Specifying Dimensions and Measures

In this section:

Reference:

PMF is primarily designed as a way to measure how organizations are doing against their stated strategy. Strategy means many things in the real world, but when it applies to creating scorecards, it specifically refers to the definition of objectives and the description of how they are linked together.

You can determine how an organization is doing against an objective by setting targets for measures and then grading those measures.

As an administrator, you work closely with scorecard authors to set up the scorecard logic and load measures as defined and approved by the strategy committee in your enterprise.

You can always change anything you set up or load into PMF at a later date. Business processes normally change over time, so you will need to change PMF to reflect external changes.

The sections in this topic describe a methodology to ensure that dimensions and measures support the objectives.

PMF Objectives

When you create objectives, remember to make them SMART:

S = Specific

M = Measurable

A = Achievable

R = Results Oriented

T = Time Based

The following table lists and describes in greater detail the principles to follow when creating objectives in PMF.

Principle

Description

S = Specific

Objective must be specific and cannot be vague. An example of a bad objective would be Do Better, whereas a good objective is Improve Sales. However, you need to define what Sales means.

M = Measurable

Objective must be measurable. If your objective is to improve sales, and there are multiple sales components, such as Sales-Dollars and Sales-Margins, then this objective can be split into two objectives, such as Improve Sales Dollars and Improve Sales Margins.

A = Achievable

Objective must be controllable. An example of a bad objective is Reduce Loading Dock Time when there is no capture of such time. There must always be data to support the objective. On a personnel performance review, a bad objective would ask the staff to achieve a goal that is not under their control.

R = Results Oriented

Objective must indicate actions to achieve a goal. In the sales example, consider Improve Sales Margins for Q2 2006 over Q2 2005. This clearly states that there are results that are measurable at the end of the two quarters for comparison.

T = Time Based

Objective must be time-based. In the sales example, consider Improve Sales Margins for Q2 2006 over Q2 2005 by end of Q2.

By following the SMART principle when working through the statement of an objective, there will be clear objectives. If any of the components are not clearly formalized in the objective, then finding the measures will not be clear.

PMF Measures

Measures (or metrics) are the data component of a scorecard. Each measure should clearly belong to at least one objective and you should be able to answer the following questions for each measure:

PMF Dimensions

How to:

PMF supports 16 user-configurable dimensions that can go to 16 levels deep, so you can describe complex multi-dimensional structures. There is only one default dimension in PMF, the Time dimension, and multiple Time dimensions are supported.

The following steps are helpful for analyzing your dimensions:

At the end of dimension gathering, an agreement must be reached on the types of dimensions, their levels, and the contents at each level. No loading of dimensions should occur prior to the sign-off of the agreement.

Procedure: How to Create a New Dimension

The following procedure outlines the steps required when creating a new dimension. For information about editing an existing dimension, see How to Design a Simple Dimension Load.

  1. In the Manage page, click the Dimensions panel button.
  2. Click New.

    The New Dimension panel opens.

    Note: You can also click New in the Edit Dimension panel to create a new dimension.

  3. Enter a name for the dimension in the field box.
  4. From the drop-down menu, select one of the following options:
    • Harvested from Data. Choose this option to immediately start creating a structure from a preexisting data source.
    • Collected from Users. Choose this option if you want to create a structure yourself.
  5. Do one of the following:
    • If you chose Harvested from Data, enter a Level Name and set its properties. As you enter information for each line, a new one opens, as shown in the following image.
      New Dimension panel
    • If you chose Entered by Users, start by creating a structure in the Hierarchy & Data tab by entering Level names, as shown in the following image.
      New Dimension panel

      Click Save.

      Click the Edit Dimension Values tab to set dimension properties for the Levels that were created.

  6. Click Save.

Procedure: How to Edit a Dimension

  1. In the Manage page, click the Dimensions panel button.
  2. Click the dimension you want to edit in the tree.

    The Edit Dimension panel opens.

  3. Using the drop-down lists for each dimension, make your desired changes and click Save.

    Note: To change the order of the dimension levels, place the mouse over the dimension you wish to move and use the Arrow icon icon to drag the dimension up or down.

Procedure: How to Preview a Dimension Load

  1. In the Manage page, click the Dimensions panel button.
  2. Click the dimension you want to edit in the tree.

    The Edit Dimension panel opens.

  3. Click the Preview tab. The Dimension Load information opens, as shown in the following image.
    Edit Dimension panel

    To refresh the Preview contents, click the Refresh icon that is shown on the Preview tab.

Procedure: How to Rename a Dimension

  1. In the Manage page, click the Dimensions panel button.
  2. Click the dimension you want to rename in the tree.

    The Edit Dimension panel opens.

  3. Click the dimension name at the top of the panel tab. The field becomes editable.
  4. Enter the new dimension name and click Save. The dimension is automatically renamed throughout PMF.

Procedure: How to Delete a Dimension

  1. In the Manage page, click the Dimensions panel button.
  2. Click the dimension you want to delete in the tree.

    The Edit Dimension panel opens.

  3. Click Delete.

    A delete confirmation dialog box opens.

  4. Click OK.

When a dimension is deleted, all measures that were linked to it have their dimensional linkage deleted. The measure records are automatically re-summarized, and any orphaned extra records are automatically cleaned up.

Note: Re-summarization is a one-time change and cannot be reversed. Once you perform a re-summarization, measure values are now permanently changed. It is strongly recommended that you take a snapshot of your PMF data mart before making this change.

Reference: Dimensions Load History

PMF keeps track of each load that is executed for each dimension in the system, regardless of whether you loaded it manually or the load was called by the scheduler. This data is stored in a special logging section of the PMF data mart.

The History tab on each dimension displays the history of all loads that have been logged.

The history of the dimension shows:

  • The dates that the loads ran.
  • The count of rows that were retrieved, inserted, updated, and deleted.
  • The count of total mismatches that occurred between the source data and the PMF metrics mart. Mismatches are source data rows that do not match to any existing keys for one or more dimensions.
  • The count of gaps in data continuity, which indicate the sparsity of the data. This does not mean there are errors, but, if paired with mismatches, can help you debug any unexpected data discontinuities.
  • Any messages returned from the load system. If there is an error, the exact error is displayed in the information shown in this tab.

WebFOCUS