Thank you very much. I'll be happy to leave a copy of my notes when I leave.
First of all, thank you for inviting me to appear today in front of this parliamentary committee. I feel honoured to be here today. I am speaking as a former employee of the federal government and not Hewlett Packard Enterprise, the company that I presently work for. All the opinions, comments, and input are mine and mine alone. The input I am providing today is based on my knowledge and expertise of Shared Services Canada up to June 14, 2015, the date of my retirement from the federal civil service.
I spent 37 years working in the federal government, all in the information management/information technology domains. I worked at the Department of National Defence from 1978 to 1988, and I started at the lowest possible level that you could start at in IT. Then I was at Transport Canada from 1988 to 2015, the last 15 years as chief information officer. I was the longest-serving CIO in this position.
To start, I am extremely proud of the time I spent working in the federal government, and in particular as the CIO of Transport Canada. We accomplished much. My employees were very skilled, passionate, and committed to service excellence in supporting departmental program delivery. They worked under considerable resource constraints and pressure to deliver the required services and to meet the performance levels required. TC's IM/IT had the lowest employee attrition rate of all Government of Canada IM/IT organizations. I am most proud of this. We valued our employees and they valued working at TC. They were kept extremely busy and were given the authority and accountability to get the job done. Under my leadership as CIO, TC was considered a best practice department in IM/IT strategic planning, operational excellence and efficiency, thin client desktop, server virtualization, operation automation, information technology service management, and information management, particularly electronic document management.
TC received the highest Treasury Board Secretariat management accountability framework ratings in IT management every year and was the only department to receive the highest MAF rating possible in IM. My management team and I spent considerable time sharing our best practices and lessons learned with other government departments. TC IM/IT received numerous departmental and Government of Canada awards during this period. TC IM/IT also leveraged our private sector partners, where it made sense, to help us transform and deliver the most efficient, cost-effective and high-value IM/IT services possible. We could not have achieved our goals, over my years as CIO, without private sector support and expertise. I am profoundly thankful to all my colleagues, internal and external, and my IM/IT partners for all the support provided to me and the department. I feel honoured to have worked with these people and very honoured to have accomplished what we were able to.
I also want to thank SSC employees and previously PWGSC employees who have supported the sharing of services. I'll get into details of the sharing of services before the actual Shared Services Canada was created. I know that many of these employees worked extremely hard to deliver the best possible services and were and are committed to service excellence. For that I am eternally grateful.
I'm going to take you back a little bit to the mid-1990s when TC went under massive transformation with the privatization of air traffic control to Nav Canada; Canadian Coast Guard transfer to the Department of Fisheries and Oceans; and the privatization of national airports, harbours, and ports.
This transformation resulted in TC's employee base shrinking from over 30,000 employees to 5,300 employees over a five-year period. These changes resulted in Transport Canada having an over-provisioned IT infrastructure. A $20-million data centre at Macdonald-Cartier International Airport in Ottawa had just been built before the devolution transformation direction was announced. This data centre and the associated IM/IT infrastructure, computers, network and data storage, and human resources were now being utilized at less than 25% capacity. The solution to this issue was to transfer the TC data centre, all associated hardware and software, all IM/IT infrastructure contracts, and 30 TC IT employees to Public Works' government telecommunications and informatics services, or GTIS as it was known, or as it was commonly referred to, baby shared services.
The end result was that TC was able to save costs and avoid $2 million annually from a $10-million spend. It had been costing us $10 million, and we were able to have the service delivered by Public Works for $8 million.
Public Works was also able to leverage the excess data centre space, associated IT infrastructure, and the transferred employees for other government departments, resulting in over tens of millions of dollars in cost avoidance for the Government of Canada. Service level standards were established, and performance measurement processes were put in place. It also should be noted that the Macdonald-Cartier data centre, built by Transport and transferred to Public Works, is one of the flagship data centres for Shared Services Canada today.
Additionally, a few years later, Transport Canada transitioned our wide area network service to PWGSC GTIS, leveraging the shared services already being delivered, and bought the WAN service back on a fee-for-service basis, based on volumetrics and performance. Again, significant savings were achieved for TC, and there was significant cost avoidance for the Government of Canada.
Tactical and strategical government committees were established, and all the associated required management operational processes. Both those shared services initiatives were a resounding success and should have been a best practice for the future Government of Canada shared services direction.
I will start by saying that I firmly believe in the shared services concept when done properly and where it makes sense. TC and Public Works proved that it does work.
In my opinion, when SSC was formally announced, line departmental CIOs were generally not aware, and minimal consultation had occurred on any planning for the implementation model. It is my belief that planning for SSC was done at such a high level that operational issues associated with the implementation were not completely understood. The implications and complexities were vastly underestimated, and impact on line departments was severely underestimated. When SSC was formally announced with no new money allocated to support implication, and in fact, significant resource reductions identified, it was clear to me at the time that SSC had an extremely minimal chance to succeed, right from the start.
My experience and knowledge told me that significant—in the billions—investment had to be done to meet SSC's and the Government of Canada's identified objectives. In addition to this, I was extremely concerned about the magnitude of transformation being proposed and the lack of strategic planning to support this transformation. Line department CIOs already had their hands full delivering existing and transformative services to support evolving departmental program delivery. The analogy of a change in car engine at 100 kilometres an hour could be used here.
It is my feeling that if departmental CIOs had been fully consulted, better informed decisions could have been made resulting in a modified shared services implementation that could have been actually implementable over a prescribed time period with the proper investment. In addition, any existing SSC business cases at the time for the most part did not include departmental transformation costs. One example is the email transition initiative. The business case did not include any departmental costs for application changes, training, or implementation. These departmental costs are significant and should have been included so an informed decision could have been made on the direction and implementation that was best for the Government of Canada.
Shortly after SSC was announced, a letter was sent to each line department deputy minister providing additional information on the SSC initiative and saying that existing IT services and the required associated service levels would continue to be delivered by SSC. This was shocking for all line departmental IT employees and CIOs, as SSC management had little idea of the actual services being delivered, and had minimal idea of the required and established service levels. SSC also made it clear shortly thereafter that they would not establish service level reporting.
Common belief was that SSC did not want to create service level standards, as there was significant concern that the service levels could not be maintained. This was very difficult for many departmental CIOs to accept, as many departments had implemented IT service level standards to meet the program requirements of their respective departments. TC was able to minimize the impact of this and maintain its required service levels as the TC employees transferred—more than 35 of them—to SSC remained on TC premises and were dedicated to TC.
TC had a strong operations management, which included accountability for reporting on SSC services levels. We established a position at the senior level to actually monitor and manage the SSC services to Transport Canada. TC would not allow any changes on SSC employee relocations until a transition plan was developed. This was just beginning to occur when I retired in June 2015.
SSC also made it very clear that its focus and priorities would be on the transformation agenda and not supporting sustaining existing services and support. This direction had resulted in critical hardware/software maintenance agreements not being renewed, data centre support systems not being maintained, critical hardware becoming end-of-life and beyond, software becoming obsolete, and skilled support resources not being replaced, both employees and contractors. This in turn resulted in many major server outages and SSC losing even more credibility.
In fact, one major outage—some critical systems down for days, many others down for 24 hours—at the Macdonald-Cartier data centre, which TC had built, was due to not having 24-hour/7-day power maintenance contracts. This is unheard of when it comes to critical data centre services. This maintenance contract was for the daytime only and was provided by a Toronto-based company, five hours away. When the outage occurred, SSC had no idea, or very little, of what systems ran in that data centre, what departments were impacted, and how to do the proper problem resolution and associated escalations to minimize impact.
Many more outages have occurred and continue to occur. Recent press and internal Government of Canada documents validate the severity of the issue. This is of critical importance to all Government of Canada CIOs, as program delivery is being directly and significantly affected.
After SSC was announced, I and members of my management team attempted to insert ourselves into as many SSC governance committees as possible, to both support and attempt to influence direction. It quickly became apparent to me that for the most part, SSC on the transition side didn't want to hear any input, and that the direction had already been set by senior SSC management and was not open to change. This was relayed over and over to me as I attempted to provide input, challenge directions, and offer recommendations. I offered to have TC take on the leadership for the ITSM initiative under SSC's authority, a critical piece of work for SSC success, but the offer was declined. Little progress has been made on this file.
I specifically remember one planning meeting that really stands out in my mind, even though it happened three years ago. I was told that TC computer applications did not meet SSC standards, and therefore would not be moved to the SSC production centre in Buckingham, as had been planned. This was the first time I had ever heard of these SSC standards. I asked what consultations had been done and why these were not Government of Canada standards. I did not get an answer. I escalated up the line in SSC and still did not get an adequate response. I could go on, but you get my point. I am sure that for some of the most senior SSC executives, I was considered a pain in the backside.
I firmly believe that when consultation did occur, it was done for SSC to tick a box rather than to hear what departments wanted to say. I also heard the same from many industry experts. At the most senior levels, SSC seemed to know all the answers before the questions were even asked or input even provided. Government of Canada partners felt marginalized and even under attack from SSC, as the rules of engagement and procurement continually changed.
It became apparent to me that line departments were considered as users and not partners, even though SSC continually indicated the opposite. To be partners, both sides must be included in the planning and decision-making process. This was simply not the case in the vast majority of instances on the transformation side.
Another one of the issues I had with SSC was that I was never able to obtain an integrated strategic plan covering off all the transformation areas. The ETI, or email transformation initiative, and network and data centre consolidations are all interrelated and impact each other. Add in the departmental implications and you have a strategic plan to begin to move forward. Without one, it is impossible to do the detailed planning required for successful implementations. In my view, this has never been done properly. I asked continually to see this plan and the associated HR plan for SSC employees. I do not believe one existed then.
After SSC was created, procurement resources were transferred from the then PWGSC to SSC, and SSC was given the authority to do their own procurement. This, in my view, was a grave mistake. It has resulted in significant issues with procurement, including issues with process, security exemption clauses, fairness, transparency, and perceived favouritism. It is my opinion that the Government of Canada should have one procurement centre of excellence, and it should reside in PSPC...what is it called now?
Yes, Public Services and Procurement Canada. It's changed since I left.
My rationale for this is the same reasons as why SSC was formed itself: efficiency, effectiveness, and delivering high-value services.
As mentioned earlier, the Government of Canada has some of the best and brightest and most competent IT/IM workers in Canada. What has happened to these workers who have been transferred to SSC for the most part has been very disconcerting to me. Many employees at the working level have been marginalized, and not consulted or involved in direction setting and/or decision-making. They are uncertain about their future and have been asked to deliver increasing amounts of work under increasing pressure. The Government of Canada employee survey indicated that SSC employees have the lowest morale of all Government of Canada departments. I believe that Government of Canada employees are the most important asset the government has. It is tragic that these issues with SSC employees have occurred. Employees are the most important asset in any organization.
My recommendations, at a very high level, as of June 14, 2015, are: identify SWAT teams to work with each department to resolve outstanding production problems. Once this is done, including ensuring hardware and software, etc., is not obsolete and the necessary maintenance agreements are in place for all IT infrastructure, including data centre support systems etc., identify and agree upon service standards and report monthly on SSC performance. Ensure monthly performance meetings are scheduled with each department and SSC production operations. Establish escalation processes and all necessary production governance processes and procedures. Implement the necessary automation tools to include asset management. I firmly believe that SSC does not know what is in the environment today and what they are managing. Identify all hardware and software, including licence compliancy, network components, application configurations, and interfaces, etc. Then there is server/storage deployment automation; operations monitoring and performance management; an integrated ITSM, building on the largest install base already implemented in line departments. This supports every facet of production operations, from help desk to change management, to configuration management, to access control, and all the required operational support processes.
Freeze, where possible, all transformation activities and reset the plan, with input from line department CIOs and industry experts. Some major IT companies have implemented massive transformations internally and have the required experience, services, and best practices to support the Government of Canada's transformation initiative. Consult and engage with these companies that have proven track records and have lived transformation activities. Align this plan with the Government of Canada's strategic IT/IM plan being done by Treasury Board and also departmental IT/IM strategic plans. Once this plan has been approved, develop an implementation plan that will identify the lowest-hanging fruit and all implications. This plan should include the target architecture that all departments should migrate to and when over the next one to three years so when systems are rebuilt or software is bought, it is bought for the target architecture. Data centre convergence will be simpler when departmental systems are standardized.
Government procurement should negotiate enterprise licence agreements for all software in this target architecture to drive down costs. Business cases should then be developed, including identifying all costs. Approval of business cases will dictate priorities, resources required, and scheduling of implementations. While this is occurring, give departments approval to move forward on IT/IM delivery priorities—because we can't stop delivering the line services, and you did—with SSC acting as a service broker.
Start small. Think big. Treat line departments as real partners and really listen to what they have to say. Involve them in the decision-making and keep them informed moving forward.
The next recommendation is to cancel the email transformation initiative due to the significant schedule delays and cost overruns, and await the results of the strategic plan. There is little credibility left with the ETI as it stands today. Once directions and initiatives are approved, develop an HR plan for all SSC employees, including training, career development, etc. For those employees whose functions may be outsourced or privatized, utilize workforce adjustment and try to place them, through retraining, etc., into other positions or transfer employees to external departments. Minimize any layoffs to the absolute degree possible. Align this plan with departmental HR IT plans and manage IT employees at the Government of Canada level. Keep employees aware of the developments affecting them and make them feel that they are part of the solution.
The next recommendation is to transfer procurement authority and associated resources back to PSPC to create a procurement centre of excellence.
The last one is to establish, augment, and strengthen the required tactical and strategic governance: tactical governance to deal with present IT/IM services and strategic governance to deal with SSC direction and project implementation. Manage expectations and communicate.
Now I'll move on to my final thoughts. It is my feeling that without immediate and direct intervention, SSC will not accomplish and meet the objectives assigned to it, resulting in significantly increased costs and extremely inefficient and ineffective IT/IM service delivery, the end result being that line departments will not only be unable to deliver their existing program delivery services, but will not be able to transform to meet evolving requirements. This will ultimately result in the federal government becoming less and less relevant, and Canada as a country becoming unable to compete in the global marketplace, and a civil service that will be not only no longer the envy of other countries but unable to support the needs of Canadian citizens, businesses, and others.
Canada is the best country in the world. Let's keep it that way.
Thank you for your time today.
I hope Chris feels better, but well done, well said.
I was there at the very beginning. With respect to the planning that was done prior to the creation in November—I moved over there on November 22, 2011—I was not involved in the consultation or in the administrative service review. I was asked, because of my experience and some of the success I'd had, to move into the operational role and, because of the relationships I had with people like Chris and within the community, to work with them to basically keep the lights on in the legacy systems. That was my role operationally, but I also had a role to deal with the people who work for us within the organization.
From a planning perspective, I think Chris's point was that there wasn't a lot of consultation, and I would support that point. I think that is valid. Often when we pull these things together and make large machinery changes in government, there's some secrecy around that, so it was a bit of a culture shock for organizations to find out that their back-office folks would be moving to a new organization.
The way that was meant to be mitigated was that those people would actually stay within those organizations. One has to understand that no HR plan had been created when SSC was created. In fact, there wasn't even an HR management system when we arrived. With regard to most of the systems we leveraged, we had agreements with people like Chris and the 43 departments that were involved, such that they would continue to pay and manage the people who were still within their buildings, supporting their former data centres. It was more of an ownership issue. They continued to work in their operational environments, but they now wore a T-shirt that said, “I work for some other new entity that I don't know a lot about”. I just want to say that as well.
From a legacy perspective, Chris waxed on a lot about some of the success he had as a CIO, and he was recognized with a long-time service award for outstanding leadership in this field. I have to say that in many of the departments whose back office I became accountable for, that maturity level, if you will, with respect to understanding the assets that they had.... I don't want to leave an impression that there was a detailed service level agreement within each CIO organization, because that's completely untrue. For the more mature organizations like Transport Canada, which had moved back offices almost a decade before the creation of SSC to ITSB and GTIS, as you mentioned, there would have been a service level agreement, but the only departments that had done that were Transport Canada and—Chris can correct me if I'm wrong, since I'm going on recall—Canada School of Public Service, but they were small entities and they represented about a couple of 100-million dollars' worth of operations on a day-to-day basis.
There was no financial system when we arrived either, so we continued to use departmental financial systems. Where we were managing contracts and had taken responsibility for contracts, we were still leveraging the community to help us manage those contracts in the legacy systems as well.
I'm trying to answer your question around HR transition consultants. You need to understand that I looked at it as a leadership opportunity. I believed the enterprise approach was the right way to go, and I think all of us did. It was the right thing to do. Chris's point was that if we could do this again, would we put more time into planning it, getting a coalition of the willing, starting small, etc.
I'm sorry if that's—
I didn't mean to...when I referred to legacy, right? It just was the existing stuff; that's just a term.
We can't forget that in 2010 there was an Auditor General's report that looked at five major systems. If you look at those recommendations, you'll see they didn't say the legacy is glowing, is working great, is all interconnected, and is sustainable. They said quite the opposite. They said you need to make major investments. The recommendation, just on the five systems studied, was two-billion dollars' worth of investment.
When SSC was created, it did not inherit, or the departments that came into the marriage for the most part did not show up with systems that were not in need of investment.
I'll use some examples, because I think Chris made some great points, and I don't disagree with anything he said.
When you look at Canada Border Services Agency and the importance of that organization from a security perspective, you see it links into 17 different departments and agencies.
During my tenure in operations, we went from hundreds of outages across that ecosystem of departments, and using and leveraging the power of working together where back office—I'll say back office—employees now were working under the same roof, we were able to analyze single points of failure.
We were able to analyze, when we make these kinds of changes from a maturity perspective and development of systems, that we have to go through the testing procedures in a certain way to make sure that, when they move to production, they actually work.
My point is that we actually uplifted the performance of some fairly dated systems around managing the border. There was no singular place where security operations came together, where all government departments were actually hooking into the Internet except for at various places, and that wasn't always the case. We were able to work with our security partners, pull the security teams from multiple partners together, and manage cyber-attacks in a coordinated way: what's the problem, who has the problem, etc.
There were some major benefits to bringing them together for the first time, for example, with the Heartbleed or whatever virus you want to talk about. You remember the one that brought CRA systems down, where someone actually infiltrated the systems. That was the first time we were able to work in a coordinated fashion and bring major systems down in order to protect government systems going forward.
The idea behind shared services to improve security is absolute. What I think Chris's point, if I could say it very briefly—
Thank you very much for the question.
Just as Kevin said, departments have different maturity levels. I'm speaking about Transport Canada. I'm not speaking on behalf of my colleagues in other departments, just to clarify.
I think really what Shared Services needs to do is take a step back. There's just too much on the platter to do what it is trying to do, and I think it's an almost impossible challenge to do that, based on how government is evolving.
What we need to do, as I indicated, is to pick where there's one transformative area that makes the most sense, that's going to perhaps cost the least and could show some success. Right now there's a serious credibility issue in line departments with their shared services. People believe in the concept, but the results are not generally there everywhere.
Don't get me wrong. People have worked extremely hard, and there have been some successes, security being a major one. I couldn't agree more with Kevin on that.
I think it needs to retool, do an analysis, set the expectation of what can be done, when it can be done, and what investments are required and then move forward with that.
The first thing, obviously, is to fix all existing production problems that are in place today. If you don't fix those today, we're going to continue to have more outages that are going to cause more issues moving forward. You don't want to spend a whole bunch of money if you're going to migrate off those in the future, so it's a balance, but you still have to make sure the right services are being delivered.
I think the question was around a senior level of support. Part of what Chris is talking about is an integrated planning regime among communities. I was going to get to the issue of trust. The way SSC was formed, it was difficult for us from a trust perspective. You can imagine: departments were losing money, losing people, losing resources, almost losing ownership of their operational elements. It became somewhat confrontational.
From a speed perspective, if you want to move faster because you believe this enterprise approach is the right way of doing business, which we all believe it is, somehow from a community perspective we need to build that trust. It can't be just the sole responsibility of those at Shared Services Canada. It has to be a very prioritized, wilful movement towards more enterprise approaches to systems.
Look at what Ontario is doing, or at what Northern Ireland is doing. It took them quite a bit of time to move forward, but the approach was slower and more paced. Clear results and outcomes were identified prior to making movement. I would just suggest a focus on rebuilding that trust and having demonstrable action around speed. In the end, I think that will bring the cost savings we are looking for as an organization.
The savings component is an issue of trust. We definitely leveraged the power of economies of scale to negotiate unprecedented rates, and competition to get a lowering of the price for very good equipment, to maximum advantage. The problem is that we assumed, for instance, that in some of the services, such as cellular telephony, the quantity would remain constant. We assumed that on what we inherited, let's say 20,000 smartphones, the queue would stay the same, the price would come down, we'd be able to leverage that service, get those savings, and invest that money in voice over IP or whatever investment initiatives. When the quantity went up to 80,000, it made it somewhat difficult and redundant, no matter how much better the price of the contract was.
I guess I remain very hopeful, but hope is not a strategy. I think what we need is a reset. I was here on Tuesday when deputy minister Ron Parker was here. He spoke to some of his plans and his intention to come back with a more fulsome strategy. I echo Chris's comments that the community should be brought into that strategy early, and it should be presented as an integrated approach and plan to uplift IT enablement across Canada.
This is a large nation and a wonderful nation. I served in the Canadian Forces for 20 years. I'm a big believer in the public service. I guess I would just say that we're a large country but we're small enough as an organization that we can work together and make this happen. I believe we have the leadership in place to do that, and the will; we just need the plan. We need it actioned and we need to follow it. We need to be somewhat flexible and agile with that plan when things come up, e.g., fires in Fort McMurray or cyber-attacks, but the most important thing is to re-establish that trust.
How far along did we get under my tenure? We focused on incident management and supporting problem management in that space.
If I could get away from the traditional cost per terabyte, etc., and a pricing strategy—and I'm going to leverage where I am now, and as the committee knows, I look after all federal infrastructure across the country—I think there is an opportunity here to look at the workspace, and technology, and the environment, and greening, and at some of the outcomes we want around alternative working arrangements, and mobility, and people being able to work from any place, anytime, anywhere. Chris mentioned some of the success they had with this at Transport Canada.
I would try to measure success against those types of parameters. How much workspace is available that's collaborative, that allows any public servant or official to go into a space, work collaboratively, hook up a VPN into a Wi-Fi system, be available in the cloud, and be operational, whether they're a police officer or an inspector who is in the middle of a farmer's field and they need to pass information back and forth; I would really encourage measuring the go-forward position for those kinds of things.
Another example could be alternative working arrangements and teleworking. Right now when we build workspace for individuals in the public service, we look at one-to-one ratios. Many of you who visit our departments know that as you walk down the hallways, there are lots of empty spaces and under-utilized space, because that's just the way we do it today. Imagine more of a place where we can reduce GHG and the 40% of GHG associated with infrastructure simply by reducing the footprint, while also allowing that mobility and flexibility so that someone who is coming from Kanata into the downtown core and hits a traffic jam can pull off and move to the Carling campus and work from there and be mobile and free. I think those are the kinds of things.
The traditional IT way of doing it is around a service catalogue, around clear service levels, around how many outages you have. That's another way of doing it. I believe Shared Services Canada has made quite a significant movement in that space, but I'm not sure about the maturity level now. I've been out of there for about nine months.