When Ansett went belly-up in September 2001, only hastily developed contingency plans, a lot of hard yakka and heaps of teamwork prevented Air NZ’s IT infrastructure from sharing the fate of its doomed Australian counterpart.
It should have been the day a long-held dream was finally realised. By mid-September 2001, Ansett and Air New Zealand’s IT infrastructure and operations finally should have been integrated, consolidated into a single data centre running out of Melbourne. Intended to be the glorious culmination of months of careful planning and preparation, the move to a common data centre would have completed a major stage in a projected three-year program of work on moving the two airlines to common systems and processes. Celebrations were planned.
Instead, with the public announcement of Ansett’s collapse on September 12, just a few days before the new data centre was to have been fully commissioned, Air New Zealand’s (Air NZ) dreams of IT integration were well and truly shattered. Not only would the data centre merger not now take place, Air NZ group vice president strategy and planning Andrew David was facing major headaches to keep his own airline’s IT operations sound.
“It had been obvious for some time that both Air New Zealand’s and Ansett’s systems needed replacing,” David says. “Before the collapse there was a program of work under way to integrate both companies’ IT infrastructure, which had actually all but happened. We were a few days away from having one data centre in Melbourne when it all went belly up.
“We’d already saved a sizeable sum. We had quite an extensive integration program under way, and we had a program planned over a three-year period to move to single systems, single business processes right across the two airlines, and we had made significant progress. We’d gone to single suppliers. We were all but ready to turn the switch on in the data centre in Melbourne and turn off our Auckland data centre.”
Then Ansett collapsed and Air NZ found itself devoting the next six months to efforts to breathe life back into the Auckland data centre. Instead of helping the 500 staff in Melbourne settle into their roles, David was forced to begin an immediate program of rebuilding, to secure the information Air NZ needed to survive. And he abruptly found himself facing the biggest test his leadership had ever faced.
“I had 500 staff in Melbourne. I felt they were part of my team,” David says. “I was in Melbourne, and had to work through that process with 500 staff and manage it the best I could.”
When the airline — which had its beginnings as a regional operator and grew to be one of the best known and respected domestic carriers in the world — collapsed some 13,000 jobs were lost, as well as many regional services considered essential to country areas.
Commentators have blamed Ansett’s demise on various causes, from government deregulation and competition policy to predatory pricing by Qantas and serious mismanagement by a long line of Ansett directors. Its impact has been clearer, starting with the $10 ticket tax designed to secure Ansett employee redundancy payments and visible in the new life given by its demise to the ambitions of new contender Virgin Blue. However, the internal fallout on Air NZ, and particularly its IT infrastructure, has been less well publicised.
David says the collapse of Ansett potentially exposed Air NZ to major risk. In the end, only some hastily developed contingency plans, a lot of hard work and the good will of vendors saved Air NZ from much more serious consequences when it had to change direction abruptly on the data centre merger.
Fortunately the IT team received some warning shortly before the collapse that Ansett’s future was looking shaky. That meant limited contingency plans were at least already in place for a rapid regroup back to the Auckland data centre. The steps needed to ensure Air NZ had the capacity to continue to run its systems had been outlined, however sketchily. Even so, David says, both the head of IT infrastructure and chief technology officer had to work very, very hard to achieve continuity.
“We did have for a period of time some problems with performance of the systems and we were exposed for a period of time where we didn’t have what I call a robust, stable platform with appropriate fall-back facilities. We now have all that in place. It took us about three or four months to recover properly but we’re now back to the situation where we have a very, very stable platform,” he says.
Best Laid Plans
When the merger between Air NZ and Ansett was sealed in August 2000 and Air New Zealand acquired 100 per cent of Ansett, it rapidly became clear that for the two airlines to be consolidated as quickly as possible they needed a single set of business processes and a single set of business systems supporting those processes. To that end David ran a series of workshops with staff from both sides of the Tasman designed to examine both airlines’ current capabilities, rated according to best practice, functionality delivered to the business and cost.
The review, which David says highlighted the different leadership and focus of the Air NZ leadership and CIO compared to their Australian counterparts, showed that in general Air NZ’s systems tended to be very strong on the operations and engineering side as well as on some aspects of revenue management. Ansett’s strengths were on the customer side. Those differing strengths were eventually reflected in decisions David made around which systems to choose. The result in the best of all possible worlds might well have been an airline running a full gamut of best practice systems.
“In some cases we adopted the Air New Zealand systems, in other cases we adopted the Ansett system, and in a few cases we decided that since neither airline had what we needed we would move to a new system and purchase that externally,” he says. “Air NZ had a pretty good track record of delivery, but sometimes at the cost of an overall view and plan of where they were going. Ansett had done a good job of overall architecture, so they had strength in things like middleware and some of that architecture to facilitate delivery of systems.
“We were starting to leverage off of each other’s strengths,” says David.
Consolidation was to be a three-stage process. The first, obvious step was to leverage the infrastructure and where possible to consolidate to a single network, single data centre, single LANs and single set of providers. Air NZ decided to merge the entire infrastructure of the two airlines, including the telephone system, WANs, LANs, mainframes, Unix boxes and databases. That work should have then facilitated the two airlines moving to a single set of systems, David says. By September 2000 the team had completed a cost benefit case for this work and had a board paper signed off and approved.
The next step was to merge the two IT organisations. By the time of the Ansett collapse David had oversight of a single IT organisation with an IT management team that was both Melbourne- and Auckland-based and comprised staff from both airlines. (Managing the Melbourne staff was a challenge, he concedes, but one which became easier as he came to know people personally and could make more use of videoconferencing and other technology to manage oversight. Even so, he says, he used to spend three days in Melbourne and two in Auckland every week.)
Finally, the team planned to move the two airlines to a single set of applications that were going to enable to two airlines to move to one set of business processes.
“Now the infrastructure, because that was under my complete control, I could move at a pace, whereas moving to a single set of applications clearly depended on the business moving at the same pace,” David says.
“Some areas of the two airlines were integrating quite rapidly; for example, revenue management. We moved quite quickly to a single revenue management system because we moved very quickly to one revenue management team. Other areas of the companies were taking a little longer to integrate and hence our plan to move to a single set of applications was spread over three years, whereas our plan to move to a single infrastructure environment was planned over one year.
“We were on target and we were in budget and we’d already saved a few million dollars in terms of that integration, and obviously there were further savings planned,” says David. “By the time we got to that week, September 13 — two days after September 11 — we were less than a week from consolidating to one data centre in Melbourne.”
Melbourne had been chosen for the data centre because its infrastructure was newer, it had a complete fall-back solution by way of an offsite data centre, and investment in that data centre had already been considerable. The parent company had already put in the bandwidth between Auckland and Melbourne, upgraded the surrounding infrastructure and made the call that it would move the various servers and Unix and mainframe boxes from the Auckland data centre across to Melbourne.
Then the corporate disaster that had been foreshadowed in the gloomiest forecasts finally struck. Ansett went into receivership.
David was left to unscramble the egg. If there were any silver linings in the clouds surrounding that collapse, they were only that the IT team had had a tiny bit of forewarning and that at least the merger of applications had not yet begun.
“We’d had some warnings that the collapse might well happen so we’d started to put in some contingency plans, but nobody knew anything for sure in such a dynamic environment. Things were changing constantly, so while we had some warning, it wasn’t very long,” David says.
“What we then had to start to do was say: Well, in the event that something does happen, how do we regroup back to the Auckland data centre and how do we ensure that we’ve got the capacity in the servers and the capability to continue to run the Air New Zealand systems?”
When it became obvious Ansett had collapsed, David immediately flew to Melbourne, where he spent a week helping staff through an “extremely traumatic period of time” while speaking often to Auckland staff nearly as upset about the whole affair. The uncertainty was compounded by the way the PricewaterhouseCoopers was first appointed administrator before the entire administration was moved to Andersen.
“This was a very emotional period for everybody,” David says. “When you’ve got a plan that’s robust and you’re making good progress, something like that is a disappointment, but it was more the people factor than anything else. It’s just disappointing to see so many people in positions where they don’t have jobs. It was very traumatic.”
Worse, there had already been redundancies when the two airlines consolidated the previous year, but for those affected then at least there were full payouts and extensive support mechanisms in place. The staff still working at Ansett when the airline collapsed, David knew, would be in a much less secure position.
“You can’t give people any guarantees, and it’s not a very pleasant situation. We just had to work through it,” he says. “I’d have to say from my own personal perspective I was very fortunate that the Ansett IT team acted in a most professional and mature way, and any anger was never directed personally at me.”
Finally, after a week of uncertainty, David was forced to make an emotional farewell and leave an acting CIO to work with the administrators, someone to decide who would be let go and who would be kept on to “unscramble the egg of Ansett”.
David had even more pressing work: setting up a team to work through what became known as “the de-merger”.
Unscrambling the Egg
Back in Auckland the team set up to manage the de-merger had to face a number of immediate and very “messy” questions about ownership of intellectual property and infrastructure, and about which Air NZ-owned servers could be brought back to New Zealand and which would be left with the administrators. Much of the equipment from the Ansett data centre also had to be returned to suppliers, including most of the data centre’s PCs which had been leased from IBM.
“There was a whole inventory collection and accounting exercise,” David says. “So we had to set up a team and a whole process to work with the administrators and with the Ansett team to unpick that whole environment which we’d spent 12 months putting together.”
He says there was a period of time where both Ansett and Air NZ people found dealing with the administrators frustrating, at least until they came up to speed. Fortunately most people acted very professionally, motivated by the common agreement that it was not in the best interests of either party to in any way undermine Air NZ’s ability to run its own organisation. At the same time, the administrators knew Air NZ had no intention of trying to claim equipment it had no right to.
“Nobody was trying to play any games here,” says David, “and once it was understood that we were all being honest and open, it flowed, and we unpicked it. It took probably through to about January or February but we did that and moved on.”
Minimising Air NZ’s exposure meant moving to put new infrastructure in place in Auckland very quickly. Those new purchases did not put a huge hole in the airline’s budget, David says, and were allowed for in the budget provision for the write-off of the costs to do with the Ansett fallout. But that is not to say that the de-merger was not ultimately a very costly exercise.
“We did provide a figure to the board that allowed for the de-merger and it did run into the millions,” David says. Perhaps the real cost, though, was in the economies of scale and scope which would have been achieved once the two airlines finally became one.
“Now that didn’t exist going forward [the economies of scale], so obviously the synergy that had been spoken about in terms of the merger and the IT savings were part of that. Once that disappeared those benefits disappeared as well,” David says. “We did get some benefits in that some of the deals we did with some of the vendors we were able to leverage off. So there was certainly upside and there were certainly benefits of that; but most of the benefits went when Ansett went.”
Talk, Talk, Talk
Along with a lot of goodwill, david says leadership was the factor that saved Air NZ’s bacon when the de-merger looked to threaten the airline’s smooth running.
During the entire crisis, and particularly in the early weeks, communication, he says, was absolutely vital. The twin teams across the Tasman had telephone hook-ups daily during the first week of the crisis, and weekly thereafter. It was a time for learning many lessons about good leadership and how to communicate.
“After a while I went to weekly hook-ups with the IT team and every week I’d get them all on a telephone conference — because the IT staff here are spread around four or five locations — and I’d keep them on the phone and just take them through what had happened that week, what I’d done, and invite them to ask any questions that they had and respond accordingly,” David says.
“On reflection, I learnt a lot about communication and about being visible and walking the talk, as they say, which we all think we do. But on reflection, none of us do it as well as we should do. So I learnt a little bit about that.”
Communication, he found, was all about being there for staff, listening to their concerns and above all, being honest. You have to deliver the truth, and that is not always what people want to hear. David believes he succeeded partly because of a determined belief that 99 per cent of people are basically good, and partly due to an equal determination to maintain the trust of those good people.
He says he remains convinced the approach to merging the two airlines was sound. The airlines, after all, had been making excellent progress until external factors undid the plans. The same with the aftermath.
“Again, in the way we managed the de-merger: we acquitted the right things, we got on the front foot,” David says.
“You certainly have to be proactive in a situation like that because you’re dealing with a third party who just doesn’t have any history or knowledge and you have got a number of people on the Melbourne side who’ve left the organisation and taken that knowledge with them. So I would do the same again there.
“As I said, I think the lessons learnt for me were more personal ones, around being visible and leading and communicating.” Would he ever wish to be in such a situation again?
“I think that’s a one-time in a career, but you learn a lot from any situation, good or bad,” he muses. “Perhaps you learn more from a bad experience than you do the good, and so I certainly say I learnt a lot. Would I want to repeat it? Absolutely no.”
Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.