Project Retrospectives No. 1: The One About the Gay Reindeer

In the Spring of 1996 I was involved in the merging of two Australian second-tier banks. Just over 6 months into this venture we were finally getting our heads around the scope and scale when a third bank came along and purchased the, as yet, newly launched entity. I was the senior Test Management Consultant on this roller coaster, driving two significant Testing initiatives in Sydney and Adelaide. The untimely intervention of this third bank certainly upped the ante with respect to scope, risk, complexity and delivery. I dedicated over 2 years of my life to this project, flew thousands of kilometres between Melbourne, Sydney and Adelaide and spent countless nights in hotels and apartments.

No matter what people may tell you, Consulting is NOT glamorous. Consulting generally means plenty of hard yakka and far too many nights away from your loved ones. Consulting often means choosing between 4am starts on a Monday morning and Sunday evenings in another airport lounge, watching all the lucky folks going home after a fun weekend away. At any given time at least 30 of my colleagues were sharing the same airline food, as we wrestled with thousands of business processes and hundreds of software systems.

If this endeavour wasn’t already challenging enough, one of my Test Managers was sacked midway through the project for misappropriating funds, but that’s a story for another time. Some of our most notable challenges on this project included building a new bank (which we tested via a Model Office strategy), crunching over 700 (collective) software systems down to a more manageable 150 and leading over 1,000 people towards a series of carefully choreographed implementations. Multiple Test phases, Dress Rehearsal implementations and data migrations were de rigour on such a complex project. I lost count of how many times I refined our Test Strategy!!

The entire project primarily followed a Waterfall approach, although we did employ iterative techniques at various times. As this was a major banking undertaking, there was also significant focus on fiscal governance and therefore appropriate documentation was critically important (and had to be kept for at least 7 years after the culmination of the project).

As I said before, we had a myriad of challenges, but what were the Top 3? COMMUNICATION, COMMUNICATION and COMMUNICATION!! Surprising? Not really. With multiple locations spread across two Australian States, over 1,000 full-time personnel and thousands more impacted by the changes, keeping everyone abreast of decisions and progress was a full time occupation. From a Testing perspective, we attended hundreds of workshops, studied thousands of requirements, organised daily (and sometimes twice daily) defect triage meetings (generally via teleconference) and managed over 50 test environments. We built many of our own bespoke tools and leased Enterprise-wide Quality Centre licences, in order to be able to provide daily statistics of our (testing) activities.

I flew relentlessly. My Mondays and Tuesdays were spent in Adelaide. Tuesday evening I would fly to Sydney and Friday afternoon I’d fly back to Melbourne. Sunday evening I’d be back at Melbourne airport and off to Adelaide again. This was essential in order that I stay across the activities of the project and, more importantly, my Test teams. Tens of thousands of test cases were written, reviewed and performed during the life of the project and thousands of defects were logged, fixed and re-tested. Our Regression Testing wasn’t state of the art in those days, but it was effective. Twenty years ago there were few industrial-strength tools, but bespoke automated data-management were essential for a project of this scale and therefore we employed several of our own developers to build and maintain our toolset.

Building relationships and trust was essential to our success and therefore as the most senior Test Manager it was left to me to identify the key influencers and work with them to ensure our (testing) outcomes were achieved with the least amount of fuss. Staying off the critical path, on such a large scale project, was difficult, but most of the time we were successful. I’ve said before on this Blog, as a Test Manager, you must keep tight control of your schedule and this was never more evident than on this project. I worked and re-worked our schedule on a daily basis as new challenges arose and problems were solved. This is where I learnt the benefits of Risk-based Testing and the importance of Context. What was important in keeping our (testing) schedule on track wasn’t necessarily relevant to the other teams, but in the context of the overall project goals, it was crucial to be single-minded. This is where having the ear of the Project Director was essential. I took time to ensure that the Project Director understood our drivers, our challenges, our goals and our processes. That way I always got a fair hearing when we were under the pump. Testing isn’t the easiest thing to explain, to most folk, so I developed context-driven storyboards that made it easier for non-technical people to get it. Most importantly, the Project Director got it.

We had diagrams and charts everywhere, in order that our journey and our progress (on that journey) was obvious, even to the most lay lay-person. The coup-de-gras came when we built a replica of a new bank branch in the foyer of one of the bank’s offices, in order that we prove every new and modified process, both manual and technology-driven. This was my first venture into the use of the Model Office concept and it was a raging success. It was our shop window and it allowed anyone and everyone interested (from the CEO to bank tellers and auditors) to come and see their newly modelled bank perform using real world scenarios. We recorded everything and used the Model Office as a training aid when we got close to implementing our various releases.

So, what has this project got to do with a gay reindeer? Is there even such a thing as a gay reindeer? Of course there is! In the crazy world of mega projects anything can exist. I know, because I was that gay reindeer. As I said earlier, building relationships and trust is crucial in creating effective communication. So, in the lead up to our first Christmas we had a team party in the Domain in Sydney, where we all dressed as Christmas characters. I chose to be a gay reindeer. Why? To be honest, I don’t remember. But what I do recall is that the sharing of information and the level of trust increased ten-fold after my appearance as that gay reindeer. Different strokes!!

What surprised me most about my reindeer outing was that I did it at all. Fortunately, we were pre-social media days and therefore there is no lingering evidence of my outrageous efforts. However, there is lingering evidence that the merged three banks have gone from strength to strength in the intervening (20) years. Information is king (and queen) and sharing that information on any project or change initiative is critical. So, if you are ever struggling to get your stakeholders on board throw a Christmas party and go as a gay reindeer.

Dateline: Wednesday February 3 2015, Melbourne

Advertisements

3 thoughts on “Project Retrospectives No. 1: The One About the Gay Reindeer

  1. Pingback: Testing Bits – 2/1/15 – 2/7/15 | Testing Curator Blog

We're here to help

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s