Deflect tickets at the support form
Answer the question before it becomes a ticket.
By the time a customer opens your contact form, they have already decided to file a ticket — even when the answer is one search away. Your team then spends a full ticket lifecycle on something that never needed to be a ticket.
The Support Form Deflector intercepts the question when the user submits your contact form, runs a non-streaming RAG query with a confidence score, and shows a cited answer before submission. High-confidence answers are shown and can deflect the ticket; low-confidence ones let the form submit normally, so nothing gets blocked.
How it works
- 01
Index the knowledge your support team answers from — help center, docs, policies.
- 02
Call the form-deflect API from your contact form when the user submits their question.
- 03
Kelu runs a non-streaming RAG query and returns an answer with a confidence score and citations.
- 04
Above your threshold, show the cited answer before submission; below it, let the form submit normally.
What you get
- Resolve tickets at the moment of creation, before they enter your queue.
- Never block a submission — low-confidence questions pass straight through.
- Tune the confidence threshold to your own quality bar.
- See which form questions deflect and which become tickets, in analytics.
Features that make it work
Got questions?
Frequently asked questions
How is form deflection different from a chat widget?
The widget waits for users to open it; form deflection intercepts the exact question a user is already typing into your contact form. It uses a dedicated non-streaming endpoint that returns an answer plus a confidence score, so your form can decide whether to show it.
What if the suggested answer is wrong or unhelpful?
The user simply submits the form as usual — deflection never blocks submission. Answers include citations so users can verify them, and low-confidence answers are not shown at all based on your threshold.
Do I need to rebuild my support form?
No. You keep your existing form and add one API call to the public form-deflect endpoint, authenticated with a client key. The response gives you the answer, score, and citations to render however fits your UI.
How long does setup take?
Most teams are live the same day. You 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. There is no model to fine-tune and no infrastructure to run.
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, and the answer text cites sources inline. Users can click through to verify any claim against the original document — which is what makes the answers trustworthy enough for support, sales, and internal use.
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. Enterprise plans add OIDC/SAML SSO, a widget without Kelu branding, and a custom SLA.
Which knowledge sources can I connect?
24 source types: websites, GitHub (docs files, source code, issues, pull requests, discussions), Notion, Confluence, Google Drive, OneDrive, Zendesk (Help Center and solved tickets), Jira, Jira Service Management, Linear, Salesforce Knowledge articles, Salesforce Cases, Discourse, Stack Overflow for Teams, YouTube, OpenAPI specs, S3, curated Q&A, and direct file uploads (PDF, DOCX, XLSX, CSV, Markdown, TXT, HTML). Sources sync on a schedule, and GitHub sources can re-index automatically via webhook on every push.
What happens when the AI is not confident in an answer?
Answers are grounded in your sources. When they do not cover a question, Kelu returns a transparent, configurable fallback — typically pointing users to your support channel — instead of guessing. Those unanswered questions are logged in knowledge gap analytics so you can close the gap in your docs.
Related use cases
Deflect repetitive support tickets
Answer common questions before they become tickets.
Learn moreFix slow first response times
Give customers an accurate answer in seconds, not hours.
Learn moreDraft replies for support agents
AI-drafted replies inside your helpdesk, cited and editable.
Learn moreReady to deflect tickets at the support form?
Connect your knowledge sources and see cited AI answers in minutes — free, no credit card required.
Free to start · No credit card required