Agile DW/BI
Addressing the Hard Problems
Scott W. Ambler
Consulting Methodologist
Ambysoft Inc.
© Ambysoft Inc. AgileData.org AgileModeling.com 1
Scott Ambler
© Ambysoft Inc. AgileData.org AgileModeling.com 2
• scott@scottambler.com
• Twitter: @scottwambler
• linkedin.com/in/sambler/
Consulting Methodologist
Ambysoft.com
Thought Leader
AgileData.org
Thought Leader
AgileModeling.com
Agenda
• Your assignment
• Agile Data ways of thinking (WoT)
• What is agile DW/BI?
• What are the “hard problems”?
• How to address each problem
• Parting thoughts
• Questions and answers
© Ambysoft Inc. AgileData.org AgileModeling.com 3
Identify the questions you want
answers for over the next two
days
© Ambysoft Inc. AgileData.org AgileModeling.com 4
Your Assignment
Agile Data Ways of Thinking (WoT)
© Ambysoft Inc. AgileData.org AgileModeling.com 5
Agile Data Ways of Thinking (WoT)
AgileData.org/essays/philosophies.html
Look Beyond Data
Collaborate Closely
Be Quality Infected
Embrace Evolution
Be Enterprise Aware
Fit-For-Purpose
Agile Enterprise Groups
© Ambysoft Inc. AgileData.org AgileModeling.com 6
Agile Data Warehousing (DW)
and Business Intelligence (BI)
© Ambysoft Inc. AgileData.org AgileModeling.com 7
Agile DW/BI is the act
of providing quality
information in a
collaborative and
evolutionary manner.
© Ambysoft Inc. AgileData.org AgileModeling.com 8
The
“Hard Problems”
in Agile DW/BI
© Ambysoft Inc. AgileData.org AgileModeling.com 9
The “Hard Problems”
1. Up-front data architecture work takes too long
2. We need to deliver value every sprint
3. We need to install required infrastructure
4. We don’t have enough agile data people
5. Sprints are too slow, we need to go faster
6. Data analytics often takes longer than a single sprint
7. My product owner doesn’t understand the data
8. We have too much data technical debt
9. Our stakeholders don’t want to work with us
© Ambysoft Inc. AgileData.org AgileModeling.com 10
Problem: Up-Front Architecture
© Ambysoft Inc. AgileData.org AgileModeling.com 11
© Ambysoft Inc. AgileData.org AgileModeling.com 12
DV2 Has Done a Lot of Thinking for You
Used with permission from Dan Linstedt and DataVaultAlliance Holdings, LLC - DataVaultAlliance.com
Key Model: Data Source Diagram
• Identify main data sources
• Goal is to understand the
physical landscape
• Support with key source
information:
• Owner
• Access strategy
• Main entity types
• Quality rating
• Refresh rate
© Ambysoft Inc. AgileData.org AgileModeling.com 13
NCC
1701
Cust
THX
1138
Lead
File
…
JB7
File
Social M
edia Data
…
Data Vault
Key Model: Conceptual Diagram
© Ambysoft Inc. AgileData.org AgileModeling.com 14
• Identify main business
terms (entity types) and
their relationships
• Goal is to understand the
conceptual landscape
• Support with source to
entity type mappings,
tying business terms to
business processes
Just Barely
Good Enough
(JBGE) Artifacts
© Ambysoft Inc. AgileData.org AgileModeling.com 15
Source: AgileModeling.com/essays/barelyGoodEnough.html
Risk
Complexity
Desire for “predictability”
More modeling Less modeling
Implementer experience & skill
Access to stakeholders
Ease of change
Uncertainty
Collaborative WoW
Problem: Deliver Value
Every Sprint
© Ambysoft Inc. AgileData.org AgileModeling.com 16
© Ambysoft Inc. AgileData.org AgileModeling.com 17
Each sprint a disciplined agile team produces a working solution
Functionality is added in “vertical slices”
Example slicing strategies for DW/BI:
• One new data element from a single data source
• One new data element from several sources
• A change to an existing report
• A new report
• A new reporting view
• A new data mart table
AgileData.org/essays/verticalSlicing.html
Vertical Slicing
© Ambysoft Inc. AgileData.org AgileModeling.com 18
Source: AgileData.org/essays/questionStories.html
Vertical Slices for DW/BI: Question Stories
A question story is a specialized user story specific to data-oriented requirements.
Examples:
• As a sales manager I want to know the level of sales by my team by the end of each day so that I know where we
stand.
• As an instructor I want to know the certification pass rate of my students so that I can update my seminar marketing
message.
• As a restaurant owner I want to know the common combinations of menu options being ordered so that I can
identify potential specials.
• As a city councillor I want to know the number of complaints about road quality so that I can determine where we
need to focus our repair efforts.
Formats:
As a [Role] I want to know [Something] by [Timeframe] because [Reason]
- OR -
As a [Role] I want to know [Something] because [Reason]
Problem: Installing the Required
Infrastructure
© Ambysoft Inc. AgileData.org AgileModeling.com 19
Adopt the Great DV2
Vendor Solutions
© Ambysoft Inc. AgileData.org AgileModeling.com 20
Embrace Evolution
• Recognize that:
• You don’t need everything right
away
• Your situation will change
• Evolutionary development
• Think about the future, but wait
to act
© Ambysoft Inc. AgileData.org AgileModeling.com 21
Sometimes The First Release
Is Bigger Than You’d Normally Prefer
© Ambysoft Inc. AgileData.org AgileModeling.com 22
Problem: Lack of
Agile Data People
© Ambysoft Inc. AgileData.org AgileModeling.com 23
Agile Data Roles
© Ambysoft Inc. AgileData.org AgileModeling.com 24
Source: AgileData.org/essays/roles.html
Generalizing Specialists
© Ambysoft Inc. AgileData.org AgileModeling.com 25
Generalist
Specialist
Generalizing
Specialist
Broad
knowledge
Deep
skills
Source: http://agilemodeling.com/essays/generalizingSpecialists.htm
Skill
1
Skill
2
Skill
4
Skill
N
Skill
3
Skill
5
Skill
6
…
Knowledgeable
Competent
Specialist
Also known as:
• T-skilled people
• Cross-functional people
Invest in Your People
• Training
• Coaching
• Collaborative, non-solo
work
© Ambysoft Inc. AgileData.org AgileModeling.com 26
Problem: Sprints Are Too Slow
© Ambysoft Inc. AgileData.org AgileModeling.com 27
Run the First Release as an Agile Project
© Ambysoft Inc. AgileData.org AgileModeling.com 28
Source: http://agiledata.org/essays/disciplinedAgileDW.html
Evolve via Lean Continuous Delivery
© Ambysoft Inc. AgileData.org AgileModeling.com 29
Source: http://agiledata.org/essays/disciplinedAgileDW.html
Also….
Negotiate stakeholder expectations
• Perhaps the real issue is
misalignment
Shorten your sprints
• Motivates you to squeeze waste out
of your way of working (WoW)
• Gets you closer to a continuous
delivery strategy
© Ambysoft Inc. AgileData.org AgileModeling.com 30
Problem: Data Analytics
Takes Longer Than a Sprint
© Ambysoft Inc. AgileData.org AgileModeling.com 31
Look-Ahead Data Analysis: Agile
© Ambysoft Inc. AgileData.org AgileModeling.com 32
Scenario: You want to implement three question stories in sprint #9
You need to:
• Have a definition of ready (DoR) indicating the amount of data analysis work required
• Guesstimate the amount of data analysis required for each one, and then perform the
analysis sufficiently before sprint #9
• Have sufficient capacity to perform look-ahead data analysis
• Interleave data analysis for other sprints into the work of the people performing it
Note: Staffing your team with specialists will exacerbates work scheduling challenges.
Source: AgileData.org/essays/lookAheadDataAnalysis.html
Look-Ahead Data Analysis: Continuous Delivery
© Ambysoft Inc. AgileData.org AgileModeling.com 33
Development – QS 9a
Development – QS 9b
Development – QS 9c
Scenario: You want to implement the same three question stories
• You are not constrained by organizing the work into sprints
• Development = Data analysis + other implementation work
• Work can be brought into the team as capacity permits
• Value is delivered when it is available
• Average cycle time to deliver stories is shorter
• Staffing your team with generalizing specialists may enable you to swarm around data
analysis work and shorten the development time of a question story
Source: AgileData.org/essays/lookAheadDataAnalysis.html
Problem: My Product Owner
Doesn’t Understand the Data
© Ambysoft Inc. AgileData.org AgileModeling.com 34
Product Owners Must Collaborate Closely
With Agile Data Professionals
© Ambysoft Inc. AgileData.org AgileModeling.com 35
Active Stakeholder Participation
© Ambysoft Inc. AgileData.org AgileModeling.com 36
Source: AgileModeling.com/essays/activeStakeholderParticipation.htm
• Adopt inclusive modeling tools and
techniques and enable stakeholders to
explore their own requirements
• Product Owners, like Business Analysts,
can miss or filter out important details
• Supports a strategy of self-serve data
• Requires stakeholders to make time
Also…
Train/coach the product owner
• Help them get the knowledge that they
need
Generalizing specialists
• It’s not just an issue with product owners,
everyone should have a better data
background
© Ambysoft Inc. AgileData.org AgileModeling.com 37
Problem: We Have Too Much
Data Technical Debt
© Ambysoft Inc. AgileData.org AgileModeling.com 38
Be Quality
Infected!
© Ambysoft Inc. AgileData.org AgileModeling.com 39
Source: AgileData.org/essays/dataTechnicalDebt.html
Data technical debt refers to quality
challenges associated with data sources.
Data technical debt impedes the ability of
your organization to leverage information
effectively for informed decision making,
increases operational costs, and impedes
your ability to react to changes in your
environment.
© Ambysoft Inc. AgileData.org AgileModeling.com 40
Source: AgileData.org/essays/databaseRefactoring.html
Customer
CustomerID <<PK>>
Fname
Customer
CustomerID <<PK>>
Fname
FirstName
SynchronizeFirstName()
Customer
CustomerID <<PK>>
FirstName
Original
Schema:
Interim
Schema:
Final
Schema:
Database Refactoring
A database refactoring is a simple change to a
database schema that improves its design while
retaining both its behavioral and informational
semantics
A database schema includes structural aspects such
as table and view definitions; functional aspects such
as stored procedures and triggers; and informational
aspects such as the data itself
Problem: Our Stakeholders
Don’t Want to Work With Us
© Ambysoft Inc. AgileData.org AgileModeling.com 41
Deliver Value Frequently
Frequent Value Delivery
• Value delivered from very early on
• Provides opportunities to gain
concrete feedback and to steer
• Flexible, enabling the team to focus
on regularly delivering most
valuable functionality
• Requires team to be skilled and
disciplined
© Ambysoft Inc. AgileData.org AgileModeling.com 42
€
“Big Release” Value Delivery
• Convenient for the IT team
• High-risk, likely leading to something
that doesn’t meet their actual needs
• Very likely to go late, be over budget,
and potentially cancelled before
delivery
€ € € € € € € € € € € € €
Also…
Question stories
• Focus on their needs, not yours
Active stakeholder participation
• Make it more enjoyable for them
© Ambysoft Inc. AgileData.org AgileModeling.com 43
Parting Thoughts
© Ambysoft Inc. AgileData.org AgileModeling.com 44
© Ambysoft Inc. AgileData.org AgileModeling.com 45
The increasing pace of change,
increasing complexity,
and increasing volume of data
demands nothing less than
complete data agility
What questions do you
want answers for over the
next two days?
© Ambysoft Inc. AgileData.org AgileModeling.com 46
Thank You!
© Ambysoft Inc. AgileData.org AgileModeling.com 47
• scott@scottambler.com
• Twitter: @scottwambler
• linkedin.com/in/sambler/
Consulting Methodologist
Ambysoft.com
Thought Leader
AgileData.org
Thought Leader
AgileModeling.com
Backup Slides
© Ambysoft Inc. AgileData.org AgileModeling.com 48
Critical Agile Data WoW
• Active stakeholder participation
• AgileModeling.com/essays/activeStakeholderParticipation.htm
• Look-ahead data analysis
• AgileData.org/essays/lookAheadDataAnalysis.html
• Question stories
• AgileData.org/essays/questionStories.html
© Ambysoft Inc. AgileData.org AgileModeling.com 49
Agile Enterprise Architecture
AgileData.org/essays/enterpriseArchitecture.html
• Collaborative
© Ambysoft Inc. AgileData.org AgileModeling.com 50
• Enterprise focused
• Architects also implement
• Evolutionary
• Consumable
The Agile Database Techniques Stack - AgileData.org/essays/techniquesStack.html
Vertical Slicing
Clean Architecture and Design
Agile Data Modeling
Database Refactoring
Database Regression Testing
Continuous Database Integration and
Deployment
Configuration Management
© Ambysoft Inc. AgileData.org AgileModeling.com 51
Organize value into small increments
Enabled by
Evolved via
Improves via
Verified via
Automated via
Pulls files from
High quality, evolvable, and flexible - Agile
Iterative, incremental, and collaborative
Safely improves quality of data sources
Verifies and enables safe evolution
Fundamental DevOps infrastructure
Manage data assets as assets

Agile Data Warehousing (DW)/Business Intelligence (BI): Addressing the Hard Problems

  • 1.
    Agile DW/BI Addressing theHard Problems Scott W. Ambler Consulting Methodologist Ambysoft Inc. © Ambysoft Inc. AgileData.org AgileModeling.com 1
  • 2.
    Scott Ambler © AmbysoftInc. AgileData.org AgileModeling.com 2 • scott@scottambler.com • Twitter: @scottwambler • linkedin.com/in/sambler/ Consulting Methodologist Ambysoft.com Thought Leader AgileData.org Thought Leader AgileModeling.com
  • 3.
    Agenda • Your assignment •Agile Data ways of thinking (WoT) • What is agile DW/BI? • What are the “hard problems”? • How to address each problem • Parting thoughts • Questions and answers © Ambysoft Inc. AgileData.org AgileModeling.com 3
  • 4.
    Identify the questionsyou want answers for over the next two days © Ambysoft Inc. AgileData.org AgileModeling.com 4 Your Assignment
  • 5.
    Agile Data Waysof Thinking (WoT) © Ambysoft Inc. AgileData.org AgileModeling.com 5
  • 6.
    Agile Data Waysof Thinking (WoT) AgileData.org/essays/philosophies.html Look Beyond Data Collaborate Closely Be Quality Infected Embrace Evolution Be Enterprise Aware Fit-For-Purpose Agile Enterprise Groups © Ambysoft Inc. AgileData.org AgileModeling.com 6
  • 7.
    Agile Data Warehousing(DW) and Business Intelligence (BI) © Ambysoft Inc. AgileData.org AgileModeling.com 7
  • 8.
    Agile DW/BI isthe act of providing quality information in a collaborative and evolutionary manner. © Ambysoft Inc. AgileData.org AgileModeling.com 8
  • 9.
    The “Hard Problems” in AgileDW/BI © Ambysoft Inc. AgileData.org AgileModeling.com 9
  • 10.
    The “Hard Problems” 1.Up-front data architecture work takes too long 2. We need to deliver value every sprint 3. We need to install required infrastructure 4. We don’t have enough agile data people 5. Sprints are too slow, we need to go faster 6. Data analytics often takes longer than a single sprint 7. My product owner doesn’t understand the data 8. We have too much data technical debt 9. Our stakeholders don’t want to work with us © Ambysoft Inc. AgileData.org AgileModeling.com 10
  • 11.
    Problem: Up-Front Architecture ©Ambysoft Inc. AgileData.org AgileModeling.com 11
  • 12.
    © Ambysoft Inc.AgileData.org AgileModeling.com 12 DV2 Has Done a Lot of Thinking for You Used with permission from Dan Linstedt and DataVaultAlliance Holdings, LLC - DataVaultAlliance.com
  • 13.
    Key Model: DataSource Diagram • Identify main data sources • Goal is to understand the physical landscape • Support with key source information: • Owner • Access strategy • Main entity types • Quality rating • Refresh rate © Ambysoft Inc. AgileData.org AgileModeling.com 13 NCC 1701 Cust THX 1138 Lead File … JB7 File Social M edia Data … Data Vault
  • 14.
    Key Model: ConceptualDiagram © Ambysoft Inc. AgileData.org AgileModeling.com 14 • Identify main business terms (entity types) and their relationships • Goal is to understand the conceptual landscape • Support with source to entity type mappings, tying business terms to business processes
  • 15.
    Just Barely Good Enough (JBGE)Artifacts © Ambysoft Inc. AgileData.org AgileModeling.com 15 Source: AgileModeling.com/essays/barelyGoodEnough.html Risk Complexity Desire for “predictability” More modeling Less modeling Implementer experience & skill Access to stakeholders Ease of change Uncertainty Collaborative WoW
  • 16.
    Problem: Deliver Value EverySprint © Ambysoft Inc. AgileData.org AgileModeling.com 16
  • 17.
    © Ambysoft Inc.AgileData.org AgileModeling.com 17 Each sprint a disciplined agile team produces a working solution Functionality is added in “vertical slices” Example slicing strategies for DW/BI: • One new data element from a single data source • One new data element from several sources • A change to an existing report • A new report • A new reporting view • A new data mart table AgileData.org/essays/verticalSlicing.html Vertical Slicing
  • 18.
    © Ambysoft Inc.AgileData.org AgileModeling.com 18 Source: AgileData.org/essays/questionStories.html Vertical Slices for DW/BI: Question Stories A question story is a specialized user story specific to data-oriented requirements. Examples: • As a sales manager I want to know the level of sales by my team by the end of each day so that I know where we stand. • As an instructor I want to know the certification pass rate of my students so that I can update my seminar marketing message. • As a restaurant owner I want to know the common combinations of menu options being ordered so that I can identify potential specials. • As a city councillor I want to know the number of complaints about road quality so that I can determine where we need to focus our repair efforts. Formats: As a [Role] I want to know [Something] by [Timeframe] because [Reason] - OR - As a [Role] I want to know [Something] because [Reason]
  • 19.
    Problem: Installing theRequired Infrastructure © Ambysoft Inc. AgileData.org AgileModeling.com 19
  • 20.
    Adopt the GreatDV2 Vendor Solutions © Ambysoft Inc. AgileData.org AgileModeling.com 20
  • 21.
    Embrace Evolution • Recognizethat: • You don’t need everything right away • Your situation will change • Evolutionary development • Think about the future, but wait to act © Ambysoft Inc. AgileData.org AgileModeling.com 21
  • 22.
    Sometimes The FirstRelease Is Bigger Than You’d Normally Prefer © Ambysoft Inc. AgileData.org AgileModeling.com 22
  • 23.
    Problem: Lack of AgileData People © Ambysoft Inc. AgileData.org AgileModeling.com 23
  • 24.
    Agile Data Roles ©Ambysoft Inc. AgileData.org AgileModeling.com 24 Source: AgileData.org/essays/roles.html
  • 25.
    Generalizing Specialists © AmbysoftInc. AgileData.org AgileModeling.com 25 Generalist Specialist Generalizing Specialist Broad knowledge Deep skills Source: http://agilemodeling.com/essays/generalizingSpecialists.htm Skill 1 Skill 2 Skill 4 Skill N Skill 3 Skill 5 Skill 6 … Knowledgeable Competent Specialist Also known as: • T-skilled people • Cross-functional people
  • 26.
    Invest in YourPeople • Training • Coaching • Collaborative, non-solo work © Ambysoft Inc. AgileData.org AgileModeling.com 26
  • 27.
    Problem: Sprints AreToo Slow © Ambysoft Inc. AgileData.org AgileModeling.com 27
  • 28.
    Run the FirstRelease as an Agile Project © Ambysoft Inc. AgileData.org AgileModeling.com 28 Source: http://agiledata.org/essays/disciplinedAgileDW.html
  • 29.
    Evolve via LeanContinuous Delivery © Ambysoft Inc. AgileData.org AgileModeling.com 29 Source: http://agiledata.org/essays/disciplinedAgileDW.html
  • 30.
    Also…. Negotiate stakeholder expectations •Perhaps the real issue is misalignment Shorten your sprints • Motivates you to squeeze waste out of your way of working (WoW) • Gets you closer to a continuous delivery strategy © Ambysoft Inc. AgileData.org AgileModeling.com 30
  • 31.
    Problem: Data Analytics TakesLonger Than a Sprint © Ambysoft Inc. AgileData.org AgileModeling.com 31
  • 32.
    Look-Ahead Data Analysis:Agile © Ambysoft Inc. AgileData.org AgileModeling.com 32 Scenario: You want to implement three question stories in sprint #9 You need to: • Have a definition of ready (DoR) indicating the amount of data analysis work required • Guesstimate the amount of data analysis required for each one, and then perform the analysis sufficiently before sprint #9 • Have sufficient capacity to perform look-ahead data analysis • Interleave data analysis for other sprints into the work of the people performing it Note: Staffing your team with specialists will exacerbates work scheduling challenges. Source: AgileData.org/essays/lookAheadDataAnalysis.html
  • 33.
    Look-Ahead Data Analysis:Continuous Delivery © Ambysoft Inc. AgileData.org AgileModeling.com 33 Development – QS 9a Development – QS 9b Development – QS 9c Scenario: You want to implement the same three question stories • You are not constrained by organizing the work into sprints • Development = Data analysis + other implementation work • Work can be brought into the team as capacity permits • Value is delivered when it is available • Average cycle time to deliver stories is shorter • Staffing your team with generalizing specialists may enable you to swarm around data analysis work and shorten the development time of a question story Source: AgileData.org/essays/lookAheadDataAnalysis.html
  • 34.
    Problem: My ProductOwner Doesn’t Understand the Data © Ambysoft Inc. AgileData.org AgileModeling.com 34
  • 35.
    Product Owners MustCollaborate Closely With Agile Data Professionals © Ambysoft Inc. AgileData.org AgileModeling.com 35
  • 36.
    Active Stakeholder Participation ©Ambysoft Inc. AgileData.org AgileModeling.com 36 Source: AgileModeling.com/essays/activeStakeholderParticipation.htm • Adopt inclusive modeling tools and techniques and enable stakeholders to explore their own requirements • Product Owners, like Business Analysts, can miss or filter out important details • Supports a strategy of self-serve data • Requires stakeholders to make time
  • 37.
    Also… Train/coach the productowner • Help them get the knowledge that they need Generalizing specialists • It’s not just an issue with product owners, everyone should have a better data background © Ambysoft Inc. AgileData.org AgileModeling.com 37
  • 38.
    Problem: We HaveToo Much Data Technical Debt © Ambysoft Inc. AgileData.org AgileModeling.com 38
  • 39.
    Be Quality Infected! © AmbysoftInc. AgileData.org AgileModeling.com 39 Source: AgileData.org/essays/dataTechnicalDebt.html Data technical debt refers to quality challenges associated with data sources. Data technical debt impedes the ability of your organization to leverage information effectively for informed decision making, increases operational costs, and impedes your ability to react to changes in your environment.
  • 40.
    © Ambysoft Inc.AgileData.org AgileModeling.com 40 Source: AgileData.org/essays/databaseRefactoring.html Customer CustomerID <<PK>> Fname Customer CustomerID <<PK>> Fname FirstName SynchronizeFirstName() Customer CustomerID <<PK>> FirstName Original Schema: Interim Schema: Final Schema: Database Refactoring A database refactoring is a simple change to a database schema that improves its design while retaining both its behavioral and informational semantics A database schema includes structural aspects such as table and view definitions; functional aspects such as stored procedures and triggers; and informational aspects such as the data itself
  • 41.
    Problem: Our Stakeholders Don’tWant to Work With Us © Ambysoft Inc. AgileData.org AgileModeling.com 41
  • 42.
    Deliver Value Frequently FrequentValue Delivery • Value delivered from very early on • Provides opportunities to gain concrete feedback and to steer • Flexible, enabling the team to focus on regularly delivering most valuable functionality • Requires team to be skilled and disciplined © Ambysoft Inc. AgileData.org AgileModeling.com 42 € “Big Release” Value Delivery • Convenient for the IT team • High-risk, likely leading to something that doesn’t meet their actual needs • Very likely to go late, be over budget, and potentially cancelled before delivery € € € € € € € € € € € € €
  • 43.
    Also… Question stories • Focuson their needs, not yours Active stakeholder participation • Make it more enjoyable for them © Ambysoft Inc. AgileData.org AgileModeling.com 43
  • 44.
    Parting Thoughts © AmbysoftInc. AgileData.org AgileModeling.com 44
  • 45.
    © Ambysoft Inc.AgileData.org AgileModeling.com 45 The increasing pace of change, increasing complexity, and increasing volume of data demands nothing less than complete data agility
  • 46.
    What questions doyou want answers for over the next two days? © Ambysoft Inc. AgileData.org AgileModeling.com 46
  • 47.
    Thank You! © AmbysoftInc. AgileData.org AgileModeling.com 47 • scott@scottambler.com • Twitter: @scottwambler • linkedin.com/in/sambler/ Consulting Methodologist Ambysoft.com Thought Leader AgileData.org Thought Leader AgileModeling.com
  • 48.
    Backup Slides © AmbysoftInc. AgileData.org AgileModeling.com 48
  • 49.
    Critical Agile DataWoW • Active stakeholder participation • AgileModeling.com/essays/activeStakeholderParticipation.htm • Look-ahead data analysis • AgileData.org/essays/lookAheadDataAnalysis.html • Question stories • AgileData.org/essays/questionStories.html © Ambysoft Inc. AgileData.org AgileModeling.com 49
  • 50.
    Agile Enterprise Architecture AgileData.org/essays/enterpriseArchitecture.html •Collaborative © Ambysoft Inc. AgileData.org AgileModeling.com 50 • Enterprise focused • Architects also implement • Evolutionary • Consumable
  • 51.
    The Agile DatabaseTechniques Stack - AgileData.org/essays/techniquesStack.html Vertical Slicing Clean Architecture and Design Agile Data Modeling Database Refactoring Database Regression Testing Continuous Database Integration and Deployment Configuration Management © Ambysoft Inc. AgileData.org AgileModeling.com 51 Organize value into small increments Enabled by Evolved via Improves via Verified via Automated via Pulls files from High quality, evolvable, and flexible - Agile Iterative, incremental, and collaborative Safely improves quality of data sources Verifies and enables safe evolution Fundamental DevOps infrastructure Manage data assets as assets