Government Operations Committee on Feb. 6th, 2012

[Denis Blanchette, NDP, Louis-Hébert, QC]

Thank you, Mr. Chair.

Mr. Gupta and Ms. Oliver, welcome. At the beginning of your presentation, you described certain conditions that were necessary for this kind of project to be successful. You also warned us of certain problems.

Allow me to use the word "project", even though we understand that you're talking about multiple projects. If we do things right, if we do things by the book, in your group's experience, when might we hope to see the project being profitable, since it's necessary to invest up front to obtain a return on investment?

What do your members have to say about the time needed to make the operation profitable for everyone?

[Karna Gupta]

It's an interesting question because there's no clear answer, largely because the financials vary from project to project or from overall program to program on a transformational project, and they come in different sizes and shapes.

I can only talk about the one I was personally involved in. When we looked at it, we saw that it was between a three-year and five-year payback. That's why I mentioned that we fell within that benchmark of a three-year to five-year payback.

Payback came in a couple of fashions. One was your resources, with “resources” meaning the people on your staff. You have merged companies and you're simplifying the whole IT and delivery mechanisms from a consolidation point of view, but what do you do with these people?

You have two choices. One, if your company is in a growth mode, meaning you're doing more things more effectively, you can redeploy the people to do something differently. You'll always have some talent that does not fit the need, but at the end of the day, you've finally found a set of people who are doing other work that they didn't get done before.

The second part of the benefit, coming purely from the private sector, is how you deal with the end market and the customers. From the customer's side, the benefit is absolutely direct, because the last thing any private organization would want is to deal with 16 different customers in 16 different ways and have training done differently in trying to deal with that.

So that benefit, in terms of both effectiveness and dealing with customers, is absolutely direct. There are no two ways about it.

[Denis Blanchette, NDP, Louis-Hébert, QC]

Thank you very much.

You said earlier that the exercise we're talking about here was the same as merging 43 companies. What the Government of Canada wants to do is a tremendous undertaking.

In your experience and the experience of your members, which of Shared Services Canada's models are the most adapted to that type of challenge?

[Karna Gupta]

If there is one significant challenge that the shared services would face, it would be that you're talking about 43 departments that are all operating differently, with different cultural tones. Merging them into one organization with similar values and cultures—that's where the challenge would be for the leadership.

That is a significant challenge, and I just cannot imagine it. In the private sector we don't have something this large and complex in terms of merging organizations on the cultural side. But that needs to be paid attention to. This is where you may need some outside help in order for the shared services to deal with these kinds of mergers.

You're talking very significant change here. It's not the technical side, it's not the execution side, it's the people side. The people side is very critical here, and that needs to be paid attention to.

[Denis Blanchette, NDP, Louis-Hébert, QC]

You are talking about organizational culture and governance with a view to implementing a project that might be somewhat successful.

I'll repeat the question I just asked, but I'll ask it slightly differently.

Are there models that, given the size of the operation, should right away be excluded to better target what needs to be done and how it needs to be done? Do you have any recommendations for us in this respect?

[Karna Gupta]

I cannot recommend specifically what would work for the shared services, because I don't know the composition of the organization. I can play back to you what I have seen work, which is to have the projects very well defined with a start and an end date, a very “blocking and tackling” approach. If a project starts on a certain date, it needs to end in six months or nine months from now. It has to yield x number of dollars in benefit, and certain customers will be dealt with in a certain way. Those milestones must be met.

The governance of all of these projects, how they'll be managed, is where a lot of attention needs to go up front. If I were to translate that, with the complexity that Shared Services faces today with 43 departments, they should spend all of the necessary time up front planning it rather than jumping into execution tomorrow. Significant effort needs to go into planning, and in this case, I would say, consulting the private sector to get the input of what has worked and what hasn't worked. Whether it's the Ontario government or IBM or HP, have those consultations. Then assess what works best for that specific instance. That's where a lot of investment needs to be made.

For most of us not doing this, sitting outside as observers, we'd feel that a lot of time has been going by, and a lot of investment going in, with no output. But this planning stage is absolutely critical to get it on the right footing. The governance itself and how it needs to be executed: that is where the time needs to be spent.

[The Chair, NDP, Pat Martin]

Thank you, Mr. Gupta.

Thank you, Mr. Blanchette.

We now go to Kelly Block from the Conservatives.

[Kelly Block, Conservative, Saskatoon—Rosetown—Biggar, SK]

Thank you very much, Mr. Chair.

I just want to thank you for coming here today. It is a very interesting study around Shared Services Canada.

Mr. Gupta, you mentioned in your opening remarks that collaboration and trust between the government and the private sector are key, and that both the Government of Canada and the private sector must be satisfied that they are in a win-win IT relationship.

Do you believe this win-win relationship has existed between the government and the private sector in the past? And how do we ensure that it exists in the future as we move forward?

[Karna Gupta]

Thank you for that question.

I think in a lot of the areas in procurement, we have worked very closely with the government at the federal and provincial levels. When you look at the shared services side, it's not just procurement. A significant amount of effort is upfront planning and setting the right model that would work. That early consultation to deal with the private sector is very important.

In terms of the current leadership of the shared services, definitely that has been the approach so far, with us and with other industry groups. In that sense, I believe, at the initial starting block we're off on the right track. It needs to be followed through in terms of getting the right people around the table.

Often these consultations need to be in smaller groups. We could facilitate five or six key people from various industries to meet with the other side to engage in a dialogue where the actual knowledge transfer and information exchange is a lot more robust than having a group of 40 people sitting around, trying to figure out what to do.

Those changes are probably subtle, but I think having a smaller group in a specific case, on a specific basis, will increase the robustness of the engagement.

[Kelly Block, Conservative, Saskatoon—Rosetown—Biggar, SK]

I also want to follow up on another comment you made.

You didn't make this comment, but I've heard this in other places: “Culture eats strategy for breakfast.” You referenced the culture of an organization and the need to attend to it and to communicate it. I'm wondering when you think about the need for a shift in culture....

You also said that SSC must be able to build a team for success. I just want to give you an opportunity to comment further on that, and perhaps describe for us how an organize like SSC would start to build a team for success, paying attention to the culture or the need to shift the culture.

[Karna Gupta]

Thank you.

I won't make the supposition that the cultures are all bad—it would be a bad premise to start with—but they are different. If we start with the assumption that there is some difference in culture in different departments, it would take some amount of time within each of the department's leadership to meet with the SSC's current leadership and establish what the protocol is, how they would operate, and that there is transparency within the management, up and down. That's where a lot of time needs to be spent.

It looks like it takes a lot of time and that it will be a time sink for management, but it is necessary for future success. I can only comment on it based on, again, my personal experience. It's a tremendous amount of time sink. In my case, we had direct employees in 65 different countries, so we had to move people around from country to country to do that. It is a tremendous investment in time, and you often question the economics of it. There are no clear economics in that sense. It is, absolutely, just time with people to ensure they understand: one, what outcome you want, so there is clarity; two, how you are going to get it done, so there is a process in place; and three, how you are going to measure them, so that people are accountable to deliver what you ask them to.

[Kelly Block, Conservative, Saskatoon—Rosetown—Biggar, SK]

Do I have any more time?

[The Chair, NDP, Pat Martin]

Thirty seconds, Kelly, if you like.

[Kelly Block, Conservative, Saskatoon—Rosetown—Biggar, SK]

No, that's fine.

Thanks.

[The Chair, NDP, Pat Martin]

Thank you very much.

We will move then to the Liberals, and John McCallum.

[John McCallum, Liberal, Markham—Unionville, ON]

Thank you.

And thank you both for being with us.

You gave a timeframe of three to five years for getting a project done, but the PricewaterhouseCoopers consulting study that we've seen has more in the order of 10 to 20 years. So given the scale of this, with the 43 departments, are you disagreeing with this study? Are you saying three to five years for this specific project, with the 43 departments?

[Karna Gupta]

No. I think the projects we have done in the private sector are single-threaded. They were not 43 departments or 43 companies merging at the same time, and the scale and scope were not as large as what you're seeing here. I would not dispute the PricewaterhouseCoopers study. What I'm saying is, transformation-wise, IBM should have been larger than HP, just given the sheer size of the organization.

It is a function of the organization and a function of the type of project you're undertaking. In smaller cases, what I have gone through is three to five years.

[John McCallum, Liberal, Markham—Unionville, ON]

Okay, thank you.

Now let's take the 10 to 20 years as a given, then, and given that, presumably one needs to make some investments before one reaps the savings. So if the whole thing is 10 to 20 years, how many years do you think it would take before there would actually be net financial savings?

[Karna Gupta]

That's a very good question. I think this is where, in setting up the projects, it's very critical how you break them up. A big part of the initial stage is investment, but a big part of the downstream investment should be fed by the benefits you garner in the early years. That's why the breaking up of projects is also critical from a purely financial metric point of view.

A project is not just the work. It has to embody the financials with it. If it is generating some financial benefit in the first two, three, or four years, that net benefit should go towards the investment of future years. I don't see it as a 20-year investment and your payback starts on the 21st year, or a 10-year investment with payback in the 11th year. If I were looking at this, I would look at it in a slightly different way. How do I break it up so that at some point my surplus, so to speak, funds the future investment?

[John McCallum, Liberal, Markham—Unionville, ON]

Okay. I'm certainly not saying the net benefit would have to wait until year 21, but if we assume, for example, it's a 20-year project, you scale up the length. Given your experience, are you saying it would take until year four, or three, or six, or...?

[Karna Gupta]

I don't know in this case. I couldn't answer that question. But if it is a five-year project...for the ones we have done, I'd say by about 20% to 30% of the way in, you should starting getting some significant benefits.

[John McCallum, Liberal, Markham—Unionville, ON]

You're saying about 20% to 30%. So if it's 20 years, that would be four to six years in?

[Karna Gupta]

Yes.

[John McCallum, Liberal, Markham—Unionville, ON]

Okay.

I'm going to change the subject. We had some union people in the other day, and they naturally have an interest in minimal outsourcing. I assume your clients have an interest in maximum outsourcing. So in both cases, I think it's fair to say there's a personal interest.

Can you tell us what factors the government should consider when looking at whether to have a lot of outsourcing or a little bit of outsourcing?

[Karna Gupta]

They're probably to some degree dependent upon the outcome model. If you go to a cloud for various aspects, then a significant amount of internal management could be moved out. That's one scenario.

If you replicate what you have today, but just modernize it, then you continue to have a more skilled labour force of the same number or maybe a smaller number, because you consolidate functions. My speculation at this stage is that Shared Services Canada would probably fall somewhere in between, given the state of technology. So outsourcing would free up some employment within the government sector. The question is whether the talent at that point would be reusable for something else.

There will be a requisite change on the private sector side. I couldn't tell whether there would be a net loss or a net benefit, but most of the time this kind of generates a little bit of a net benefit on both sides.