My first Publication Agile-Data-Warehouse-Design-eBook | Page 49

28
Chapter 2
Data Stories
Data stories are to
agile DW design as
user stories are to
agile software
development
Event stories use
the narrative of a
business event to
discover BI data
requirements
Data stories are comparable to user stories: agile software development's lean
requirements gathering technique. Both are written or told by business stake-
holders. While user stories concentrate on functional requirements and are written
on index cards, data stories concentrate on data requirements and are written on
whiteboards and spreadsheets.
Business events, because they represent activity (verbs), have strong narratives.
BEAM ✲ uses these event narratives to discover their details (nouns) by telling data
stories. BEAM ✲ events are the archetypes for many similar data stories. "Employee
drives company car on appointment date." is an event. "James Bond drives an
Aston Martin DB5 on the 17 th September 1964" is a data story. By following five
event story themes, event stories, a specific type of data story, succinctly clarify the
meaning of each event detail and help elicit additional details.
Story Types
Event stories are
discrete, evolving or
recurring depending
on how they
represent time
Figure 2-1
Story type timelines
BEAM ✲ classifies business events into three story types: discrete, evolving, and
recurring based on how their stories play out with respect to time. Figure 2-1 shows
example timelines for each type. Retail product purchases are examples of discrete
events that happen at a point in time. They are (largely) unconnected with one
another and occur unpredictably. Wholesale orders are evolving events that repre-
sent the irregular time spans it takes to fulfill orders. They too occur at unpredict-
able intervals. Interest charges are recurring events that represent the regular time
spans over which interest is accrued. They occur in step with one another at
predictable intervals.