- The Chat Interface (Includes Slack Chats)
- TextQL Playbooks
Ana Chat Best Practices
- 🤖 Treat Ana as you would a coworker: Have interactive conversations to clarify or refine requests, just as you would with a real analyst.
- 🧩 Break complex requests into smaller steps: Language models can get overwhelmed when given too many steps at once.
- ❓ Ask follow-up questions: This helps uncover deeper insights or correct misunderstandings along the way.
- 📈 Request charts and visualizations: These can be easy ways to understand data and insights, as well as verify Ana’s assumptions.
- ❗ Expect occasional misunderstandings or assumptions: Ana might interpret questions differently, so be prepared to rephrase or specify your request when needed.
- 🔄 If a chat gets too long or isn’t going well, start fresh: Ana has access to many tools and may stumble when handling too many things at once. Sometimes, the best solution is to start a new chat.
TextQL Playbooks Best Practices
- 📝 Write comprehensive prompts: TextQL playbooks are designed for deep asynchronous analysis, triggered on an event or schedule. Unlike Ana chats, which are designed to expect follow-up messages, TextQL Playbooks generally assume that you have provided a comprehensive description of the task (at a high level), so that they can continually work without the need to ask for clarifying questions. To that end, Playbooks tend to perform best with long prompts, that include as many steps, actions, and corner cases as the user is aware of. After a Playbook completes its action, you will have the option of continuing the playbook conversation with follow-up messages through the chat interface.
- 🕵️♀️ Preview your analysis before activating the playbook: Previewing playbook runs allows you to quickly iterate and refine your prompt, ensuring your playbook is able to reliably execute your desired task. Generally, we recommend running at least 3 previews before activating your playbook.
- 🎁 Attach any relevant datasets you’re aware of: If you believe that a Playbook will regularly use a particular dataset of interest, you should attach that dataset explicitly to the playbook at setup time. This helps reduce ambiguity and streamline the Playbook’s planning at runtime. If you aren’t aware of any such datasets, or you have low familiarity with your company’s data warehouse, you can safely disregard this step. TextQL Playbooks have the ability to search your data warehouse at runtime to identify the data of interest relevant to your query, it just might take longer to get to the result.
- 🧾 Direct Ana to relevant parts of a dashboard: If your playbook uses Tableau as a datasource, specify the sheet or sub‑figure relevant to your question or task. Call out any non‑obvious filters or business logic that affect your dashboard: for example, “The orders dashboard only visualizes orders jobs with Status = Active and Region = USA.” Remember, you see the dashboard visually, but the playbook also sees the underlying data tables used to generate the dashboard: If Ana’s results differ from what you see visually, your dashboard likely shows a restricted data view.
- 🖼️ Specify output format and include code snippets: For reporting use cases, where a Playbook is asked to perform the exact same analysis on a schedule, it’s useful to explicitly specify your desired output format so that plots and reports are consistent across playbook runs. Consider prototyping your playbook in a chat and refining output plots until they fully capture your desired report; then, copy the visualization code to a playbook prompt.
- 📅 Be mindful of dates and data updates: Users often want to ask “Tell me how we are doing today” or “Send me a report on today’s changes.” These types of analysis are supported and encouraged; however, certain datasets within your company’s data warehouse may update on a daily or hourly basis. To that end, in certain instances, complete results are only attainable by asking about previous days or time periods, for example “Tell me how we did yesterday, and compare this to previous days and weeks at the same time.”
Sample Playbook Prompt Template