My first Publication Agile-Data-Warehouse-Design-eBook | Page 79

58
Chapter 2
Summary
Business events represent the measureable business activity that give rise to DW/BI data
requirements. BEAM ✲ uses data storytelling techniques for discovering business events by
modelstorming with business stakeholders.
BEAM ✲ defines three event story types: discrete, evolving and recurring. They match the three
fact table types: transaction fact tables, accumulating snapshots and periodic snapshots.
Each BEAM ✲ event consists of a main clause, containing a subject verb object, followed by
prepositional phrases containing prepositions and detail nouns. Each subject, object and detail
is one of the 7Ws; i.e., a noun that potentially belongs in a dimensional database design.
BEAM ✲ modelers use the 7Ws to discover a business event and document its type, granularity,
dimensionality and measures — everything needed to design a fact table.
BEAM ✲ modelers avoid abstract data models. They “model by example”: ask stakeholders to
describe their BI data requirements by telling data stories. BEAM ✲ modelers document these
requirements using example data tables.
Event stories are example data stories for business events.
Event stories are sentences made up of main clause and preposition-detail examples.
Event stories succinctly describe event details and clarify their meaning by providing examples
of each of the five themes: typical, different, missing, repeat and group.
Typical and different themed stories explore data ranges and exceptions.
Typical and repeat stories describe event uniqueness (granularity).
Missing stories help BEAM ✲ modelers to discover mandatory details and document how BI
applications should display missing values.
Group stories uncover event complexities including mixed business models and multi-valued
relationships.
BEAM ✲ short codes are used to document mandatory details, granular details and story type.
Other elements of the BEAM ✲ notation document W-type, units of measure and completed
event models.