Encatch AI

AI that builds, reads, and routes your feedback.

Draft forms from a sentence, understand open-text answers, send the right responses to the right team, and translate into 100+ languages. Encatch AI picks the best-fit frontier model for each task, or you can add your own AI provider and run every feature on it.

Start free trial

Every plan

includes AI credits

100+

languages with AI translation

6

AI providers you can bring

Build

Onboarding feedback form

5 questions · 2 pages · drafted from one sentence

Read

Top topics in open text

Onboarding 38% · Performance 27% · Navigation 18%

Route

"Reports a checkout crash"

Matched by AI filter → sent to Jira

Best-fit frontier model picked for each task

AI across Encatch

AI lives in your workflow — not just a sidebar

AI is built into the features you already use. Here's where it shows up.

Real-time Alerts

AI filters

Plain-language rules decide what reaches each destination

Privacy Controls

AI Refine

Respondents can neutralise their writing style, on every plan

Product Insights

Ask your AI

Analyse with your AI or connect the MCP server

Global Settings

Bring your own key

Run every Encatch AI feature on your own AI provider account

How encatch AI works

Build, read, route, and more.

Six ways AI takes the busywork out of feedback, from the first question to the right team.

From one sentence to a ready-to-edit form.

  • Describe what you want to learn and Encatch AI drafts the questions, question types, and page flow.
  • Or have your AI assistant build it from Claude, ChatGPT, or Cursor through the Encatch MCP server.
  • Prefer a head start? Pick from 100+ ready-made templates instead.
  • You review and edit every question before anything is published.

Design your form

Encatch AI
Ask new users why they didn't finish onboarding
Welcome screen
How clear was setup? · Rating
What stopped you? · Multiple choice
Anything else? · Long answer
Thank-you screen
Or build it from Claude, ChatGPT, or Cursor through the MCP server

Bring your own key

Run Encatch AI on your own AI provider.

Connect your organisation's AI provider and every Encatch AI feature runs on it: the form builder, Feedback Intelligence, AI filters, translation, Refine text, Feedback Q&A, sample data, and External Insights chat. Read the setup guide

Your agreement, your bill
Requests run under your organisation's contract and data terms with the provider. Usage shows up on your provider bill, and Encatch deducts no AI credits.
Six supported providers
Connect Vercel AI Gateway, OpenRouter, Amazon Bedrock, Google Vertex AI, Azure OpenAI, or Azure AI Foundry, and test the connection before you save.
Every AI feature, mapped your way
Set a default connection and model per project, then set each capability to Use default, Override, or Disabled. Tune reasoning, max output tokens, and retries to balance quality and cost.
Keys locked away, requests not kept
Keys are encrypted at rest and write-only. Encatch sends your provider only what each feature needs and doesn't store the requests or raw responses.
Growth and Enterprise, no SDK changes
An org admin turns on Bring your own key in AI settings, and each project adds its own connections from the dashboard.

AI Provider

0 AI credits

Vercel AI Gateway

Secret •••••••• · anthropic/claude-sonnet-4.5

Default

Azure OpenAI

Secret •••••••• · gpt-4.1

Capabilities

Topic categorisationUse default
AI form builderUse default
Destination AI filter (Jev engine)Override · Gateway
Sample dataDisabled
Keys are encrypted and write-only, so no one can view them after you save

What's included

AI that builds, reads, and routes

Every plan includes AI credits each month. On Growth and Enterprise, add your own AI provider and use none.

Best-fit frontier models

Encatch AI uses existing frontier models and picks the best fit for each task, so you don't have to.

Prompt to form

Describe what you want to learn and get questions, question types, and page flow you can edit.

Build through MCP

Your AI assistant builds forms as working copies you preview; publishing stays with you.

Feedback Intelligence

Topics, aspects, and sentiment on open-text answers, with trends over time.

AI filters

Plain-language rules per destination, with Jev for fast decisions or Encatch AI for deeper reasoning.

AI translation

Translate forms into 100+ languages and review every string before you publish.

AI Refine

Respondents can refine long answers so their writing style doesn't give them away.

Bring your own AI

Add your own AI provider and every Encatch AI feature runs on your account, with no AI credits used.

Product journey

Keep following the signal.

Previous chapter

Product Insights

Individual responses become patterns your team can see.

The next question

You found the answer. How quickly can the right team act?

Continue to Real-time Alerts

A little more detail

encatch AI, answered

Common questions about encatch's AI features, credits, and adding your own AI provider.

What does encatch AI do?

Encatch AI is the AI built into encatch. It drafts feedback forms from a plain-language description, groups open-text responses into topics with sentiment (Feedback Intelligence), decides which responses reach each destination (AI filters), translates forms into 100+ languages, and lets respondents refine long answers. Behind the scenes it uses existing frontier models, picking the best fit for each task.

Does encatch AI use credits?

Yes. Every plan includes AI credits each month: 100 on Free, 500 on Team, and 1,000 on Growth, and you can add more with AI Credits add-ons. Form generation, AI translation, Feedback Intelligence (1 credit per analysis run), and Encatch AI filters use credits, while Jev filters have no extra cost while Jev is experimental. On Growth and Enterprise, you can add your own AI provider instead, and AI features that run on it use no Encatch AI credits.

Can I use my own AI provider with encatch?

Yes, on Growth and Enterprise. An org admin turns on Bring your own key in AI settings, then each project adds its own connection to Vercel AI Gateway, OpenRouter, Amazon Bedrock, Google Vertex AI, Azure OpenAI, or Azure AI Foundry. Every Encatch AI feature then runs on your provider under your own agreement and billing, with no Encatch AI credits used and no SDK changes. You set a default model per project and can override or turn off each feature.

Is my AI provider key safe with encatch?

Keys are encrypted at rest and write-only, so no one can view them in encatch after you save, including admins. Encatch sends your provider only what each feature needs and doesn't store the requests or raw responses. If your provider is down or the key is revoked, encatch doesn't fall back to its built-in AI; the affected features pause until you fix the connection.

Can AI build a feedback form for me?

Yes. Describe what you want to learn and Encatch AI drafts the questions, question types, and page flow, which you can edit before publishing. Or have your AI assistant build it through the MCP server, or start from 100+ ready-made templates.

How do AI filters route responses?

Each destination can have its own filter. Manual filters match on scores or answers. AI filters check each response against a rule you write in plain language: Jev makes fast decisions, and Encatch AI reasons more deeply for nuanced rules.

How does encatch analyse open-text feedback?

Turn on Feedback Intelligence for a text question and encatch groups responses into topics and aspects, scores sentiment for each, and shows trends, so you see what people talk about and how they feel without reading every response.

Your next step

AI does the busywork. You make the call.

Draft forms, read open text, and route what matters automatically, on Encatch AI or on your own AI provider.

Start free trialRead the docs