Tip

Ten steps in launching a data mining application

Alex Berson, Stephen Smith and Kurt Thearling

Ten steps in launching a data mining application
Alex Berson, Stephen Smith and Kurt Thearling

Data mining and analysis is more about ROI and cost savings than the technical aspects of data mining. The technology is important, but having a good plan seems more so, according to Berson, Smith and Thearling's book

    Requires Free Membership to View

Building Data Mining Applications for CRM (McGraw-Hill).


We have defined ten important steps that you should follow when creating and launching your data mining application. They are as follows:

  1. Define the problem
    Data mining can be used to either solve new problems or streamline manual analytical processes.
    • Find something that matters
    • Define the deliverables
    • Pick something well-defined and small
    • Understand the existing CRM process
  2. Define the user
    • Know the user
    • Understand the complexity or simplicity of features that the user requires
    • Build a profile of each user
    • Use a program to educate your future user and elicit needs and desires
  3. Define the data
    • Locate the data dictionary
    • Locate the data librarians
    • Define the metrics
  4. Now, really define the data (cleansing, organizing, data dictionaries)
    • Assess levels of data integrity
    • Validate data sources
  5. Scope the project
    • Contain scope creep through a launch document
    • Scope data cleansing
    • Scope data movement, modeling and storage
    • Scope data mining
    • Scope experimental design and measurement
  6. Trial
    • Don't wait too long
    • Start small, but go end to end
  7. Quality assurance
    • Make quality assurance a process
    • Validate and communicate model results
  8. Education

     

  9. Launch
    • Select your initial users
    • Keep things under wraps until all results are in
    • Help users interpret the results
  10. Continuation

Notice that none of the major ten steps talks about actually mining the data. The reason for this is that the technical areas--such as creating the database to store the data for the application, and the installation and application data mining--are fairly routine tasks if these other ten steps are followed. Users rarely get hung up on the execution of the data mining. They worry much more often about what the results of the data mining will be and how the data mining application will move from a one-off special project to part of the critical path for some important business process. For that reason, even technical steps such as quality assurance will focus here on how to perform and present the quality assurance so that it makes sense to the end user and they feel comfortable in sing the results of the data mining application.


Click here to learn more about Building Data Mining Applications for CRM.

Did you find this tip useful? Email and let us know.


This was first published in July 2001

There are Comments. Add yours.

 
TIP: Want to include a code block in your comment? Use <pre> or <code> tags around the desired text. Ex: <code>insert code</code>

REGISTER or login:

Forgot Password?
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy
Sort by: OldestNewest

Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to:

Disclaimer: Our Tips Exchange is a forum for you to share technical advice and expertise with your peers and to learn from other enterprise IT professionals. TechTarget provides the infrastructure to facilitate this sharing of information. However, we cannot guarantee the accuracy or validity of the material submitted. You agree that your use of the Ask The Expert services and your reliance on any questions, answers, information or other materials received through this Web site is at your own risk.