AI Agents & Assistants
Cognigy
EU origin, foreign-owned Report an error0–5 in half steps. 5 means the rubric's top anchor is met on the evidence.
by Cognigy GmbH · www.cognigy.com
Report an error on this page Is this your product? →
Link to this page
Add ?theme=dark to the image address for dark backgrounds.
The badge shows the category and the month of the latest evaluation, and EU-Made where
that is the product’s provenance. It never shows a score.
Cognigy at a glance
- Vendor
- Cognigy 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
- Panel verdict
- The AI panel rates Cognigy 2.5/5, the mean of 6 judges. Cognigy is an AI-agent platform for contact centres on voice and chat from Cognigy GmbH of Düsseldorf. how scores are made
- As of
- Vendor pages captured 26 Oct 2026; evaluated by the panel 28 Oct 2026. all sources
Read this page as one judge. Each weighs the same scores by what they care about.
The panel's verdict
Cognigy is an AI-agent platform for contact centres on voice and chat from Cognigy GmbH of Düsseldorf. Scoring is lopsided: integrations and knowledge and agent capabilities lead at 7-8, covering prebuilt connectors to CRM, ticketing, payments and 100+ more systems, full MCP client support with per-node tool filtering, and a Simulator suite running synthetic customers with success criteria, regression testing and version comparison. Data use and training is the weakest scored area outside pricing at 2-3: the published privacy policy governs the vendor's websites and its own support bot, and judges found no public information on platform retention or a data processing agreement; the spread reflects different weight on the website retention specifics. Processing and models scores sit at 4 with no variation — model vendors are named and selectable, but we found no public information on where inference runs. Sovereignty scores read 3-4: the contracting entity is German, the site brands the product NiCE Cognigy, and we found no public statement on where customer data is hosted. Pricing transparency reads 0-1, with only a free trial and a demo.
Speaks for it
- Prebuilt connectors to CRM, ticketing, payments and 100+ more systems, with a documented REST API and developer CLI
- Full MCP client support with automatic schema updates and per-node tool allow- and deny-lists
- Simulator evaluation suite with synthetic customers, scheduled and regression testing, automatic scoring with full transcripts, and version comparison
- Model orchestration across OpenAI, Azure OpenAI, Anthropic, Google and Aleph Alpha with the LLM of your choice
- Contracting entity and GDPR controller is Cognigy GmbH of Düsseldorf with the North Rhine-Westphalia authority as regulator
Held against it
- The published privacy policy governs the vendor's websites and its own support bot, not the deployed platform
- We found no public information on where inference runs, EU processing options or bring-your-own-key deployment
- We found no public statement on where customer data is hosted; the subprocessors listed cover website tools, with Standard Contractual Clauses stated only for HubSpot
- We found no public information on SCIM provisioning, audit logs or group-level permissions over models, agents and connectors
- We found no public information on human approval before consequential agent actions or on production triggers and schedules
Best for
- You run a contact centre on voice and chat and need prebuilt connections into CRM, ticketing and payments systems
- Your team must prove agent quality before release through synthetic customer simulations, regression runs and version comparison
- You want to orchestrate models from several vendors, including the European option Aleph Alpha, through one interface
- You need agents portable across development, staging and production with tool allow- and deny-lists at each node
Avoid if
- You must review platform-grade data processing terms and retention commitments before signing — the published privacy policy covers the vendor's websites and its own support bot
- You must pin inference and hosting to the EU — the pages name providers including OpenAI, Anthropic and Google, and we found no public statement on where inference or hosting run
- You plan agents running unattended in production systems — the vendor states that MCP authentication and permission handling are still evolving
- You need published prices to budget — the only entry points shown are a free trial and a demo request
The scores
| Criterion | Panel median |
|---|---|
| Data use, training & retention | 1 of 5 |
| Models & processing location | 2 of 5 |
| Admin, identity & access controls | 2.5 of 5 |
| Integrations, connectors & knowledge | 4 of 5 |
| Agent capabilities & guardrails | 3.5 of 5 |
| European sovereignty | 2 of 5 |
Data use, training & retention
Show reasoningHide reasoning
How this is scored
What happens to prompts, uploaded files, connected data and outputs: whether any of it trains a model (the vendor's or a model provider's), how long it is kept and where logs live, whether zero-retention is available, and whether a data processing agreement covers it.
0 — No statement on training or retention; the privacy policy is a website policy that does not cover the product.
3 — A general statement that business data is "not used for training", with no word on retention, on the model providers behind the product or on logs.
5 — Business data excluded from training by contract, retention periods stated for chats and files, and a DPA offered to business customers.
8 — No-training commitments stated for the vendor and every model provider it routes to, retention configurable by the admin (including deletion of chat history), zero-data-retention terms with the model providers named, and abuse-monitoring retention described.
10 — Data use fully accounted for: training, retention, logging and abuse monitoring stated per model provider, customer-controlled retention and deletion with an audit trail, outputs and connected data treated as customer data in the DPA, and an independent audit (SOC 2, ISO 27001 or ISO 42001) covering the product.
The Head of IT
The privacy policy governs the websites and the support widget, not the platform: it sets concrete retention for contact forms, free trials and Support AI Agent inputs (deleted after 30 days), and notes Azure OpenAI will not train foundation models on support-agent inputs. We found no public information on how the platform itself treats customer prompts, transcripts or knowledge files — training use, retention windows, admin-configurable deletion or a customer data processing agreement are all absent from the captured pages. That is website-grade disclosure, not product-grade. 4
The Data Protection Officer
The privacy policy gives retention periods for website touchpoints — contact form data after nine months, support-agent input after thirty days, trial subscription data after three years — and notes that personal data shared with Azure OpenAI for the support agent will not train the foundation models. For the platform itself we found no public information on training commitments, retention of customer conversations and files, log storage, or a data processing agreement covering the product. 4
The Automation Lead
The privacy policy governs the websites and the vendor's own support bot: 30-day retention for the Support AI Agent, a no-training note scoped to Azure OpenAI on that bot, and website retention periods. It says nothing about prompts, files or outputs on the production platform, and no product data processing agreement or admin-configurable retention appears in the captured pages, so the training and retention posture of the actual product is unaccounted for. 4
The Customer-Service Director
The captured privacy policy governs the websites and the vendor's own support bot: it sets retention for website data (nine months for contact forms, fourteen for Google Analytics, thirty days for the Support AI Agent) and states that input shared with Azure OpenAI "will not be used to train the foundation models". For the platform my forty agents would actually run on, we found no public information on conversation retention, zero-retention options, or a data processing agreement covering customer chats and files. 4 6
The Works Council Member
The privacy policy is written for the vendor's websites and its own trial, demo phone bot and support assistant, not for the platform a company deploys: it states that input to the support assistant will not be used to train Microsoft's foundation models and is deleted after 30 days, but we found no public word on training, retention or prompt and transcript logging for the enterprise product itself. Order processing agreements are mentioned for website service providers, which is not a product data processing agreement a buyer can rely on. 4
The Skeptic
The privacy policy governs the websites, not the platform: it sets deletion periods for contact forms, analytics, trial data and the website support agent, and states that support-agent inputs shared with Azure OpenAI will not train the foundation models. For prompts, files and outputs inside Cognigy.AI itself we found no public information on training, retention, or a data processing agreement; the order-processing agreements described cover website service providers. 4
Models & processing location
Show reasoningHide reasoning
How this is scored
Which models sit behind the product and who runs them, where inference and storage happen, whether the customer can pin processing to the EU, choose or exclude model providers, bring its own model or key, or run the product in its own cloud or on premises.
0 — The models behind the product and the processing location are not stated.
3 — The model family is named, but where inference runs and through which provider (the model maker, a hyperscaler, the vendor) is unstated.
5 — Models and their providers named, with an EU processing option stated for storage and at least the default model.
8 — Each available model listed with provider and inference region, EU-only processing selectable by the admin and enforced across models, model providers excludable, and bring-your-own-key or own-cloud deployment documented.
10 — Processing fully under the customer's control and documented: EU-only or on-premises inference for every feature including agents and connectors, open-weight or self-hosted models supported, data flows per feature published, and a residency commitment in the contract rather than on a marketing page.
The Head of IT
The orchestration page names the model vendors behind the product — OpenAI, Azure OpenAI, Anthropic, Google and Aleph Alpha — and customers can use the LLM of their choice to balance experience, data security and cost. We found no public information on where inference or storage actually run, on an EU-processing option, or on excluding specific providers, so I cannot pin anything to Europe from these pages. 6 8 9
The Data Protection Officer
Model vendors are named — OpenAI, Azure OpenAI, Anthropic, Google and Aleph Alpha — orchestrated through one interface, and customers can use the LLM of their choice. We found no public information on where inference runs, an EU-only processing option, excluding individual providers by policy, or deployment in the customer's own cloud. 6 8
The Automation Lead
The orchestration page names the vendors behind the product — OpenAI, Azure OpenAI, Anthropic, Google and Aleph Alpha — and the knowledge page says you use the LLM of your choice, which is real model choice including a European option. But no captured page states where inference runs, offers an EU processing option, or documents own-key or own-cloud deployment, so processing location stays a blank. 5 6 8 9
The Customer-Service Director
The pages name the model vendors the platform orchestrates — OpenAI, Azure OpenAI, Anthropic, Google and Aleph Alpha — say I can use "the LLM of your choice", and describe a model-agnostic approach plus Cognigy's own Transformer-based language model for intent work. We found no public information on where inference runs, an EU processing option, or the ability to exclude specific providers for the platform itself. 6 8 9
The Works Council Member
The platform openly orchestrates models from OpenAI, Azure OpenAI, Anthropic, Google and Aleph Alpha and lets the customer choose the LLM, which is more than naming a model family — but we found no public information on where inference runs, through which provider, or how to pin processing to the EU. A truly European option exists in the model list, yet no residency statement accompanies it. 6 8 9
The Skeptic
The model providers are named and selectable — OpenAI, Azure OpenAI, Anthropic, Google and Aleph Alpha through a central orchestration interface, with the stated ability to use the LLM of your choice. But we found no public information on where inference runs, whether processing can be pinned to the EU, or bring-your-own-key and own-cloud deployment. 6 8 9
Admin, identity & access controls
Show reasoningHide reasoning
How this is scored
What the IT department controls: SSO and SCIM provisioning, roles and groups, which models, agents and connectors each group may use, retrieval that respects the permissions of the source system, audit logs, usage reporting, and policies such as data-loss prevention — and whether reporting on staff use is scoped and documented.
0 — No admin features documented; accounts are individual sign-ups.
3 — A team workspace with an owner and member invitations, but no SSO, no roles and no audit log.
5 — SSO, admin and member roles, user management, and an admin view of usage.
8 — SAML/OIDC SSO with SCIM provisioning, group-based permissions for models, agents and connectors, retrieval that inherits source-system permissions, exportable audit logs, and usage analytics with documented scope.
10 — Governance a regulated company can operate: fine-grained policies (DLP, content filters, prompt and output logging rules), audit logs streamed to a SIEM, per-agent ownership and approval, usage reporting that can be aggregated or pseudonymised for works-council agreements, and every admin action logged.
The Head of IT
The platform page lists SSO, role-based access control, end-to-end encryption and fine-grained data controls as out-of-the-box features, which is the floor for an eight-hundred-seat rollout but is a single marketing line. We found no public information on SCIM provisioning, group-level control over which models, agents and connectors each team may use, retrieval that respects source-system permissions, audit logs or SIEM export; the analytics page documents customer-conversation reporting (intents, containment, drop-off, handoffs) rather than scoped staff-usage reporting. 2 7
The Data Protection Officer
Single sign-on, role-based access control and fine-grained data controls are documented out of the box, admins get intent, containment and handoff analytics, and MCP tools can be restricted per node with white- and blacklists. We found no public information on SCIM provisioning, audit logs, or group-level rights to models, agents and connectors. 2 7 9
The Automation Lead
SSO and role-based access control are listed as out-of-the-box enterprise features, conversation analytics exist, and white- and blacklisting restricts which tools any MCP node may call — the kind of scoped permissions I want on agents. But we found no public information on SCIM provisioning, group-level control over which models or connectors a team may use, audit logs, or how reporting on staff use is scoped. 2 7 9
The Customer-Service Director
The features page lists SSO and role-based access control as out-of-the-box enterprise security, tool white- and blacklisting is available per MCP node, and analytics cover containment, drop-off and handoff reporting. We found no public information on SCIM provisioning, exportable audit logs, or group-level control over which models, agents and connectors each team may use. 2 7 9
The Works Council Member
SSO, role-based access control, end-to-end encryption and fine-grained data controls are documented out of the box, and dashboards report intents, volumes, containment and hand-offs — a real admin view of operations. We found no public information on SCIM provisioning, audit logs, export of logs, or usage reporting on staff that can be aggregated or pseudonymised for a works-council agreement, which is what I would need in writing before agreeing to this tool. 2 7
The Skeptic
A single features-page bullet promises SSO, role-based access control, end-to-end encryption and fine-grained data controls, and the documented analytics are conversation metrics such as containment and drop-off rather than staff-usage reporting. We found no public information on SCIM provisioning, group-level permissions over models, agents or connectors, audit logs, or data-loss prevention policies. 2 7
Integrations, connectors & knowledge
Show reasoningHide reasoning
How this is scored
How the product reaches company knowledge and systems: connectors to file stores, wikis, mail, CRM, ticketing and ERP, how fast and how completely they sync, the API, support for open protocols such as MCP, and how the assistant cites the sources it answers from.
0 — No integrations; users paste text or upload files one at a time.
3 — File upload and one or two connectors (for example a drive), with no API and no statement on sync or citations.
5 — A connector library covering the common file stores, wikis and collaboration tools, answers citing sources, and a documented API.
8 — Connectors to file stores, wikis, mail, CRM and ticketing with incremental sync, custom connectors or MCP support, an API and webhooks for embedding assistants and agents in other tools, and source citations on every grounded answer.
10 — The product sits in the company's systems: read and write connectors with documented scopes, MCP servers and clients, contact-centre or telephony integration where relevant, a knowledge layer that can be audited and corrected, and the data, agents and configuration exportable when the customer leaves.
The Head of IT
This is the strongest area: a hundred-plus prebuilt connectors across CRM, ticketing and payments, a documented REST API with full platform coverage, a developer CLI, an extension framework with hundreds of free plug-ins, and full MCP client support with per-node tool allow- and block-lists. Knowledge AI traces every generated answer back to its source and lets teams categorise sources and filter search scope. We found no public information on connector sync frequency or on webhooks, which keeps it just short of the top band. 2 8 9
The Data Protection Officer
Prebuilt connections to CRM, ticketing and payments plus a hundred more systems, a documented REST API with full platform coverage, an extension framework and CLI, full MCP client support with automatic schema updates, and grounded answers traceable to their source with configurable citations. We found no public information on sync behaviour for connectors or on exporting data, agents and configuration when the customer leaves. 2 8 9
The Automation Lead
MCP client support with an MCP node in the flow editor, automatic schema updates and per-node tool filtering, prebuilt connectors to CRM, ticketing and payments among 100-plus systems, a documented REST API covering the platform, a CLI that plugs into CI/CD, contact-centre and voice integration as the core of the product, and answers traceable to their sources with a correctable knowledge layer. The captured pages give no statement on sync speed or incremental sync, webhooks, or exporting data and agent configuration on exit. 2 3 8 9
The Customer-Service Director
This is what I need for order and ticket systems: prebuilt connectors to CRM, ticketing and payments plus 100+ more systems, MCP client support with no custom code and automatic schema updates, a documented REST API with full platform coverage, and an extension framework with hundreds of free plug-ins. Knowledge answers are traceable to source with output configurable "with or without sources", and deep backend integrations handle transactional queries from start to finish; we found no public information on sync speed or export on exit. 2 8 9
The Works Council Member
Prebuilt connectors to CRM, ticketing, payments and 100+ systems, MCP client support with white- and blacklisting of tools, a documented REST API with full platform coverage, a developer CLI, an extension framework with hundreds of plug-ins, and every generated answer traceable back to its source. We found no public statement on sync cadence or webhooks for embedding agents in other tools, which keeps this just short of the top band. 2 9 8
The Skeptic
This is documented in detail: prebuilt connections to CRM, ticketing, payments and 100+ more systems, a documented REST API with full platform coverage, a developer CLI into CI/CD, and full MCP client support with automatic schema updates and per-node tool allow- and deny-lists. Knowledge AI traces every generated answer back to its source with configurable output with or without citations; we found no public information on sync cadence or export of data and agents when a customer leaves. 2 8 9
Agent capabilities & guardrails
Show reasoningHide reasoning
How this is scored
What an agent can actually do and where it stops: multi-step tasks, tool use and actions in other systems, scheduling and triggers, human approval before consequential actions, run logs, evaluation and testing before release, and the limits the vendor states rather than hides.
0 — No agents; the product is a chat window, whatever the marketing says.
3 — Saved prompts or custom assistants with instructions and files, but no actions in other systems and no run history.
5 — Agents that call tools or APIs and run multi-step tasks, built without code, with a log of each run.
8 — Agents that act in connected systems with scoped permissions, human approval configurable per action, triggers and schedules, test and evaluation tooling before release, and documented limits and failure handling.
10 — Agents a business can put in production and account for: versioning and staged rollout, evaluation against test sets, monitoring of quality and cost per agent, handover to a human with full context, every action attributable to the agent and the person who deployed it, and the vendor's own statement of what agents must not be trusted with.
The Head of IT
Agents act in external systems through MCP with tool filtering, handle transactional queries end to end, and the Simulator suite — personas, missions, success criteria, scheduled regression runs, mocked third-party APIs, automatic scoring with full transcripts, version comparison and change-impact prediction — is genuinely strong pre-release evaluation. The vendor also states plainly that MCP authentication and permission handling are still evolving. We found no public information on human-approval gates before consequential actions or on live triggers and schedules, which is exactly what I would need before letting agents touch production systems. 9 10 11
The Data Protection Officer
Simulations run against specific agent versions with personas, missions, success criteria and conversation-turn limits, scored automatically with full transcripts, tracked over time for regressions, with version comparison, impact prediction and mocked third-party failures — real pre-release evaluation. Agents act in external systems through MCP with tool white- and blacklists; we found no public information on human approval before consequential actions or on per-agent cost monitoring. 2 9 10 11
The Automation Lead
Agents act in connected systems with no custom code, tools scoped by white- and blacklists, transactional queries handled end to end, environment portability across development, staging and production, live debugging and full run transcripts — and the evaluation suite is exactly what I ask for: personas and success criteria, synthetic customers, automated scoring, regression tracking, version comparison and mocked APIs so testing never touches production. We found no public information on human approval before consequential actions or on triggers and schedules for production runs, which is what keeps an agent from going live unattended in my stack. 2 8 9 10 11
The Customer-Service Director
The strongest area for my go-live requirement: synthetic customers with persona, mission and success criteria, automatic scoring against configurable criteria, simulations on demand, on schedule and as regression testing, mocked third-party responses, and version comparison plus environment portability from staging to production. Agents act through MCP servers and 100+ prebuilt tools with scoped tool filtering, and we found no public information on human approval gates per action, production triggers, or cost monitoring per agent. 2 9 10 11
The Works Council Member
The evaluation suite is unusually strong: synthetic customers with persona, mission and success criteria, automatic scoring with drill-down into failed conversations, a full transcript per run, scheduled and regression testing, version comparison and mocked failure scenarios — exactly the testing-before-release discipline I want to see. But agents act in connected systems with tool filtering and no custom code, and we found no public information on a configurable human-approval step before consequential actions, on triggers in production, or on the vendor's own statement of what agents must not be trusted with. 10 11 9 2
The Skeptic
The evaluation story is unusually strong and honest: synthetic customers with personas, missions and success criteria, simulations on demand, on schedule and as regression tests, automatic scoring with full transcripts, mocked third-party failures, and version comparison before production. Agents act in connected systems with tool filtering per node, but we found no public information on human approval before consequential actions or on triggers that run agents unattended. 9 10 11
European sovereignty
panel opinion
Show reasoningHide reasoning
How this is scored
Who the contracting entity and the parent company are, where customer data is hosted, and who the subprocessors are — model providers above all, then hosting, search, telephony and email. Independently sourced by the sovereignty pipeline; for an AI product the model provider is a subprocessor, so a European interface over a non-European model counts as such.
0 — Non-EU vendor and contracting entity, hosting unstated, subprocessors (including model providers) unnamed.
3 — A privacy policy exists, but hosting location is unstated or non-EU and the model providers behind the product are not named as subprocessors.
5 — EU hosting stated and an EU contracting entity, but prompts reach non-EU model providers or other subprocessors without a stated safeguard or EU region.
8 — EU hosting on named infrastructure, EU contracting entity, every subprocessor listed with its location including model providers, EU inference available, and a DPA offered.
10 — Sovereign end to end and evidenced: European ownership, entity, hosting and model processing, every subprocessor European or optional, data export and deletion documented, and a security or certification statement published rather than implied.
The Head of IT
The contracting entity is solidly European — Cognigy GmbH in Düsseldorf with a commercial register entry, a German VAT number and the North Rhine-Westphalia data protection authority as regulator — though the site brands itself NiCE Cognigy as part of the NiCE group. We found no public information on where customer data is hosted, and the model vendors appear on marketing pages rather than as named subprocessors with stated safeguards; the subprocessors actually listed are website tools such as Google Analytics, Microsoft Clarity and ZoomInfo, plus Azure OpenAI for the support widget. 3 4 6 11
The Data Protection Officer
The contracting entity is Cognigy GmbH of Düsseldorf, registered under HRB 79167 with a German VAT number and the North Rhine-Westphalia data protection authority named as regulator, and website subprocessors are named with order-processing agreements in place. However hosting for the product is unstated, the model providers behind the platform — including the non-European names OpenAI, Anthropic and Google — are not listed as subprocessors with their regions, and we found no public information on EU inference; the NiCE Cognigy branding appears without ownership details on the captured pages. 3 4 6
The Automation Lead
The contracting entity is Cognigy GmbH in Düsseldorf with a German commercial register number, but the site brands itself NiCE Cognigy under the NiCE group, and no captured page states where the production platform hosts customer data. The subprocessors the privacy policy lists — Google Analytics, Microsoft Clarity, Zendesk, ZoomInfo, Cloudflare — are website and marketing tools; the model providers behind the platform are named as orchestration options but not as subprocessors with locations or an EU inference statement. 3 4 6 11
The Customer-Service Director
The contracting entity is well documented and German — Cognigy GmbH in Düsseldorf, HRB 79167, with the LDI NRW as regulator — though the site brands itself NiCE Cognigy within the NiCE group. The named subprocessors (Google, Microsoft, Zendesk, ZoomInfo, Cloudflare, plus Azure OpenAI for the support bot) are website and marketing services, and we found no public information on where customer data is hosted, an EU inference region, or a subprocessor list for the platform itself. 3 4 11
The Works Council Member
The contracting entity and controller is Cognigy GmbH in Düsseldorf, with a German commercial register entry, German VAT number and the North Rhine-Westphalia data protection authority named — that much is solid German ground, though the site brands the product NiCE Cognigy as part of the NiCE group. We found no public statement on where customer data is hosted, and the model providers named include OpenAI, Anthropic and Google with no stated EU region or safeguard, so sovereignty is proven only in part. 3 4 6 11
The Skeptic
Cognigy GmbH of Düsseldorf is the GDPR controller with the North Rhine-Westphalia authority as regulator, yet the site brands the product NiCE Cognigy and consent forms reference NiCE. We found no public information on where customer data is hosted, and the providers named for model orchestration — OpenAI, Anthropic, Google among them — are not named as subprocessors; the subprocessors listed are website tools such as Google Analytics, Zendesk and ZoomInfo, with Standard Contractual Clauses stated only for HubSpot. 3 4 6 11
Pricing transparency
not rated — the vendor publishes no price
Show reasoningHide reasoning
How this is scored
Whether a company can compute the real annual cost from public pages alone: price per seat, per agent, per conversation or per usage unit, what is included in each tier, usage limits and overage, which models, connectors or agents cost extra, minimum seats and term.
0 — No public prices at all; every customer is a sales call.
3 — A starting price or "from" figure exists, but usage limits, included models and what scales the cost are unclear.
5 — Prices per tier are public with the main inclusions, but usage limits, overage, agent runs or enterprise features (SSO, audit logs, EU processing) are unpriced or "on request".
8 — Every tier priced publicly with seat or usage units, usage limits and overage rates, add-ons listed, minimum seats and term stated.
10 — Complete price computability: annual cost derivable for a given number of seats, agents or conversations, with every usage rate, model surcharge, add-on and support tier and the cancellation and data-return terms published.
The Head of IT
The captured pages document a free trial and a "See a demo" button; we found no public information on prices — no seat figures, tiers, usage limits, agent-run rates or enterprise add-ons such as SSO or EU processing. A company cannot compute an annual cost from these pages; every commercial question starts with a sales conversation. 4 10
The Data Protection Officer
The captured pages advertise instant access to a free Cognigy.AI trial and a demo, but we found no public prices at all — no figure per seat, tier, conversation or usage unit. A company cannot compute any annual cost from the public pages. 4 10
The Automation Lead
The captured pages show an instant free trial and a demo gate, and we found no public prices beyond that — no tiers, no per-seat or per-conversation figures, no usage limits, no minimum term. Real annual cost is a sales conversation, not computable from the pages we read. 4 10
The Customer-Service Director
The only commercial facts on the captured pages are a free trial and a "See a demo" button — we found no public prices at all, no per-conversation or per-seat rate, no tiers, limits or add-ons. My CFO cannot forecast a single number from this; every quote will be a sales conversation. 4 10
The Works Council Member
We found no public prices on any page captured: no per-seat or usage figure, no tiers, no inclusions, only a free trial and a demo request as the entry points. A company cannot compute even a rough annual cost from public information; every quote is a conversation with sales. 4 10 2
The Skeptic
We found no public prices on any captured page: no per-seat, per-conversation or per-agent figures, no tiers and no usage limits. The only pricing-adjacent facts are an instant free trial and a "See a demo" invitation, so the annual cost for a given number of seats or conversations is not computable from what is published. 4 10
European sovereignty — proven facts
0 of 4 dimensions provenBuilt only from facts shown on the vendor's own pages. A dimension we could not prove is left open, not scored as zero.
| Legal entity | Not determined ⚠ unverified | — | uncited Report an error |
|---|---|---|---|
| Ownership | Not determined | — | uncited Report an error |
| Data residency | Not determined | — | uncited Report an error |
| Subprocessors | Not determined | — | uncited Report an error |
Questions about Cognigy
Is Cognigy European?
Partly. whats-best.ai classifies Cognigy as EU origin, foreign-owned.
Where does Cognigy store data?
Not stated on the vendor’s pages as of 26 Oct 2026.
Does Cognigy have a free plan?
Not stated on the vendor’s pages as of 26 Oct 2026.
What are the best alternatives to Cognigy?
As of 28 Oct 2026, the panel rates Langdock (4/5), Dust (3.5/5) and Unique AI (3.5/5) 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 (EU-Made). Ratings are the AI panel’s opinion, formed from public evidence: how scores are made. The full list, with European products first: Cognigy alternatives.
Where this could be wrong
- Evidence ages. The oldest capture behind this page is from 26 Oct 2026. Vendors change pricing and policies without notice; every fact reflects its source as of the capture date shown in the registry.
- Weak sourcing — Legal entity. Not confirmed on the vendor’s own pages as captured.
- AI can misread a source. Extraction and judgement are automated; a citation guarantees traceability, not infallibility. If something here is wrong, say so — no account needed, every report is decided within 5 business days, and accepted corrections are published.
What we left out
A claim that does not survive our checks costs us the claim, not the page. This is what was taken off this one.
- We found no public information on pricing on the pages we read (cognigy.com, cognigy.com/platform/ai-agent-studio, cognigy.com/legal-notice, cognigy.com/privacy-policy, cognigy.com/product-updates/cognigys-universal-language-understading-model, cognigy.com/product-updates/llm-conversational-ai and 5 more). If the vendor publishes it somewhere else, send us the page. Know more? Tell us
- 27 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. Know more? Tell us
- 4 subprocessors 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. Know more? Tell us
- 1 compliance 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. Know more? Tell us
- 1 integrations 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. Know more? Tell us
- 1 legal 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. Know more? Tell us
- 1 support 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. Know more? Tell us
- 1 sovereignty dimension could not be confirmed on the vendor’s own pages and is shown as unknown. Know more? Tell us
Sources (11)
The pages every claim on this page was read from — each one checked, dated, and kept verifiable.
- 1 Vendor homepage www.cognigy.com Checked 26 Oct 2026 Details →
- 2 Features page (AI Agent Studio) www.cognigy.com Checked 26 Oct 2026 Details →
- 3 Legal notice www.cognigy.com Checked 26 Oct 2026 Details →
- 4 Privacy policy www.cognigy.com Checked 26 Oct 2026 Details →
- 5 Models & processing location — found from sitemap www.cognigy.com Checked 26 Oct 2026 Details →
- 6 Models & processing location — found from sitemap www.cognigy.com Checked 26 Oct 2026 Details →
- 7 Admin, identity & access controls — found from sitemap www.cognigy.com Checked 26 Oct 2026 Details →
- 8 Integrations, connectors & knowledge — found from sitemap www.cognigy.com Checked 26 Oct 2026 Details →
- 9 Integrations, connectors & knowledge — found from sitemap www.cognigy.com Checked 26 Oct 2026 Details →
- 10 Agent capabilities & guardrails — found from sitemap www.cognigy.com Checked 26 Oct 2026 Details →
- 11 Agent capabilities & guardrails — found from sitemap www.cognigy.com Checked 26 Oct 2026 Details →