Home Up Next

 

 

Enabling superior business decisions

Nine Steps to DW Success

 

 

 

Carletonís Data Warehousing
Nine Steps to Success

  1. Strategic Plan
  2. If your organisation has as one of its objectives something along the lines of "to provide electronic access to accurate and timely information and services by efficiently managing data as a vital and valued asset" then one of the strategies to achieve that objective will be to establish a data environment which enables the appropriate sharing of data at all levels within the organisation and sometimes externally.

  3. Business Requirements and Business Case
  4. Cost justification for using a Data Warehouse initiative to meet the specific strategic business initiatives. The business base is a collection of analyses covering organisational benefits, costs estimates, tasks, resources requirements, key success factors, risks, return on investment and assumptions for developing the data warehouse.

    or

    What do I want to know?
    What donít I know?
    Can business intelligence increase my capability and knowledge of my business?
    What will it cost?

    or

    Assess and define information dissemination and access needs with costing and benefit metrics.

    "There are painters who transform the sun into a yellow spot, but there are others who, with the help of their art and their intelligence, transform a yellow spot into the sun." - Pablo Picasso

  5. Technologies
  6. SMP v MPP
    UNIX v MVS v Windows
    Which hardware platform?
    Which RDBMS?
    Which ETL?
    Which metadata repository?
    Which front-end query tool?
    etc.

    or

    identify the information systems required to respond to the above business requirements.

  7. Architecture
  8. Operational data stores, data warehouses, data marts

  9. Metadata "The Glue that Binds the Data Warehouse"
  10. What is metadata?
    What metadata do we have?
    What additional metadata do we want?
    Who will use it, what for, and how?
    Can we import and export it?

  11. Dimensional modelling
  12. Dimensions are the structures through which the business users will access the data in the data warehouse. Although the primary sources of data for the data warehouse are the legacy operational systems, the logical design of the data warehouse is significantly different to that of an operational system.

  13. DW design optimisation. Issues to be considered include:

Audience
Power users
"Normal" users
Subject areas
Volume of historical data
Expected growth of the data warehouse
Summarised and non-summarised data
Source data
Refresh and update strategies
DW performance
Scalability

  1. Data propagation
  2. Operational data stores, data warehouses, data marts
    Data extraction
    Data cleansing
    Data integrity
    Data integration
    Data loading
    Propagation strategies
    Redundant data
    Security

  3. Maintenance and support
  4. Changed data capture
    Administration
    Usage
    Tuning
    Inclusion of additional subjects

  5. Measuring the benefits

There must be a business benefit in designing, building and deploying a data warehouse. A great deal of time, effort and money have gone into this endeavour so it is incumbent on the builders to ensure that the expected business benefits have been realised. It is equally incumbent on the business to ensure that the business benefits are measurable and quantifiable.

Each iteration of the Data Warehouse must yield a demonstrable business benefit 

or 

"Success consists of going from failure to failure without loss of enthusiasm." - Winston Churchill

 

Home ] Up ]