Friday, February 10, 2006

Can we have properly defined scope before the detailed project plan?

Can we have properly defined scope before the detailed project plan? I guess, this is the "best case" secario if you are the project manager who is to execute the work defined in the scope statement.

But, before a properly defined scope can be obtained, there are a lot of planning and coordination work to be performed by the team. There are many rounds of follow up to be done. There are many parties to be interviewed and met to obtain all necessary important information. So, the objective of the project plan is to include that portion of work as well.

The scope statement (some times it is known as System Requirement Specification for software project) will be a "living" document. This means that, the team will continuously update the scope document to ensure the latest and most accurate condition of the requirements are documented. When there is additional scope after further assumption clarification and exploration, this new requirement will be submitted for approval for inclusion as Change Request. When we say new requirement it means that it is totally outside the original intention of the requirement. For example, originally the scope statements specifies that there are two units of servers are required for hosting the database. After the project is started, some external government authority dictates that for all transaction audit, the system needs additional two units of servers to store archive-data. In this case, the two additional servers are considered as new requirement.

In another scenario, in the original scope statement, if it reads "The system shall be able to keep the sales transaction for 3 months on the production server." After further study, the team finds out that the "sales transaction" means sale enquiry record, quotation record, sales order record and payment record. But, the original team assumes the sales record is equivalent to sales order record only. In this case, the problems arise: For these four types of records, for three months, the production server storage capacity is not sufficient to store these records. So, additonal storage media or shorter duration of the record keeping needs to be decided. In this scenario, it is "wrong assumption or intepretation" of scope definition. It is not an additional requirement. It is an under-estimate of scope complexity. In this case, it is the vendor responsibility to "find out" what does "sales transaction" means, what to do after the "3 months", and where the "production server" is.

Then the next question will be, how can the vendor find out this requirement in a very short duration during tendering (or proposal) stage? Usually, the detailed study will be performed only after the contract is awarded to the vendor. But, the tolerance level for scope difference should be defined in the project flexibility. Some times, the buffer for the scope is built in to the quotation or proposal. It is part of the risk assessment. The buffer for wrong estimate is crucial especially for fixed price contract.

So, the conclusion is ... it is very rare chance you will have a properly define scope before you start developing your detailed project plan :-)

2 comments:

Anonymous said...

You first need to differentiate between a Turnkey project from a configurable application system. The approach to scope def & user requirement gathering are quite different. So deferentiation will reduce your communication challenges.

In Turnkey projects we use Logical spec heavility, data flow of critical info, screen and report layout, sometimes with prepared data to illustarte a point. This improves our undertstanding of users SME needs more precisely... it leaves less room for misintepretaion.

In configurable applications, use of user questionaire will greatly improvethe requirement capture. Use of SME in the indutry will greatly enhance effectiveness and productivity, reduce risk and contentions.

Anonymous said...

I have been following a site now for almost 2 years and I have found it to be both reliable and profitable. They post daily and their stock trades have been beating
the indexes easily.

Take a look at Wallstreetwinnersonline.com

RickJ

Converting a Physical Linux to Virtual

Hmm ... I have done a lot of work on my Linux Lubuntu 15.10 with PHP and PostgreSQL and a few other things ... it is quite time-consuming to...