# Dimension Labs Docs

> How to get customer conversations, reviews, surveys and tickets into Dimension Labs, turn them into Dimensions, join them to business data, and get answers from the analytics agent. Every page is also available as Markdown at the same path plus .md.

## Home

- [Dimension Labs Docs](https://docs.dimensionlabs.io/index.md): How to get your data into Dimension Labs, turn it into Dimensions, and get answers from the agent.

## Start here

- [What Dimension Labs does](https://docs.dimensionlabs.io/get-started.md): Dimension Labs is the analytics agent for customer language. What it does with your data, what makes it different, and the meaning layer underneath.
- [Glossary](https://docs.dimensionlabs.io/get-started/glossary.md): The terms used across Dimension Labs and these docs, with the names you see in the app.
- [Key concepts](https://docs.dimensionlabs.io/get-started/key-concepts.md): Conversation data and business data, the meaning layer, Dimensions and prompts, conversations, when enrichment runs, and projects.
- [Quickstart](https://docs.dimensionlabs.io/get-started/quickstart.md): Sign up, load one file or a sample, say what you want to know, get your first read from the agent, then run a backfill so your file gets Dimensions.

## Get your data in: files and formats

- [Choose your path](https://docs.dimensionlabs.io/data.md): Which kind of data you have, where it goes in Dimension Labs, and the fastest way to get it there.
- [Business data format](https://docs.dimensionlabs.io/data/business-data-format.md): How to prepare revenue, orders, accounts and other structured files so the agent can query and join them.
- [Check a file](https://docs.dimensionlabs.io/data/check-a-file.md): Find and fix the format problems that cause failed uploads, and test a join with business data. Runs in your browser; your file is never uploaded.
- [Conversation data format](https://docs.dimensionlabs.io/data/conversation-format.md): The universal format for conversations, reviews, surveys and tickets, column by column, with what breaks each one.
- [Convert a file](https://docs.dimensionlabs.io/data/convert-a-file.md): Map your export's columns to the Dimension Labs conversation format and download a ready file. Runs in your browser.
- [Join business data](https://docs.dimensionlabs.io/data/join-business-data.md): How conversations get joined to revenue, orders and other business data, and how to make sure the join works.
- [PII redaction](https://docs.dimensionlabs.io/data/pii-redaction.md): How to remove personal information before you send data, which libraries to use in Node, Python and the cloud, and how redacted text is handled.
- [Processing times](https://docs.dimensionlabs.io/data/processing-times.md): What happens after you upload or send data, and when you can query it and see Dimensions.
- [Reviews, surveys and tickets](https://docs.dimensionlabs.io/data/reviews-surveys-tickets.md): How to shape non-chat sources into the conversation format so they enrich correctly.
- [Sessions and conversations](https://docs.dimensionlabs.io/data/sessions-and-conversations.md): How Dimension Labs groups messages into the conversations it enriches, by sessionId, and what that means for your ids.
- [Upload a file](https://docs.dimensionlabs.io/data/upload-a-file.md): Step by step, how to upload a conversation file or a business data file in Data Sources, re-upload safely, and remove data.
- [Upload errors](https://docs.dimensionlabs.io/data/upload-errors.md): Every upload problem the file checker reports, what causes it, and how to fix it, plus why uploaded data can lack Dimensions.

## Get your data in: integrations

- [Integrations](https://docs.dimensionlabs.io/integrations.md): Bring conversations in from your help desk, chat platform, contact center, survey tool, review sites or data warehouse. A guide for each, organized by type.
- [Ada](https://docs.dimensionlabs.io/integrations/ada.md): Send AI agent conversations from Ada to Dimension Labs with the universal API.
- [Amazon Connect](https://docs.dimensionlabs.io/integrations/amazon-connect.md): Send chat transcripts and call transcripts from Amazon Connect to Dimension Labs with the universal API.
- [App Store customer reviews](https://docs.dimensionlabs.io/integrations/app-store-reviews.md): Send App Store customer reviews from App Store Connect to Dimension Labs with the universal API.
- [Google BigQuery](https://docs.dimensionlabs.io/integrations/bigquery.md): Send conversation messages stored in BigQuery to Dimension Labs with the universal API.
- [Files from any tool](https://docs.dimensionlabs.io/integrations/csv-and-excel.md): Upload an export from any system: support tools, survey platforms, review sites, CRMs and spreadsheets.
- [Databricks](https://docs.dimensionlabs.io/integrations/databricks.md): Send conversation messages stored in Databricks to Dimension Labs with the universal API.
- [Freshdesk](https://docs.dimensionlabs.io/integrations/freshdesk.md): Send Freshdesk tickets and their conversations to Dimension Labs with the universal API.
- [Front](https://docs.dimensionlabs.io/integrations/front.md): Send Front conversations and their messages to Dimension Labs with the universal API.
- [Gladly](https://docs.dimensionlabs.io/integrations/gladly.md): Send Gladly conversation items to Dimension Labs with the universal API.
- [Google Play reviews](https://docs.dimensionlabs.io/integrations/google-play-reviews.md): Send Google Play reviews of your Android app to Dimension Labs with the universal API.
- [Google Business Profile reviews](https://docs.dimensionlabs.io/integrations/google-reviews.md): Send Google reviews of your business locations to Dimension Labs with the universal API.
- [Gorgias](https://docs.dimensionlabs.io/integrations/gorgias.md): Send Gorgias tickets and their messages to Dimension Labs with the universal API.
- [Help Scout](https://docs.dimensionlabs.io/integrations/help-scout.md): Send Help Scout conversations and their threads to Dimension Labs with the universal API.
- [HubSpot help desk and inbox](https://docs.dimensionlabs.io/integrations/hubspot.md): Send HubSpot help desk and conversations inbox threads, with their ticket ids, to Dimension Labs with the universal API.
- [Intercom](https://docs.dimensionlabs.io/integrations/intercom.md): Connect Intercom so conversations flow in automatically, and upload history as a file.
- [Kustomer](https://docs.dimensionlabs.io/integrations/kustomer.md): Send Kustomer conversations and their messages to Dimension Labs with the universal API.
- [LiveChat](https://docs.dimensionlabs.io/integrations/livechat.md): Send LiveChat chats and transcripts to Dimension Labs with the universal API, on a schedule, with history and backfill.
- [Medallia Experience Cloud](https://docs.dimensionlabs.io/integrations/medallia.md): Send survey feedback from Medallia Experience Cloud to Dimension Labs with the universal API.
- [NICE CXone](https://docs.dimensionlabs.io/integrations/nice-cxone.md): Send digital and voice interactions from NICE CXone to Dimension Labs with the universal API.
- [Any other system](https://docs.dimensionlabs.io/integrations/other-sources.md): Connect any system with an API by pulling its messages and sending them to Dimension Labs with the universal API.
- [Qualtrics](https://docs.dimensionlabs.io/integrations/qualtrics.md): Send survey responses from Qualtrics to Dimension Labs with the universal API.
- [QuestionPro](https://docs.dimensionlabs.io/integrations/questionpro.md): Send survey responses from QuestionPro to Dimension Labs with the universal API.
- [Amazon Redshift](https://docs.dimensionlabs.io/integrations/redshift.md): Send conversation messages stored in Amazon Redshift to Dimension Labs with the universal API.
- [Salesforce Service Cloud](https://docs.dimensionlabs.io/integrations/salesforce.md): Send Salesforce Service Cloud case emails to Dimension Labs with the universal API, plus messaging transcripts and opportunity data as files.
- [Snowflake](https://docs.dimensionlabs.io/integrations/snowflake.md): Send conversation messages stored in Snowflake to Dimension Labs with the universal API.
- [SurveyMonkey](https://docs.dimensionlabs.io/integrations/surveymonkey.md): Send survey responses from SurveyMonkey to Dimension Labs with the universal API.
- [Trustpilot](https://docs.dimensionlabs.io/integrations/trustpilot.md): Send Trustpilot service reviews for a business unit to Dimension Labs with the universal API.
- [Twilio Conversations and Flex](https://docs.dimensionlabs.io/integrations/twilio.md): Send messages from Twilio Conversations, including Flex messaging, to Dimension Labs with the universal API.
- [Typeform](https://docs.dimensionlabs.io/integrations/typeform.md): Send form responses from Typeform to Dimension Labs with the universal API.
- [Data warehouses](https://docs.dimensionlabs.io/integrations/warehouses.md): Send conversation messages stored in your data warehouse to Dimension Labs with the universal API.
- [Zendesk Support](https://docs.dimensionlabs.io/integrations/zendesk.md): Send Zendesk Support tickets and comments to Dimension Labs with the universal API.

## Get your data in: transform your data

- [Transform your data](https://docs.dimensionlabs.io/transform.md): How to turn any export into one row per message in the Dimension Labs format, with recipes, the toolkit script and a checklist.
- [Before you send it](https://docs.dimensionlabs.io/transform/checklist.md): The checks to run on a converted file before you upload it or send it to the API, the counts to compare, and the mistakes that most often slip through.
- [Attach business fields](https://docs.dimensionlabs.io/transform/recipe-business-fields.md): Recipe: add customer, account or order fields to conversations as extra columns, with key cleanup, conflict checks and a match rate.
- [Turn call transcripts into turns](https://docs.dimensionlabs.io/transform/recipe-call-transcripts.md): Recipe: split call transcripts that name each speaker into one row per turn, with the right direction and real timing.
- [Clean ticket and email text](https://docs.dimensionlabs.io/transform/recipe-clean-text.md): Recipe: strip HTML, quoted replies, sign-offs, odd spaces and placeholder text from message bodies before upload.
- [Combine several exports](https://docs.dimensionlabs.io/transform/recipe-combine-exports.md): Recipe: merge chat, email and monthly exports into one upload file without id collisions or double-counted messages.
- [Excel and Excel dates](https://docs.dimensionlabs.io/transform/recipe-excel.md): Recipe: read an .xlsx export directly, pick the right sheet and header row, convert Excel dates and keep long ids intact.
- [Fix timestamps](https://docs.dimensionlabs.io/transform/recipe-fix-timestamps.md): Recipe: convert any date format to Unix milliseconds in the right time zone, fix seconds and wrong years, and keep messages a second apart.
- [Nested JSON to rows](https://docs.dimensionlabs.io/transform/recipe-nested-json.md): Recipe: flatten nested JSON exports from a help desk or chat platform into one row per message, joined to conversation-level fields.
- [Split a large file](https://docs.dimensionlabs.io/transform/recipe-split-large-file.md): Recipe: split a large file into smaller parts with the header in each, without cutting a conversation in two, when your own tools can't handle the whole file.
- [Split a transcript into turns](https://docs.dimensionlabs.io/transform/recipe-split-transcript.md): Recipe: turn one row per conversation, with the whole transcript in a cell, into one row per message.
- [Reshape a wide survey](https://docs.dimensionlabs.io/transform/recipe-wide-survey.md): Recipe: turn a survey export with one row per respondent into question and answer rows, with scores kept as columns.
- [Recognize your data's shape](https://docs.dimensionlabs.io/transform/shapes.md): The export shapes you're likely to meet, how to tell them apart, and how each one maps to one row per message.
- [The toolkit](https://docs.dimensionlabs.io/transform/toolkit.md): One configurable script, the toolkit, that turns CSV, TSV, JSONL, JSON or XLSX exports into upload-ready files, with a report of every row it dropped or fixed.

## Dimensions

- [What a Dimension is](https://docs.dimensionlabs.io/dimensions.md): Dimensions are the typed fields Dimension Labs extracts from every conversation. What they are, the types, and how to read them.
- [Backfill](https://docs.dimensionlabs.io/dimensions/backfill.md): Run a prompt, custom or standard, over conversations that weren't enriched by the daily run, from the Backfill tab.
- [Config reference](https://docs.dimensionlabs.io/dimensions/config-reference.md): The JSON structure of a prompt, field by field, with a complete example.
- [Design custom Dimensions](https://docs.dimensionlabs.io/dimensions/custom.md): What a custom prompt is, the output types and field patterns to use, and how to design Dimensions that give reliable, useful labels on your own data.
- [How enrichment works](https://docs.dimensionlabs.io/dimensions/how-enrichment-works.md): What AI data enrichment is, how a prompt turns raw text into labeled columns, and why it works better than keywords or trained models.
- [Predicted scores](https://docs.dimensionlabs.io/dimensions/predicted-scores.md): Infer relevance, satisfaction, CSAT or NPS-style scores for every conversation, not just the few with a survey answer, with prompts you can copy.
- [Standard Dimensions](https://docs.dimensionlabs.io/dimensions/standard.md): The full list of Dimensions Dimension Labs has already written, at session and section level, and how to get them run on your data.
- [Test before you run](https://docs.dimensionlabs.io/dimensions/test-before-you-run.md): Check a prompt on a sample before you enrich everything, and what to look for in the results.
- [When enrichment runs](https://docs.dimensionlabs.io/dimensions/when-enrichment-runs.md): Which data is enriched every day, which needs a backfill, and why a new upload or a new prompt doesn't show Dimensions straight away.

## Ask and build: the agent

- [What the agent does](https://docs.dimensionlabs.io/agent.md): What the Dimension Labs agent can do with your data, how it works through a question, how it differs from a chatbot or BI tool, and what it won't do.
- [Analysis plans](https://docs.dimensionlabs.io/agent/analysis-plans.md): Turn a business question into a reviewed, reusable research plan, step by step, then run it in a fresh chat and run it again on new data.
- [Ask good questions](https://docs.dimensionlabs.io/agent/asking-good-questions.md): How to phrase questions so the agent gives you a specific, usable answer.
- [Context engineering](https://docs.dimensionlabs.io/agent/context-engineering.md): Move repeatable instructions out of the chat and into files, so the agent produces consistent work no matter who runs it.
- [Your first chat](https://docs.dimensionlabs.io/agent/first-session.md): Open the agent, ask your first question, read the answer, and recognize when the agent needs more data.
- [Quick answers and analysis plans](https://docs.dimensionlabs.io/agent/modes.md): Two ways to work with the agent: ask for a quick answer, or ask for an analysis plan for structured, repeatable deliverables.
- [Projects and standing instructions](https://docs.dimensionlabs.io/agent/projects-and-instructions.md): Use projects and AGENTS.md so the agent starts every chat with the context, files and rules it needs.
- [What the agent can see](https://docs.dimensionlabs.io/agent/what-it-can-see.md): The tables the agent works from, what's in each, and how to check what your workspace holds.

## Ask and build: reports, decks and sharing

- [Reports, decks and sharing](https://docs.dimensionlabs.io/outputs.md): Reports, decks and dashboards the agent builds from your data: publishing and revoking a link, download formats and schedules.

## Ask and build: causal analysis

- [How we prove a driver](https://docs.dimensionlabs.io/causal.md): What "prove" means in Dimension Labs, how the agent tests whether a driver is real, and how to read the result.
- [How reliable the labels are](https://docs.dimensionlabs.io/causal/validation.md): Two validation studies on how consistent and how accurate the model's labels and predicted scores are, and why enrichment is reliable.

## Ask and build: use cases

- [Chatbot and AI agent analytics](https://docs.dimensionlabs.io/use-cases/chatbot-and-ai-agent-analytics.md): Chatbot and AI agent KPIs: containment, escalation, task completion, journeys and bot health, with what to ask the agent for each.
- [Omni-channel analysis](https://docs.dimensionlabs.io/use-cases/omni-channel-analysis.md): Bring chat, phone, email, SMS, surveys, reviews and social into one stream, so you can analyze every channel together or compare them.

## Reference: API

- [When to use the API](https://docs.dimensionlabs.io/api.md): Choose between file uploads, one-click connectors and the API, and what each API endpoint is for.
- [Export API](https://docs.dimensionlabs.io/api/export.md): Pull your Dimensions out of Dimension Labs into your own systems, by integration and date range.
- [Intents and events](https://docs.dimensionlabs.io/api/intents-and-events.md): Send the intent your bot matched with each message, mark messages it didn't understand, and track events such as clicks, shares and revenue.
- [Scheduled sync](https://docs.dimensionlabs.io/api/scheduled-sync.md): Run an integration guide's code once for history, then every 15 minutes, on AWS Lambda, cron or GitHub Actions.
- [Universal API](https://docs.dimensionlabs.io/api/tracker.md): Send messages to Dimension Labs from any system, one POST per message, to the real-time or the historical tracker.

## Reference: troubleshooting

- [Troubleshooting](https://docs.dimensionlabs.io/troubleshooting.md): Find the fix by what you're seeing: upload problems, empty Dimensions, odd numbers, or an agent run that stopped.
- [Agent runs](https://docs.dimensionlabs.io/troubleshooting/agent.md): The agent stopped part way, ignored a file, or you can't find what it built.
- [Troubleshooting enrichment](https://docs.dimensionlabs.io/troubleshooting/enrichment.md): No Dimensions yet, Dimensions that are empty or partial, or values you didn't define. Causes and fixes.
- [Numbers look wrong](https://docs.dimensionlabs.io/troubleshooting/numbers-look-wrong.md): Why counts can be higher or lower than your file or your BI tool, and why a date can be a day off.

## Reference: security and compliance

- [Security and compliance](https://docs.dimensionlabs.io/admin/security.md): Dimension Labs certifications and deployment options, inviting users, single sign-on, the personal information policy, and how data is removed.

## Reference: changelog

- [Changelog](https://docs.dimensionlabs.io/changelog.md): What changed in Dimension Labs and in these docs, including retired features and pages.