The ability to generate timely and accurate change estimates to kick off a software project is most defintely an art. Experienced project managers will rely on their team of experts to provide them with input, then have to rely on how well they in turn, know their team in order to adjust each affected sub-system estimate with a “Murphy’s Law buffer” in order to come up with an overall project estimate. For example, Charlie may always over-estimate the size of a change, while Jill may always aim to please and consistenly under-estimate the amount of effort required to deliver a completed enhancement. Sound familiar?
With ever increasing demands and cost-contrained limited resources, those same developers are often smack dab in the middle of the high-stress tasks involved in completing the last release while being asked to look at the next potential change. Either that, or they are brand new employees or outsourced contractors hired specifically for the project who may not be very familiar with the existing code. As a result, they may not have the experience or spend the time necessary to explore all possible design alternatives so that the best possible option for the change is selected in the first place … something that is often discovered when deep in the middle of implementing the change itself!
In the end, using traditional tools and methodologies may come close, but unfortunately more often then not, “estimates” are just that … an educated best guess. Just not good enough.
How can we, as software project managers or stake holders that rely on custom software to run our business ensure we are:
- selecting the most cost effective change design?
- running the least risk of scope creep, missed target deadlines and cost overruns?
- producing estimates that are based on objective facts, rather then eager to please Joe-programmer’s “gut feel”?
I came across some interesting articles that repeat the same message and provide some ideas using traditional tools and techniques. However, even with all of the research and progress in software engineering over the decades, more than 85% of medium to large software enhancement projects miss target delivery dates and/or experience high cost overruns. In fact, over half of large software change projects experience cost over-runs that are almost double original estimates. One research study even showed that 75% of respondents believe their IT projects are doomed even before they start.
One of the keys to creating better estimates and delivering better quality code is to be able to objectively identify ALL affected code, accurately and quickly. This is critical given time constraints and pressure to do more with less with today’s constrained IT budgets. Automation tools that can create comprehensive change lists across all components of a software project deliver. They provide programmers and analysts with the tools necessary to explore and select the most cost-effective change design … as well as providing a complete scope of work “to do list” for the option selected. Reducing costly reworking and avoiding that dreaded Murphy’s law effect.
Find it EZ is just such a tool. Complete and automated change impact analysis across an entire sofware body of work ( core program modules, reports and throughout the database ). Find it EZ can reduce impact analysis time by up to 80%, and directly increase developer team productivity by 15 to 20%. Save time, do more, build better software today.
For some interesting related reading, check out:
- Software Metrics – Estimating Data
- ZDnet IT Project Failures Research
- Cortex IT Project Failure Rates
We are not alone, but we CAN do something to stand out from the crowd and rise to the top with a consistent record of delivering on time and budget. Find it EZ