Tag
The right chart makes a finding obvious; the wrong one buries it. These prompts help you pick a chart type for your data and write the explanation that goes underneath it.
6 prompts
Reviews a chart — its type, intended message, underlying data, and any suspected issues — before it goes into a deck or report. - Checks: whether the chart type fits the data relationship, honest visual encoding, data-to-ink ratio - Tests: whether the key insight lands within 5 seconds and holds up in black and white or for colorblind readers - Output: concrete before/after fixes, not just a list of complaints
Plans a dashboard for a given audience and business area before it gets built in a BI tool, based on a metrics list, user goals, refresh frequency, and target tool. Specifies the layout, chart type per metric with reasoning, filter and drill-down controls, a color-coding scheme, alert thresholds, and mobile considerations, all built around whether a viewer can pass the 5-second glance test.
Specifies an interactive exploration tool for a dataset's dimensions and metrics, covering layout, filter and drill-down behavior between named levels, comparison views, hover details, transitions, and the default view a user sees before touching anything. Adds a tool-specific implementation plan and performance notes for the expected row count, plus working Plotly Dash or Streamlit prototype code when the target tool is Python-based.
Describe your DataFrame's columns and sample rows, list up to four charts you need, and specify a color palette, figure size, and where the charts will be viewed — slides, web, or print. The prompt returns complete, runnable matplotlib and seaborn code for all four charts at once, each with a descriptive title, labeled axes with units, a legend only when needed, and annotations calling out key data points, without unnecessary chartjunk.
Name the reporting cadence, the business area, and up to four report sections, and this prompt plans an entire recurring report's worth of visuals at once. For each section it recommends 2-3 charts with the exact source columns and encoding, writes a chart title that states the finding rather than just describing the data, and adds a fill-in-the-blank caption template, then generates a rerunnable Python script that builds every chart from one DataFrame and exports consistently styled PNGs each period.
List the filter dimensions users can apply, the key metrics shown, the drill-down hierarchy, the comparison periods needed, the database engine, and the relevant tables, and this prompt writes the parameterized SQL query layer behind an interactive dashboard. Each panel's query accepts filter parameters using {{parameter_name}} syntax, returns data shaped for its visualization, is written for sub-second interactive response, and handles edge cases like empty filters meaning "all data" or invalid date ranges.