Wave Template

System Innovations

--  Integration modeling, data warehousing, data strategy --


Data Strategy

Managing Data
Why Data Management?
How data problems grow
Assert Data Principles
What is the cost of not treating Data as a Corporate Asset?
How to build a world-class data organization & an adaptive data environment
Getting Started (What to do?)
For example: Best Practices for Data Access
For example: Principles for On-Line Transaction Processing Systems (OLTP)
Getting Started (Who is needed?)
Roles Business People Play
Roles Technical People Play
What does it all mean?
When should I begin - treating data as a corporate asset?

The goal for information systems in today's technical environment is to treat data as a corporate asset by managing it like any other resource (such as property, cash, equipment and accounts receivable)

Managing Data Involves:

  • Capturing (validation, cleansing, security)

  • Storing (structure, rules of use, efficiencies)

  • Tracking (source, quality, timing, meaning)

  • Maintaining (updates, additions, deletions)

  • Retiring (archival, purging, etc.)

    data, so that you can:

  • Share data across the organization

  • Utilize data as a tool to support the business and function of the organization

  • Report data accurately to multiple user groups

Why Data Management?
Why do companies need to develop data management discipline?  Because of data problems,
such as:

  • Low quality data (inconsistent validation, wrong data in wrong fields)

  • Mismanaged data (inconsistent backup & recovery, faulty replication)

  • Inaccurate data definitions and inconsistent calculations

  • Inadequate or incomplete access to data

    (For example: Department A calculates daily service charges whereas Department B calculates monthly – this makes it hard to reconcile the two)

    Data problems can prevent a company from effectively managing its other assets (such as property, cash, equipment and accounts
    receivable)

How data problems grow:

Data problems start innocently enough.  A Customer Service Rep enters a customer phone number incorrectly.  The system picks up and applies that incorrect customer phone number due to inadequate input edits.  Various programs are executed to copy and share that customer phone number with other applications, thus proliferating the problem through multiple systems.  Finally, reports are generated which show the incorrectly entered, captured and shared data as if it were accurate.  Click below to see an illustration of how data problems grow. 

 To avoid data problems, companies must ...

Assert Data Principles (such as the following):

  • Customers demand privacy

  • Data must be accurate, consistent & predictable

  • Data must be available on a timely basis

  • Data must be secure

  • Data must be maintainable

  • Data must be recoverable

    All of the above - REQUIRES data management

What is the cost of not treating Data as a Corporate Asset?

  • Incomplete data derived separately from multiple customer channels

  • Data queries yield unpredictable results

  • Unable to reconcile information on reports generated from different sources

  • Unable to go to a single source to get answers

How to build a world-class data organization & an adaptive data environment

  • By providing rules for managing, organizing and storing data so that it is:

    Easy to access, Clearly defined, Properly managed, Secure,
    Integrated and Accurate

  • By adopting a pragmatic approach:

    Not a quick fix – but a lasting improvement

    Implemented through projects –project by project

    Takes effort in the beginning – but gets easier as you go along

Getting Started (What to do?)

First - establish rules

  • Principles & best practices

  • Process (defined terms, agreed-upon structures)

  • Engagement model

  • Architectural approach

And then… Follow the rules!


For example: Best Practices for Data Access

  1. Establish a data infrastructure that can accommodate rapid changes in data models based on changes in business requirements (i.e. use data access layers)

  2. Centralize data that needs to be shared and current

  3. Design databases to be modular, business driven and aligned with application services, not monolithic

For example: Principles for On-Line Transaction Processing Systems (OLTP)

  1. Protect data through data access rules

  2. Validate data at every practical level to ensure data quality and avoid unnecessary network traffic

  3. Minimize the replication of data within operational applications by replicating only stable data when necessary and based on business requirements.

  4. Normalize the data model for OLTP Systems

Getting Started (Who is needed?)

Achieving the data management goal is a collaborative effort requiring business and technology participation.

Roles Business People Play

Own the data and help establish the RULES

  • Establish data definitions & data refresh schedule

  • Decide where data should come from:

  • Internally – System of Record

  • Externally – Designated Source

  • Define data quality & integrity requirements

Roles Technical People Play (Not new people – what's  needed is a new philosophy)

  • Data Architecture
    Define strategy, architecture, models (RULES)

  • Application Development
    Apply data architecture rules

  • Infrastructure
    Provide data environment, hardware, support

What does it all mean?

Data Strategy changes the way your IT shop does business today:

  • Project Management: Projects add data-related steps in their project plans

  • Methodology: Application development methodology is adjusted to reflect data-related activities & deliverables

  • Engagement Model: Application development teams engage data management team at key points in the life cycle of a project

Data Strategy changes the way the data owner does business today:

Data owners proactively define the RULES

  • Data definitions & data refresh schedule

  • Data sources: internal & external

  • Data quality & integrity requirements

Through the necessary organizational supports

Data Strategy requires incurring certain costs:

Tools – for data cleansing, profiling and analysis
Infrastructure - to create a fully functional data environment (secure, recoverable, etc.)

Remember…
It takes effort in the beginning – but gets easier
as you go along

Data Strategy requires a change of mindset

From “my data” ...To “shared data”


How do they differ?

  • “My data” A single application defines and uses the required data without reference to other programs, apps, teams, needs, etc.

  • “Shared data” Any application that requires shared or common data utilizes enterprise definitions, coordinated across multiple business uses & systems

When should I begin - treating data as a corporate asset?

TODAY!

(For more information, or to get started on your data management strategy, call Laura Brown at 770-953-0534 today.)

Add a comment

Top of Page

[Company Profile] [Services] [Home] [Books & Articles] [Laura's Bio]

Copyright 1998-2005.  Laura Brown, LBPI, Inc. (DBA: System Innovations)
Last Updated: August 23, 2005