Discussion Locked

Methodology for Implementing Packages

0 helpful
5 replies
Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.

First my apologies for what may be a misleading Title (I'm limited to 40 characters!!) Now to my question.
I have been asked to research the following:
1) A project methodology adapted specifically for the implementation of "off-the-shelf" application solutions. I have access to several methodologies for solutions developed "in-house" but have not had much success finding one that's tailored to the unique requirements of package implementation.
2) Related to my first question, can anyone share some experiences on the typical duration of the unique tasks involved in package implementation?
3) I'd appreciate any wisdom you can impart on projects of this type - risks, pitfalls, critical success factors.

Thank you. I look forward to some enlightenment from all you gurus out there.

5 replies

by jcharris

A packaged solution will inherently help define your project, simply because of the predefined constraints of the package. I apply the "who/what/why/when/where/how" questions to the project that I was introduced to in journalism.

Who - List the clients / list the support people / list the installation people.
You will probably want to identify a pilot group up front who are willing to spend the extra time to give any feedback required prior to the rollout. Successfull rollouts require a champion, and you may find this person during the pilot phase of the project. Often the project manager becomes the champion.

What - Obviously the package, but there are also Server requirments, client hard disk space required, minimum platform requirements, infrastructure issues, backup requirements, training materials, required modifications,

Why - If you are going with a packaged solution, you will have a reason for doing this (usually cost effective solution) and its good to have a formal sheet on the reasoning behind 1) using packaged software and 2) how it will be implemented (overview).

When - Here you get to the PM guts and glory. Get the input from the concerned parties and leave plenty of slack in your schedule. Now double the slack again!

Where - If this is a localized or departmental rollout, this will be a no brainer, but if you are doing a multisite rollout, your 'where' will need to be thought out very carefully.

How - How will this impact current practices for (refer to Who). Get this straight early on. You don't want to find out that a package you are replacing with an off-the-shelf solution lacks a critical function that is required for month end closing.

-------
(continues on next answer)

by jcharris

(continued from Answer 1)

Duration of a project is dependent on the complexity and size of the rollout.
A departmental rollout involving an office suite and 20 people might be a weekend job with two days on-site support and seven days follow-up phone support.
A company wide switch from Lotus 1-2-3 to MS Excel involving 3000 seats will, of course, be a much different scenerio. You will need to determine what critical functions will be effected by the change, how to support script changes from 123 to VB for Apps, pre-rollout training, post-rollout support, departmental freeze periods, and may be a six week to one year process, depending on available resources and budget.

-------

The generality of the question makes the risks/pitfalls/critical success factors question difficult to answer.

Risks: The biggest risk will be a loss in functionality. Due diligence must be followed to determine what functions are critical to your user communities.

Pitfalls: Make sure there is a recovery plan. If you strand users with a new system and an unforseen glitch prevents a critical function from working. There has to be a failsafe to go back to the previous working solution. Remember, a new app might cause unexpected side effects on other applications.

Critical Success Factors: Buy-in. If the end user community doesn't buy in to a new solution, it's not going to go well, no matter how flawlessly the execution goes technically.

-------

Well, that's my 2 bitson your question. I haven't seen (or looked for) a formal methodology for "off-the-shelf" packages. I think the subject is too broad, due to the wide variations in implementations both in size and subject. Often you will find some strong guidencein the software documentation which will help you define the scope of your project.

Best of luck,
Jim Harris

This discussion is closed to new replies.