> ## Documentation Index > Fetch the complete documentation index at: https://v1-docs.adapt.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Ask > Ask questions in natural language and get answers from your business data The first step in Adapt's Ask-Understand-Act framework. Ask any question about your business in natural language—no SQL, no dashboards, no complex queries. ## How It Works Simply type your question as you would ask a colleague: ``` Why did revenue drop last quarter? ``` ``` Which customers are most likely to churn? ``` ``` What's our top-performing marketing channel? ``` Adapt's AI agent interprets your question, identifies which data sources to query, and gathers the information needed to provide a comprehensive answer. ## What You Can Ask ### Revenue & Financial Questions ``` What's our current MRR and how has it changed? ``` ``` Break down revenue by product line for Q4 ``` ``` Compare this year's revenue to last year by month ``` ``` What's our average deal size by segment? ``` ### Customer & Sales Questions ``` Which accounts are at risk of churning? ``` ``` Who are our top 10 customers by lifetime value? ``` ``` What's the status of deals in our pipeline? ``` ``` How long is our average sales cycle? ``` ### Operational Questions ``` How are we tracking against our quarterly goals? ``` ``` What are the most common support issues this week? ``` ``` Which team members are at capacity? ``` ``` What's our customer satisfaction trend? ``` ### Research & Analysis ``` What are competitors doing in the AI space? ``` ``` Summarize recent news about our industry ``` ``` What features are customers requesting most? ``` ``` Analyze sentiment in our recent reviews ``` ## Question Types > [card] Factual > > **"What is..."** > > Direct questions about metrics, counts, or current state > > *"What's our current pipeline value?"* > [card] Analytical > > **"Why did..."** > > Questions that require investigation and analysis > > *"Why did conversions drop last week?"* > [card] Comparative > > **"How does X compare to..."** > > Questions comparing metrics, time periods, or segments > > *"How does Q3 compare to Q2?"* > [card] Predictive > > **"Which will..."** > > Forward-looking questions based on patterns > > *"Which deals will likely close this month?"* ## Tips for Better Questions ### Be Specific About Time > [tab] Good > > ``` > What was our revenue for Q4 2024? > ``` > > ``` > How did signups change over the past 30 days? > ``` > [tab] Less Effective > > ``` > What was our revenue? (which period?) > ``` > > ``` > How are signups doing? (compared to what?) > ``` ### Include Context > [tab] Good > > ``` > Check HubSpot for deals that haven't been > updated in 2 weeks > ``` > > ``` > Look at our Snowflake data for customer > cohort retention > ``` > [tab] Less Effective > > ``` > Show me stale deals (from which system?) > ``` > > ``` > What's our retention? (which metric definition?) > ``` ### Ask for Specific Formats > [tab] Good > > ``` > Show me a table of top customers by revenue > ``` > > ``` > Create a chart of monthly growth rates > ``` > > ``` > Give me 3 bullet points on pipeline health > ``` > [tab] Less Effective > > ``` > Tell me about customers (how much detail?) > ``` > > ``` > What's our growth? (in what format?) > ``` ## Follow-Up Questions Adapt maintains context throughout your conversation. After getting an initial answer, drill deeper: **Initial question:** ``` What were our top-performing campaigns last month? ``` **Follow-ups:** ``` Why did the email campaign outperform others? ``` ``` Break that down by audience segment ``` ``` Compare it to the same campaign last year ``` ``` What should we do differently next month? ``` ## Multi-Part Questions Ask complex questions that require multiple analyses: ``` Compare our Q4 performance to Q3, identify the biggest changes, and explain what drove them ``` ``` Find customers who haven't logged in for 30 days, check their support history, and flag any that might be at risk of churning ``` Adapt will break these down into steps and address each part. ## What Happens Next After you ask, Adapt moves to the **Understand** phase: 1. **Plans** the steps needed to answer your question 2. **Queries** relevant data sources 3. **Analyzes** the results 4. **Synthesizes** an answer with citations Then, if appropriate, Adapt can **Act** on the insights—drafting communications, creating reports, or triggering workflows. > [card] [Understand](https://v1-docs.adapt.com/docs/features/understand.md) > > Learn how Adapt investigates and synthesizes answers > [card] [Act](https://v1-docs.adapt.com/docs/features/act.md) > > See how Adapt can take action on your behalf This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.