Storytelling Session

(Activity) for Tier: Data Analytics

PURPOSE

To provide insights into the data using plain-English stories. To guide data scientists in understanding the data in a way that anyone can understand.

WHEN

After Visualization Design

PARTICIPATING ROLES

INPUTS

  • Premises and Conclusion(s) in the form of Visuals (e.g. Dashboard, Presentation, Storyboards, or GUI).

Todo

need to be define work products for all inputs.

ENTRY CRITERIA

  • Describe any required pre-conditions that must exist before beginning this activity.

SUB-ACTIVITIES

  1. Data Scientists present a story about what they learned during the sprint. The story should be compelling, taking the audience on a journey. This narrative is key to conveying insights, with the data and visuals being important proof points.
  2. Visuals are presented if they exist. The audience understands metrics, trends and patterns better with visuals such as graphs, charts, GUI, etc.

OUTPUTS

  • Updated Visuals or Python Notebook

Todo

need to define work products for all outputs.

EXIT CRITERIA

  • Story has been shared and discussed.
  • Questions have been discussed and recorded.

SEE ALSO

../../_images/BubbleChart.png ../../_images/Choropleth(FilledMap).png ../../_images/sunburst.png ../../_images/TreeMap.png

Process Guidance Version: 10.4