You have complete control over the measurements made by this system because you create
the questions using the forms provided. When a work item is completed, the metrics manager
selects the appropriate question list for that task. The Metrics Center allows a manager
to define lists of questions appropriate for development, documentation, systems
maintenance and any other work category. Individual questions can require a numerical response (how much, how many), a
multiple-choice response or a text response (comment or customer code, for example). Each
question can provide instructions
Metrics Center is an agent-driven subsystem that provides an automated method for
defining, acquiring and evaluating experience that can be put to immediate use in the
estimation and planning process for future work.
Defining Questions
The questions you choose to ask will depend on the kind of work you do. If you are doing classic IT systems development, then you may want users to specify how many function points were created in their work. On the other hand, if you do development with a high-level tool like Notes, you may prefer to ask how many forms, agents or views were produced, because these items are easier to count. For documentation work, your measurements might include pages, chapters or help screens.
In addition to asking people to define how much was produced, you may want to ask about the type of work completed: Did you write new chapters, or edit existing ones? Did the scope of your work change during the project? How much time did you spend learning to use new tools versus actually working on the task? Project Gateways Metrics Center makes all of this information working knowledge for the entire enterprise.
The measurement process begins whenever an assignment, phase or project is completed. The metrics manager is then notified and instructed to select a questionnaire from the library maintained in Metrics Center. Project Gateway sends the questionnaire to the appropriate participant or project manager for his or her input.
The results of the questionnaire are combined with repository data, including planned and actual hours, dates and other information, to produce a set of detailed measurement reports. These raw reports are automatically reviewed for quality, and those, which fall outside anticipated norms, are posted for manual inspection, correction or rejection.
Approved measurements are then used to construct an Integrated Knowledge Record for each completed project, phase or activity. These records store the information gathered from the Repository Center database, including planned and actual dates, effort, project names and information gathered directly from questionnaires, such as function points produced or chapters written.

Metrics Center provides two integrated functions that can be used to ask questions about the accumulated experience: query and estimating. The query function returns a set of records with particular characteristics. The estimating function returns an estimate based upon the available measurements for a particular question. Many selection and tabulation options are provided. These functions work from both Notes and Web clients.
A typical estimating activity might be to find out how long it has taken to produce 50 pages of system documentation. The result would show the number of prior cases and provide a range of estimates for 50 pages derived from the most productive, average and least productive experiences. You can then refine the selection criteria or examine the individual record to provide more insight.
Depending on the scale of your organization, Project Gateway will produce useful collections of metrics records within a few months. After a year, you will have well-documented data points for common types of work that can greatly improve your estimating and provide real insight for process improvement.
If you would like to know more, read the users manual chapter entitled "Using Metrics" which you can find on the support page.