FOR PRODUCT TEAMS

In-product answers at the moment of confusion

Embed cited AI answers into your app with the Cmd+K widget and REST API, and mine every question as product feedback — where users get stuck, in their own words.

Sound familiar?

Users hit friction and leave, silently

Most confused users never file a ticket — they poke around, fail, and downgrade their opinion of your product. The help center is three tabs away and might as well not exist.

In-app help is a static tooltip graveyard

Tours and tooltips answer the questions you predicted, not the one this user has right now. Maintaining them is a release-time chore that always slips.

Product feedback arrives pre-filtered and late

By the time confusion becomes a support ticket or an NPS comment, it has been summarized, sanitized, and delayed. You never hear how users actually phrase their problems.

How Kelu helps product & onboarding

01

Native in-product assistant

The embeddable widget (chat, sidebar, and Cmd+K modes) and the streaming REST API put cited answers inside the product, at the moment of confusion.

02

Answers that ship with the release

Docs and changelog sources re-sync automatically, so the in-app assistant answers about the feature you shipped this morning — no tooltip maintenance pass.

03

Questions as a product research stream

Every question is logged verbatim with analytics on question volume, languages, and gaps — a continuous record of where users get stuck, in their own words.

04

Related articles and suggested questions

The public API serves suggested questions and related articles per knowledge base, powering contextual help surfaces beyond chat.

Relevant sources & integrations

Common knowledge sources and delivery channels for product & onboarding.

Web WidgetREST APIWebsite crawlerGitHubNotionYouTube

Got questions?

Frequently asked questions

How much engineering work is the in-app integration?

The widget is a single script tag with theming options. For deeper integration, call the public REST API directly or use the React and JavaScript SDKs on npm.

Can we build a fully custom UI?

Yes. The public streaming chat and search APIs (client-key authenticated) expose everything the widget uses, so you can build any experience — command palettes, inline helpers, onboarding checklists — on the same RAG pipeline.

How do we use the question data for product decisions?

Analytics roll up question volume per day, languages, and knowledge gaps per knowledge base, and full conversation logs preserve the user's own phrasing — a research stream you can review alongside session recordings and tickets.

How long does setup take?

Most teams are live the same day. Connect a source — a docs URL, GitHub repo, Notion workspace, or a file upload — and Kelu crawls, normalizes, chunks, and embeds the content automatically. The website widget is a single script tag, and the Slack bot installs in a few clicks.

Are answers backed by citations?

Yes. Every answer from your docs includes a structured citations list with the URL, title, and excerpt of each source used, so users can click through and verify any claim against the original document.

Is my content used to train AI models?

No. Your indexed content, embeddings, and conversations are isolated to your workspace and are only used to answer your own questions. Content is never shared across customers, and data is encrypted at rest and in transit.

Is there a free plan?

Yes. The Free plan includes 1 knowledge base and 100 AI questions per month — no credit card required. Pro comes with a 14-day free trial, and Enterprise plans add SSO, source groups, and hiding the widget's Kelu branding.

Bring cited AI answers to your product & onboarding

Connect your knowledge sources and see cited AI answers in minutes — free, no credit card required.

Free to start · No credit card required