agent and connector, answers that respect SharePoint permissions, audit logs for the SIEM, and a deployment option that is not "trust us".
- **The Data Protection Officer** (v1.0): Signs the DPA and gets asked where a prompt goes. Wants every model provider named as a subprocessor with its region, training excluded for all of them, retention the company controls, and an EU-only switch that also covers agents and connectors — not just the chat window.
- **The Automation Lead** (v1.0): Builds the agents that file tickets, draft offers and chase approvals across the ERP, CRM and wiki. Wants tools and actions with scoped permissions, MCP and an API, approval steps before anything is sent, run logs, and a way to test an agent before the whole company uses it.
- **The Customer-Service Director** (v1.0): Runs a contact centre with forty agents and wants half the routine calls and chats answered by an AI agent. Wants handover to a human with the full context, actions in the order and ticket systems, evaluation before go-live, and a price per conversation the CFO can forecast.
- **The Works Council Member** (v1.0): Has a co-determination right the moment a tool can monitor staff, and an AI assistant logs every prompt. Wants usage reports that can be aggregated or pseudonymised, prompt logging rules set out in writing, a human approving what agents do to colleagues, and no training on staff input.
- **The Skeptic** (v1.0): Has read "European AI" on a wrapper that sends every prompt to a US model, "agents" that are saved prompts, and "we never train on your data" with no word on the model provider behind it. Asks which model answers, where it runs, what it may do unsupervised, and what a thousand seats cost.

## European sovereignty — proven facts

Built only from facts on the vendor's own pages; a dimension we could not prove is left open, never scored as zero. Headline: not determined.

| Dimension | Finding | Points | Source |
|---|---|---|---|
| Legal entity | Not determined | — | — |
| Ownership | Not determined | — | — |
| Data residency | Not determined | — | — |
| Subprocessors | Not determined | — | — |

## Key facts

Each fact is the vendor's own statement, cited to the source page it was read from ([S*n*] → Sources).

**Compliance**
- Certifications: ISO 27001:2022 · ISO 17422:2020 · SOC 2 Type I · SOC 2 Type II · PCI DSS · HIPAA · GDPR · DORA [S12]
- Certifications: ISO 27001:2022 · ISO 17442:2020 · SOC 2 Type I & II · PCI DSS · HIPAA · GDPR · DORA [S13]

**Hosting**
- Provider: Microsoft Azure [S7]
- Regions stated: app.parloa.com · app.parloa.us [S7]

**Integrations**
- Enterprise systems: CRMs · knowledge sources · telephony [S13]

**Legal**
- Arbitration refusal: no [S5]
- Footer links: Privacy policy · Cookie settings · Imprint [S6]
- Trust center: Trust center page exists [S6]

**Product**
- Agent builder blueprint: Agent Blueprint single agent definition, updates propagate to all deployed agents [S13]
- Agent management platform modules: AI Agent Management Platform with Agent Builder, Performance Lab, Parloa Lens, and Parloa Navigator [S13]
- Agent skills: Agent Skills connectors configured once for the whole agent fleet [S13]
- Amp platform: Parloa offers an AI agent management platform (AMP); models must pass a benchmark before becoming available to build on in AMP [S8]
- Analytics regional: yes [S7]
- Audit trails: CCPA and GDPR accountability requests · yes [S12]
- Benchmark dimensions: instruction-following · latency [S8]
- Benchmark gate criteria: model must perform equivalently or better than current model within predefined acceptance margins; every critical use case and rule group must meet the bar on its own [S8]
- Benchmark gate stages: prompting guidance · applied validation · production capacity · testing on real calls [S8]
- Benchmark latency definition: end-to-end per turn, includes orchestration, network, tool execution; compared only under equivalent infrastructure and concurrency [S8]
- Benchmark mocked tools: yes [S8]
- Caller authentication: yes [S13]
- Context intelligence: Context Intelligence carries shared customer data across conversations [S13]
- Control runtime plane split: agent building · phone numbers · channels · phone calls · chat sessions · conversation data [S7]
- Crossplane datastores: yes [S7]
- Custom API integration: custom APIs and integration-capable connectors · yes · yes [S12]
- Customer identification decathlon: 20 · 74 [S12]
- Data governance controls: column-level access policies for AI agent data lake reads [S12]
- Data layer queries: CRM for identity/transactional state; data lake for behavioral and journey signals [S12]
- Data residency: yes [S12]
- Demo booking: Book a demo option [S1]
- Design surface: plain-language briefings or visual flows with change history on every edit [S13]
- Documentation location: prompting guidelines and best practices are published in Parloa's product documentation [S8]
- Guardrails escalation: scope limits and warm transfer with conversation context [S13]
- Handoff payload: transcript, classified intent, CRM record changes, authentication state, sentiment read [S12]
- Integration support: yes · CCaaS platforms · CRMs · backend systems [S12]
- Judge calibration: LLM-as-judge calibrated against human labels; small-margin decisions re-judged with per-conversation majority [S8]
- Kinoheld results: Kinoheld · 65 · 50 · 30 [S13]
- Lifecycle phases: Design and Integrate, Test and Iterate, Deploy and Scale, Monitor and Improve, Secure [S12]
- Llm region verification: OpenAI · Azure · yes [S7]
- Market local settings: per-market settings for local variation [S13]
- Migration audit: yes [S8]
- Model guidelines: model-specific prompting guidelines identify exact model version, configurations tested, evidence, and known limits [S8]
- Muenchener verein results: Münchener Verein · 10 · 3 [S13]
- Navigator sop input: Navigator converts natural-language guidance and SOPs into production-ready agents [S13]
- No code design: yes [S13]
- Observability tracing: per-conversation tracing and drift detection [S13]
- Parloa lens capabilities: always-on observability, advanced diagnostics, BI export hub [S13]
- Performance lab capabilities: simulations/evaluations across dozens of failure scenarios, A/B testing, agent tuning [S13]
- Pre conversation retrieval: supports pre-conversation retrieval of CRM and data lake fields at call connect [S12]
- Rag capability: yes [S12]
- Region constraints: yes [S7]
- Runtime caching: yes [S7]
- Runtime independence: yes [S7]
- Stamp architecture: Each stamp is an Azure subscription, an AKS cluster and all services in Parloa's runtime layer [S7]
- Stamp manifest automation: yes [S7]
- Studio central management: yes [S7]
- Subtask agents: subtask agent splitting with deterministic routing [S13]
- Tech stack: GitOps · Argo CD · yes [S7]
- Voice and chat: voice · chat · yes [S13]
- Zones data residency: yes · transcripts · PII · analytics [S7]

**Subprocessors**
- Contract requirement: models only from subprocessors with signed contracts can become available in AMP [S8]

**Support**
- Go live time: first use case typically goes live in a few weeks [S13]

## Questions

### Is Parloa European?

Yes. whats-best.ai classifies Parloa as EU-Made.

### Where does Parloa store data?

Not stated on the vendor’s pages as of 26 Oct 2026.

### Does Parloa have a free plan?

Not stated on the vendor’s pages as of 26 Oct 2026.

### What are the best alternatives to Parloa?

As of 28 Oct 2026, the panel rates [Langdock (4/5)](https://whats-best.ai/software/ai-agents/langdock), [Dust (3.5/5)](https://whats-best.ai/software/ai-agents/dust) and [Unique AI (3.5/5)](https://whats-best.ai/software/ai-agents/unique-ai) highest among the other 15 AI Agents & Assistants products evaluated. Ratings that look equal are rounded to half stars; the order uses the unrounded rating, then the name. The highest-rated European alternative is [Langdock](https://whats-best.ai/software/ai-agents/langdock) (EU-Made). Ratings are the AI panel’s opinion, formed from public evidence: [how scores are made](https://whats-best.ai/methodology#ordering). The full list, with European products first: [Parloa alternatives](https://whats-best.ai/software/ai-agents/parloa/alternatives).

## Head-to-head

- [Parloa vs PolyAI](https://whats-best.ai/compare/parloa-vs-polyai.md)

## What this page leaves out

- We could not confirm any pricing information on the vendor’s own pages as captured, so this page shows none rather than a statement we cannot stand behind.
- 64 product facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 30 integrations facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 12 compliance facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 12 legal facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 2 data facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 2 support facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 1 hosting fact could not be confirmed on the vendor’s page as captured and was left out of this page and of the panel’s material.
- 1 sovereignty dimension could not be confirmed on the vendor’s own pages and is shown as unknown.

## Sources

- [S1] Vendor homepage: https://www.parloa.com/ (checked 2026-10-26)
- [S2] Features page (agent builder): https://www.parloa.com/platform/agent-builder/ (checked 2026-10-26)
- [S3] Security and governance page: https://www.parloa.com/platform/enterprise-security-governance/ (checked 2026-10-26)
- [S4] Pricing page: https://www.parloa.com/platform/pricing/ (checked 2026-10-26)
- [S5] Legal notice: https://www.parloa.com/imprint/ (checked 2026-10-26)
- [S6] Terms and privacy notices: https://www.parloa.com/terms-policies/ (checked 2026-10-26)
- [S7] Models & processing location — found from sitemap: https://www.parloa.com/labs/insights/building-parloa-zones-repeatable-regional-deployments-with-central-control/ (checked 2026-10-26)
- [S8] Models & processing location — found from sitemap: https://www.parloa.com/labs/insights/how-parloa-benchmarks-models-for-enterprise-ai-agents/ (checked 2026-10-26)
- [S9] Admin, identity & access controls — found from sitemap: https://www.parloa.com/de/plattform/enterprise-security-governance/ (checked 2026-10-26)
- [S10] Admin, identity & access controls — found from sitemap: https://www.parloa.com/knowledge-hub/ai-governance/ (checked 2026-10-26)
- [S11] Integrations, connectors & knowledge — found from sitemap: https://www.parloa.com/platform/integrations/ (checked 2026-10-26)
- [S12] Integrations, connectors & knowledge — found from sitemap: https://www.parloa.com/knowledge-hub/conversational-ai-integration/ (checked 2026-10-26)
- [S13] Agent capabilities & guardrails — found from sitemap: https://www.parloa.com/knowledge-hub/ai-agent-builder/ (checked 2026-10-26)
- [S14] Agent capabilities & guardrails — found from sitemap: https://www.parloa.com/solutions/authenticated-agent-actions/ (checked 2026-10-26)

---

Cite as: whats-best.ai, "Parloa — AI Agents & Assistants", AI panel analysis, as of 2026-10-29. https://whats-best.ai/software/ai-agents/parloa

Corrections: https://whats-best.ai/correct/1028?target=page