Generate and explore visualizations with PowerMetrics AI

Have a conversation with your business data! Using a natural language interface, PowerMetrics AI interprets your questions to automatically create the visualizations you need to make effective decisions.

PowerMetrics AI helps you investigate your data and build visualizations. It can be used in combination with manual personalized configurations, or it can automate the entire experience – the choice is yours.

It all happens in Explorer, a space for free-form analysis of single and multiple metrics. Here, you can manually add visualizations, or prompt PowerMetrics AI to add them for you. Regardless of how you explore your data, you’re free to investigate and dig in without affecting the metric’s underlying data or its default display for others. When you discover something you want to share, save the visualization to a dashboard for easy access. Learn more about exploring your visualizations here.

Note:

  • PowerMetrics AI uses OpenAI and, as a result, is available to the countries listed here.
  • PowerMetrics AI is included in the 30-day premium trial for all new PowerMetrics customers. When the trial ends, you can continue to use this feature by purchasing the PowerMetrics AI Add-On for our Professional plan or by contacting our Success Team to learn more about our Enterprise plan.

This article includes:

Why PowerMetrics AI?

Working with PowerMetrics AI is a collaborative experience where business users interact with the AI to iteratively derive insights from their company’s data. Context is key. PowerMetrics AI only refers to data in your account so, while the AI may sometimes misinterpret your questions, the data used in its answers is always accurate and reliable.

Data security is assured:

  • No data from within your metrics is sent to any external service. All data processing, queries, and analysis is performed within the PowerMetrics platform.
  • We do not send customer data to OpenAI or other external services for model training purposes.
  • PowerMetrics AI only sends a minimal amount of metadata information, like metric names in prompts, to a large language model, for example OpenAI.

Generating and exploring visualizations with PowerMetrics AI

To generate and explore visualizations with PowerMetrics AI:

  1. Open Explorer by clicking Explorer button in the left navigation sidebar.
  2. Add a single metric or multiple metrics to Explorer:
    • Automatically, by typing a prompt and pressing Enter (or by clicking the visualize button button).
      Tip: When automatically adding metrics, you can visualize metrics that currently exist in your account – you can’t create new metrics.
      Note: The working example, as shown below, is dynamically based on the metrics in your account. Note also, that the prompt entry field is limited to 200 characters.
      Explorer UI
    • Manually, by clicking the add metrics to Explorer button button or by clicking the Add metrics link.
      Tip: This can be a great way to get started. With the manual option, you can select an existing metric from your list of metrics or create a new metric to explore. After adding a metric or metrics manually, you can use a combination of manual selections and AI prompts or rely solely on your prompts to the PowerMetrics AI to visualize your data.
  3. Continue entering prompts and, optionally, making manual choices (using the properties pane on the right) and watch as your visualization dynamically updates.
    Learn more about exploration options.
  4. Click the History button History button as you explore your metric visualization to see a list of the prompts you’ve entered. You can go back to any step, and see the visualization as it was displayed at that point, by clicking the entry in the History list.
    Note: The Explorer is designed for fluid, free-form investigation. As such, if you select a previous prompt in the History list and enter a new prompt, it will replace the entries that previously displayed after the item you selected. For example, if there are 5 entries in your history list and you click the 2nd entry and enter a new prompt, the new entry will replace the pre-existing 3rd, 4th, and 5th entries in the History list. It’s a convenient way to go back a few steps and take your exploration in a different direction.
  5. If you want to keep a record of your prompts, click Download history log. This downloads a text file with a list of each prompt and the time and date it was entered.
  6. To start over, click History button History > Reset history button Reset history or type “Reset” into the prompt entry field.
    Note: Resetting clears the prompt history only. To clear the visualization you’re exploring, click the Reset button in the top-right corner of the navigation bar. (See below.)
    reset Explorer button
  7. In Explorer, only one exploration can take place at a time. To save a visualization and its settings, add it to a dashboard by clicking the drop-down arrow in the Add to dashboard button. (See below.) Once the visualization has been added to a dashboard, you can easily share it with others.
    Add to dashboard
    Note: The history log information is not included when sharing a visualization.

Prompt tips and suggestions

The language you use when conversing with PowerMetrics AI is only limited by your imagination. Here are some ideas to get you started.

  • Want to empty Explorer and start fresh? Enter words like “reset”, “clear”, and “start over”.
  • Explorer supports multi-metric charts (like scatter, bubble, and combination charts) and can display up to 5 metrics in a single, exploratory visualization. To visualize multiple metrics, enter words like “show me new users and returning users”.
  • To change the chart type, enter, for example, “switch to a line chart”.
  • To see a visualization for a specific time period, enter the date range you’re looking for, like ”show me revenue for the last 20 days”.
  • If you want to see data for a specific dimension, use words like “break down” or “segment”, for example, “show me sales segmented by region”. Remember to use the exact spelling and case when entering dimension/member names.
  • You can show data with a trendline by entering something like “show moving average”.
  • You can visualize data as a ranked table by entering text like “show me reps with the most sales”.
  • PowerMetrics AI usually finds the metrics you request but it sometimes needs a little help. When you’re looking for a specific metric, you can do so using our auto-complete feature. Start by entering $. A list of up to 10 metrics will display. Optionally, enter characters to refine your choices. The list returns metrics with names that contain the characters you enter. When you see the metric you’re looking for, select it to add it to the Explorer window.

Note: For more ideas, click the Tips button. It displays beside the Visualize button, on the right-side of the prompt entry field, when the Explorer window is empty.

Working with PowerMetrics AI

PowerMetrics AI does its best to build the visualization you’re looking for. Here are a few tips to help you work together:

  • If you enter a prompt and see “No data for the selected date range”, open the associated metric to make sure it has data for the date range you requested.
  • If you prompt PowerMetrics AI to add a specific metric, for example, “Add wins metric”, and it adds more than one metric to Explorer, both with the same name:
  • You may have metrics with duplicate names. Fix this issue by opening your list of metrics and searching by metric name. Then, give the metric(s) unique names by selecting and editing them.
  • Sometimes duplicate metric names make sense, for example, you could have MRR metrics that include data from different services. In such cases, when entering the prompt, specify the service associated with the metric, for example, "Show me MRR from dbt vs MRR from Google Drive".
  • When entering metric names and keywords, aim for accuracy. Using correct spelling and capitalization is particularly important when requesting dimensional data.
  • When applying filters to a visualization, be explicit in your instructions by including the words “filter” and the “name of the dimension” you want to filter by. For example, to visualize a metric that displays Canadian sales only, it helps to enter “Filter sales by Canada”.
  • For data privacy reasons, we don’t pass all member values from all dimensions for your metrics to any external LLM. As a result, questions that ask for a member value but don’t include the metric and dimension for context may not return accurate results.
Have more questions? Submit a request