Bloggo back to the blog
Three Weeks and Counting …..-->
Three weeks to go, and the journey into PROD continues. There are some significant on-going challenges, late user decisions that have a large impact on time lines, and a UAT environment that runs like a dog. Are we winning? I am not even sure that we are coming second, but we ARE still in the race to implement into PROD in mid-May! The story continues …………
Another interesting week as days slip by all too quickly. “Days in the run up to a holiday pass by at snail’s pace, whilst those approaching a software implementation at almost the speed of light” – seldom have I agreed with Brett Gonzales more!
In my previous piece (“The View from 4 Weeks Out”), there were some figures embedded in the text. I am tabulating these to give a summary week by week. Statistics can be useful, but they do not always tell the whole story, and although the totals in the various columns need to be ascending (“UAT passed”) or descending (“System Test cases outstanding”), overall details need to be treated with caution – “There are lies, damned lies and statistics” – Benjamin Disraeli, UK 19th Century Prime Minister. So treat with caution.
In particular, I need to mention what ‘defects’ applies to. This is items that do or may need to change. Everything is recorded here, including some things that are thought to be errors, but where the error was with me! Around 3 weeks from implementation day was when business identified 3 key changes (which, as described now, are recorded as ‘defects’). ‘Defects’ may be technically incorrect, but it does mean that everything is recorded in one place, and it gives a list of matters outstanding. So there are some items in the 13 raised this week that are business generated changes, but as a project, it is almost down to single figures, having averages high 20’s for the previous 3 weeks. Not sure what the totals will be for next week!?!
The project view is that the business users are king. They pay the money, and so they determine what is delivered, in what can be loosely described as an Agile development methodology. Even in my days in a very heavy-weight Public Sector waterfall project, I always fought against the “fortress IT” school of thought – we in IT know what is best for you. The mid-May implementation will provide positive benefit to the business. It will enable three key reporting areas (Analysis of Change, Lapse Analysis and Experience Investigation) to be carried out from the same integrated data warehouse for the 5 vertical business segments, and this done “in the same way”. Data behind the reports for individual business segments may be very different, but when looking at the reporting areas and business segments, the matrix will be complete. Previously, these activities were impossible, non-standard and/or cumbersome. Reporting will also be quicker, and enable quarterly pictures to be assembled, not just end-of-year outfall.
Two of the business generated changes / possible changes have resulted from seeing data pass through the warehouse into reports – that’s what UAT is all about, adding value, whilst preparing the user base for the forthcoming implementation. Two of the three big ticket items are concerned with data quality, whilst the other reverses their former decision of now wishing to load 2 years of historical data for one of the four monthly data feeds. The original intention was to load the two quarterly feeds (one significantly amended, one new) from 03/2013 and the four monthly feeds from 01/2013 onwards. The aim of having this all loaded by May month-end seemed easily achievable – and why a mid-month implementation date was chosen. To additionally accomplish data-loading for one of the monthly feeds for all of 2011 and 2012 (and this must be done as ‘monthly loads’ rather than a single large load because of the input file structure) adds to the complexity – and resulted in changes to schedules for both existing and new job sequences. The data quality ‘changes’ have yet to be finalised – we are not sure what will be done to resolve difficulties.
The additional challenge this week has revolved in part around the loading of historical data for the one monthly feed, as described above. This new data feed requires data for the month being loaded from one of the business areas already implemented. Whereas genuine data is available for all months for 2011 and 2012 in the PROD database, there was only 6 months of this data for each of 2011 and 2012 in UAT (why would you necessarily want to keep UAT up-to-date with PROD data?) The project team are attempting to enable all 2011 data for the new feed to be loaded into the UAT environment, but that involves pre-loading 6 months of this other data first. Not included on the sign-off UAT plan? Absolutely! Gives user confidence for PROD if undertaken in the UAT environment? For sure!
So, not only is there data for the agreed UAT schedule that is being loaded into the (shared) UAT environment, but also the extra data to enable the 2011 back-log data to be loaded, and the 2011 back-log data itself (in due course). It just so happened that with using the environment in a very intensive way for the first time in many months, it was found to be running like a blind dog with three wooden legs and the other one missing – almost so slowly as to be faster walking.
Two of the ‘carried forward’ items from System Test were passed, as well as a significant number of the genuine UAT test entries. The other important activity in the week was the sign-off for the second implementation into PROD next week. This is still a preparatory implementation, ensuring that all is in place for the main implementation in mid-May, and the implementation is technical, with no business impact. As such the business delegated sign-off to myself on their behalf. Hopefully, some of the outstanding defects (12 at the end of the week) will be closed after this next implementation. Incidentally, the additional historical data to be loaded necessitated two minor amendments to existing PROD schedules, which were identified this week, and will be in the second preparatory implementation. The whole aim is to enable the real implementation into PROD (the “3 Weeks” in the title) to be as smooth and efficient as possible, so that several of the data files can be loaded that day. Or so the plan goes.
I said that the figures don’t tell half the story, and in that I was not wrong. The key message is that we are 3 weeks out, and although the ride is bumpy, the train is still on the tracks. I hope it continues that way!
Peter Morgan is a testing professional who has been involved in the ICT industry for more than 30 years, and worked in the freelance marketplace for much of that time. His time has sometimes moved from testing to ‘development’, but he would add “always using the mindset of a tester”. He is passionate about testing and a firm advocate of testing qualifications. An enthusiastic speaker and author, Peter tries to base his output on hands-on experience, attempting to relate fine sounding ideas back to how it will affect Joe or Jane Tester in their everyday working lives in the war of attrition that we call software testing.