Software implementation: transforming pain in gain!
When addressing software implementation problems the first issue that comes to my mind is related to “bugs”. This is natural and every programmer or system analyst knows that even a well planned project may carry out bugs that are only identified at the installation stage. However, several other issues may arise during the implementation process that will lead to pitfalls.
Some of most common downturns in software application are caused by: inaccurate documentation; users that overestimate their IT skills; absence of training; changes in working practices; disinterest of top management to the project; poor evaluation of software capabilities and its relevance to the company; wrong supplier.
Moreover we cannot underestimate how the cost of a new software system impacts on its implementation. While positive aspects are easily identified, a project may end up not being financially feasible. “Most organisations do not understand the costs associated with an ERP system when they first commence the implementation” (Turbit, 2012).
Another serious problem is related to the compatibility between software and hardware. During the pre-Windows age, Holtzman (1994) faced the following issue when installing a new software protocol for bus communication: “After carefully mounting and installing everything, and after ensuring that all the pin 1’s aligned with the striped edge of the cable, I powered up the system. The BIOS chip on my SCSI adapter recognized the new devices, but I couldn’t get at them through DOS. It turned out that my SCSI adapter needed a firmware upgrade” (Holtzman, 1994).
Perhaps the Murphy’s Law explains why such small details are responsible for major problems that may drive us crazy when implementing new software. On the other hand, I personally understand that in technology the issue is the fact that the relationship between men and machines is quite complex, while one (computer) acts logically, the other thinks too broadly.
I usually associate software bugs with problems related to software documentation and programming codes. In 2006, a linguistic study identified that 123,417 words were used across 200,000 bug reports. From this dataset, 82,181 words were nouns which “54% were proper nouns that represented codes, file names, acronyms and version numbers” (Ko, Myers & Duen, 2006).
However, although tests are extensively run to avoid the above issues, it all comes down to company’s business culture. Without the active participation of top management is nearly impossible to think about software implementation. It is important to highlight that a new system will impact on the way how employees work, communicate and tackle business opportunities. Hence, if senior managers do not make part of this process, even “well documented” software which is manipulated by the best programmers in the market is likely to fail.
On this regards, data management requires special attention due to the fact that an organization might have to handle data in a different way than it has ever done. “In order to continue to make progress in software measurement, as it pertains to reliability and maintainability, we must shift the emphasis from design and code metrics to metrics that characterize the risk of making requirements changes” (Schneidewind, 2001).
At last, but not least, choose the right supplier is certainly essential! It is important to make sure that the product which was promised to you will really be delivered on time and on budget. Accordingly to CHOAS Standish Reports the percentage of Failed and Challenged (delayed and/or over budgeted) projects are incredibly higher than the number of Succeeded projects.
Hence, when procuring or purchasing new software or enterprise system, make sure that your partner is your real partner!
Galorath, D. (2008) Software Project Failure Costs Billions.. Better Estimation & Planning Can Help. SEER. http://www.galorath.com/wp/software-project-failure-costs-billions-better-estimation-planning-can-help.php
Holtzman, J (1994) It may not be sexy–but it sure is SCSI. Popular Electronics, Jun94, Vol. 11, Issue 6, pp 74.
Ko, A.J.; Myers, B.A. & Duen, H. C. (2006) “A Linguistic Analysis of How People Describe Software Problems,” IEEE Symposium on Visual Languages and Human-Centric Computing, 2006.pp.127-134
Schneidewind, N.F. (2001) “Investigation of the risk to software reliability and maintainability of requirements changes,” Proceedings. IEEE International Conference on Software Maintenance, 2001 pp.127-136
Turbit, N. (2012) ERP Implementation – The Traps. [Online] Available from: http://www.projectperfect.com.au/info_erp_imp.php. Last accessed: 19 January 2012.