Agentic enterprise search

Find any answer.Then let it work for you.

Referent connects every tool your company uses, answers any question across all of it, and turns those answers into drafted emails, daily reports, and ready-to-ship work.

Start free trial
Book a demo

Search was the start. Action is the difference.

Three layers. One brain. Built for teams that need answers and outputs, not more dashboards to check.

01 · Find

Ask anything across your stack.

Slack, Drive, Confluence, Jira, GitHub, email, Notion. One question. One sourced answer. Real-time, no data duplication.

02 · Act

Turn answers into actionable items.

Daily reports, drafted emails, meeting summaries, content briefs, campaign ideas. Agents that don't just tell you — they do.

03 · Adapt

Onboarding is the product.

We set it up with you, train your team, and tune the agents to your business. Most search tools sell licenses and disappear. We stay.

One brain for your whole company.

Connected and ready to act on what your company already knows.

0%

Faster than searching manually

0+hours

Back to your team every week

0sec

From question to sourced answer

Integrated into your workflow

Referent seamlessly integrates with the tools your team already uses.

Connected and active enterprise knowledge

Referent connects emails, chats, documents, wikis, dashboards... for accurate and secure answers in seconds.

Try Referent
Contact Us
Solutions

Agentic enterprise search for every role.

Same brain. Different work. Each team gets the answers they need — and the agents that turn them into output.

Sales

Find

Past deals with similar customers. Pricing precedents. Objection responses. Competitive intel.

Act

Draft follow-ups. Generate proposal first drafts. Pre-call briefs in your inbox. Daily pipeline summary.

Referent · Sales
Demo with Lumio at 11:00. Brief me.

30-person agency, found us via Google. Pain: client reporting eats 8 hours a week. Decision-maker is Priya (CEO). Last call she pushed on price. Pulled 2 comparable wins, drafted a 3-slide deck and added prep notes to your calendar event.

HubSpot · Gmail · Calendar
How we compare

Real-time answers. Real control.

See how Referent stacks up against most enterprise search tools on data protection, governance, and connecting to the stack you already use.

Indexing model

  • Most enterprise search tools

    Duplicates all your data into another index

  • Referent Agentic Search

    Real-time retrieval. Your data is never duplicated

Data freshness

  • Most enterprise search tools

    Slow re-indexing keeps answers out of date

  • Referent Agentic Search

    Granular indexing controls keep answers current

Connector coverage

  • Most enterprise search tools

    Limited connectors leave critical context behind

  • Referent Agentic Search

    Broad coverage of the tools you already use

Deployment & LLM keys

  • Most enterprise search tools

    Vendor-managed only, with rigid integration options

  • Referent Agentic Search

    Bring your own cloud (BYOC) and your own LLM API key

Retention & governance

  • Most enterprise search tools

    Shares prompts with providers and one-size-fits-all permissions

  • Referent Agentic Search

    Zero retention with leading LLMs and access that mirrors your existing permissions

Pricing

  • Most enterprise search tools

    Inflated, underused features that drive up the price

  • Referent Agentic Search

    Affordable and scalable for teams of any size
Trust by design

Your data stays your data.

Referent protects your company's knowledge in the AI era. We enforce policies, safeguard sensitive information, and keep you compliant, without slowing your team down.

Your question
Real-time fetch APIs
Your connectors
Sourced answer
Action taken
Search personal files, emails, and chats with real-time retrieval. Personal connectors run at the user level and pull answers on demand, so private data sources are never indexed. Your most sensitive information never sits on our system or gets exposed to other employees.
Decide what gets indexed and what doesn't. Per-app controls let you choose exactly which data is shared with Referent and which stays untouched. A layered security strategy lets you tune the trade-off between speed and privacy on every connector.
Zero retention agreements with leading LLM providers ensure that prompts, responses, and metadata are never stored after they're used. Nothing is logged, nothing is reused for training. Your queries serve the moment, then disappear.
Run Referent in your own AWS, Azure, or Google Cloud environment with your own LLM API key. You keep full control over access, logs, and rate limits, with complete visibility into every model interaction. Your data never leaves your perimeter.

Agents that automate your workflow.

Book a demo
Try Referent Free

FAQs

Referent is an AI assistant that connects to all your company's tools (Slack, email, documents, CRM, and more) and delivers instant, sourced answers to any question. It also generates a personalized daily briefing of key KPIs, decisions, and risks, so executives and teams always have what they need to act.
Referent connects seamlessly with 50+ enterprise tools including Slack, Gmail, Google Drive, Notion, Confluence, Salesforce, HubSpot, Jira, Microsoft Teams, and many more. Setup typically takes minutes per integration using secure OAuth or API key authentication.
Yes. Referent is built for regulated environments: GDPR and ISO 27001 aligned, full encryption in transit and at rest, single-tenant architecture, granular audit logs, and zero use of your data for model training. Enterprise customers can also self-host or deploy in their own cloud (BYOC) for complete data residency control.
By default, Referent automatically routes each query to top-tier models from OpenAI, Anthropic, and Google (GPT-4 class, Claude, Gemini) and picks the best one for the task, optimized for quality and cost, so you don't manage API keys or model infrastructure. Enterprise customers can bring their own LLM provider and API keys to keep all inference under their existing AI agreements.
Permissions in Referent mirror exactly the permissions from your connected systems. Each user only sees information they already have access to in the original tools, with granular controls to protect sensitive data.
Generic AI tools like ChatGPT don't have access to your company's data. Referent is specifically built to connect to your internal tools and deliver answers grounded in your actual business information, not general knowledge. Every answer comes with citations so you can verify the source.
Both. Referent can answer questions about your business data instantly, and it can also take actions: schedule meetings, create calendar events, draft documents, and trigger workflows in connected tools. The level of automation is configurable per team.
Yes. Referent maintains persistent memory of your preferences, recurring queries, and business context. Over time it learns what matters most to you and surfaces relevant updates proactively in your daily briefing.
Most teams are up and running within a single day. Connecting your first integrations takes minutes. For Enterprise deployments with custom configurations and SSO, implementation typically takes 1–3 business days with the help of our onboarding team.
Referent only answers from your connected data sources and always cites its sources, so accuracy is tied directly to the quality and completeness of your integrated tools. You can see exactly where every answer came from and drill into the original document or message.
No. Your company data is never used to train any AI model. Referent processes your data only to answer your queries, and all data remains encrypted and under your control at all times.
Yes. Referent integrates natively with both Slack and Microsoft Teams, allowing your team to ask questions and receive answers directly inside the chat tools they already use, with no separate app required.
Yes. Referent doesn't wait to be asked. It continuously monitors your connected tools and proactively surfaces what matters: emerging risks, pending decisions, and key metric changes. Every morning it delivers a personalized briefing with the updates most relevant to you, so you stay ahead of your business rather than catching up to it.
Yes. Referent has a Free plan with no credit card required, so you can connect your tools and start getting sourced answers right away. Paid plans start at $9/month for Premium (introductory price) and scale up to Pro and Max as your usage grows.
Referent inherits the access controls from every connected source, so sensitive documents and messages are only accessible to users who already have permission in the original system. Admins can also mark specific data sources or channels as off-limits for AI retrieval.
The standard plans (Free, Premium, Pro, Max) include all core features: 50+ integrations, sourced AI answers, daily briefings, and managed top-tier LLMs optimized for cost and quality. Enterprise adds bring-your-own LLM (custom providers and API keys), self-hosted or bring-your-own-cloud deployment, SSO/SAML, custom integrations, volume discounts, an enterprise SLA, and a dedicated customer success manager.
Referent is available in English and Spanish. You can ask questions and receive answers in either language, and your daily briefing is delivered in your preferred language. Additional language support is on the roadmap.

Be the first to hear about Referent news.

By signing up, you agree to receive marketing emails from Referent. See our Privacy Policy.