Using
Top-Down Estimating
A top-down estimate allows a project manager to take a very
similar project’s budget, work some financial math magic, and arrive at a
reasonable budget for the current project. Top-down budgets are often used by
organizations that complete IT projects for other companies. Consider IT
integrators who install servers, network cable, and network equipment. They’ll
have similar projects they can refer to when predicting the cost of current
projects.
Within an organization, IT project managers also have projects
that are similar to other projects they’ve completed in the past. Consider a
project to roll out a new operating system using a disk-imaging server. If the
project manager has rolled out other operating systems in the past using the
disk-imaging server, he’ll have a pretty good idea of how the current project
will go. This historical information on proven, completed applications allows
the project manager to save the time of doing a bottom-up estimate; he can work
from prior successful projects
The problem with top-down estimates in the IT world, however, is
that most IT projects have never been done before. Specifically, because IT
changes so quickly and each environment is generally customized, top-down
estimates are not as reliable or useable as bottom-up estimates.
Using Analogous
Estimating
If you find that you’re launching projects that are similar
to past accomplishments, analogous estimating may be your best bet. Analogous
estimating relies on historical information to predict the cost of the current
project. It is a type of top-down estimating. The process of analogous
estimating takes the actual cost of a historical project as a basis for the
current project. The cost of the historical project is applied to the cost of
the current project with respect to the scope of the current project, its size,
and other known variables.
This estimating approach takes less time to complete than other
estimating models do, but is also less accurate. This top-down approach is good
for fast estimates to get a general idea of what the project may cost.
Here’s an example of analogous estimating: You completed the
design and installation of an application for the sales department to track
incoming phone calls from clients. Your IT help desk now wants you to create an
application to track phone calls from internal users. The project deliverables
are technically different, but both have fundamental characteristics that can
guide you to create a reasonably reliable project cost estimate.
Using Parametric
Modeling
Another approach to top-down estimating is Parametric
Modeling. Parametric Modeling uses a mathematical model based on known
parameters to predict the cost of a project. The parameters in the model can
vary based on the type of work being completed. A parameter can be cost per
cubic yard, cost per unit, and so on. A complex parameter can be cost per unit
with adjustment factors based on the conditions of the project. Further, the
adjustment factors may have additional factors depending on additional
conditions.
For example, if you’re managing an application development
project, you may create a cost estimate based on the number of years of
experience the application developer has with a given software language. Bob may
have eight years of experience while Sam only has two years of experience. Sam
doesn’t cost as much as Bob because he’s considered less experienced than Bob.
Sam can still get the work done; it just may take him slightly longer than if
Bob did the work.
When you think of parametric modeling, a parameter is
generally used: cost per unit installed, cost per machine delivered, and so on.
This approach doesn’t always lend itself to IT projects because of the variables
within the technology. Consider function point analysis—lines of code are not
always reflective of the productivity, the number of servers, or even the number
of programmers assigned to an activity.