Government Operations Committee on Oct. 17th, 2017

[Erin Weir, NDP, Regina—Lewvan, SK]

I'm certainly not suggesting that IBM is evil—

[Voices]

Oh, oh!

[Dan Murphy]

No, not at all.

[Erin Weir, NDP, Regina—Lewvan, SK]

—but should it not perhaps have broken the project down into...?

[Dan Murphy]

Don't get me wrong: no, IBM is extremely smart, but that's the environment they're in.

[Erin Weir, NDP, Regina—Lewvan, SK]

Sure, but I guess I'm wondering, should IBM not have broken down the Phoenix project into smaller components that were conducive to agile?

[Dan Murphy]

Oh. No, because they were asked—although maybe they did on their side—to respond to a tender. There are mandatories and there are desirables. They want the lowest cost per point, right? They have to play the procurement game. But if you put in agile...if you do that, you're betting the farm on those bids. So don't bet the farm. Implement in increments, and as you go along the procurement people might learn a few more things as well.

Industry right now, through ITAC, is reaching out to procurement, to work in collaboration with procurement, through some implementations to figure out how they can do procurement better so that it's a win-win, and everybody gets a fair shot.

Again, we're looking at what outcomes are being set for procurement. What are the outcomes that procurement wants to get? Is it the lowest price or is it the best value? In a lot of cases, you'll get the lowest price but you won't get the best value. There's a big difference.

[Erin Weir, NDP, Regina—Lewvan, SK]

Given a choice between contracting out in these small increments versus having an agile team working within government, what would be the better model, or would it depend on circumstances?

[Dan Murphy]

I think you can do both. I think you can contract out. You can use existing procurements and buy people, or you can say, “We're trying to get this outcome,” and you can write a procurement around that outcome and then implement it in increments. You can say, “We want you to come in, but we have an out every 90 days. We're going to have three vendors and you're going to go for the first 90 days. If you're pathetic, we're going to go to vendor number two.” Vendors might push back a bit and say, “Well, you're pathetic on the government side”, and you could have some of that. But I think this would work a lot better and it would mitigate risk.

The stakes aren't as high. It's not $50 million, where stakes are really high. If the stakes are $100,000, maybe in the long term, if I perform as a vendor, I will get that big prize over time. But I'm only going to get $250,000 in the first 90 days.

[The Chair, Conservative, Tom Lukiwski]

Thank you very much.

Colleagues, I was planning to suspend around 12:30 to go into committee business. If there's still interest, we perhaps have time for two more five-minute interventions.

Mr. Whalen, would you like to take a stab at firing out a question?

[Nick Whalen, Liberal, St. John's East, NL]

Absolutely.

It sounds as though, if you try to get yourself into an agile mindset in order to implement agile itself, you'll have to do that using an agile approach. If people are used to doing $40-million contracts, getting them down to $1-million capped contracts can't happen overnight. It's going to take some type of an approach.

When you're adopting agile to cultural change like this, what types of percentage change are we looking at doing in each iteration? Are we looking at a situation where, for example, we're going to shorten the length by 5% every quarter for the next 10 quarters, or are we going to reduce the dollar value by 20% every quarter? How much do you bite off in terms of growing the teams? How many teams are you going to try to build, and what team rate of growth are we looking at in order to implement this type of a cultural shift across government? How does that look?

[Dan Murphy]

All of that stuff comes out in the retrospectives at the end of the quarter. We do an assessment at that point of how things went.

Effectively, it's the same planning process we've always had, except we're going back to revisit inside the release every two weeks. At the end of the release, you're going back every three months to revisit your planning, to check if your assumptions were correct.

In terms of that planning process, they have a thing in agile called “velocity”, which means how fast are we moving? How much are we providing? How much value are we creating? The team's getting a sense as they go along of how much they can create. They'll get to a certain point where they'll find their steady state, and then it becomes very predictable. We know this team can produce about this much every quarter, and if everybody's happy with that we can keep going; or we consider whether there is anything else we can do to produce a bit more, or to do less. But it's not driven top-down; it's driven by the team. It's not the top saying, “Now you have to do it two weeks quicker”—

[Nick Whalen, Liberal, St. John's East, NL]

Dan, is it driven by the team or is it driven by the data that arises out of their performance?

[Dan Murphy]

Both. Effectively what you're creating is a discovery and learning organization that's building capability, and credibility and capacity internally. It's not about getting an outside contractor to do all this. You really need to build this capability inside the government.

[Nick Whalen, Liberal, St. John's East, NL]

I guess maybe related to this, one of the things I found surprising about the SSC approach and the Phoenix approach is that it's really sole-sourced. With such a broad organization, you'd like to see a competitive spirit help drive some success. Does agile allow us through its metrics to approximate or to create some type of a cultural competition or a culture of excellence, or does it provide opportunities for actual competition where other people can more frequently come in and bid on projects? How does that interplay work in agile?

[Dan Murphy]

You can absolutely create an environment of competition, but again, I'd go back to this: Is value your number one driver? Do you want value or do you want competition? Does competition drive value? Yes. Okay, so let's do more competition.

You can always set it up and structure it. If I have to be in an environment where I have to have competition, because that's part of my mandate, that's the world I live in, okay, then we do it. However, too often the outcome is driven by internal requirements versus the value that we want to drive for citizens.

A citizen says, “I want to get my cheque.” You say you can't do that, because you have to be competitive. Well, maybe you do, but underlying the real thing you're trying to do is create value for the citizen through program delivery. If you stay focused on that, then everything else comes into perspective underneath it. If you lose track of that, then competition or the lowest price becomes the most...then you're going to run into trouble, I think.

[The Chair, Conservative, Tom Lukiwski]

You have less than a minute left.

[Nick Whalen, Liberal, St. John's East, NL]

Is there an opportunity for people who work in government already to take this two-day course that Minister Brison is taking? For instance, how do I take it?

[Dan Murphy]

It's through ITAC. You can contact ITAC. They have the stuff up.

It's important to understand this, and there is a lot of stuff going on from an education point of view on how to run an iteration and how to run a team and get the team dynamics, but for the leadership part across the industry, there is a bit of a gap there. It's still maturing. That's why we want to come forward with this through ITAC, so it's industry and not representing one vendor but all kinds of vendors.

Regardless, just dive into it and understand it. If you take the course, great.

[Nick Whalen, Liberal, St. John's East, NL]

Thank you.

[The Chair, Conservative, Tom Lukiwski]

Finally, Mr. McCauley.

[Kelly McCauley, Conservative, Edmonton West, AB]

Thanks again.

You really summed it up when you talked about how we focus on internal requirements, not outcome. I'm sure as long as we have all the boxes ticked, if the bus drives off the cliff it doesn't matter—as long as we get the boxes ticked.

[Dan Murphy]

Right.

[Kelly McCauley, Conservative, Edmonton West, AB]

I can imagine how difficult it must be to change just the whole attitude and the way we think within government. I want to go back to, say, Estonia or the U.S. or Britain. Are they approaching it as a department-by-department focus, or is every department trying to tackle a small bit internally?

[Dan Murphy]

Government in the U.K., from what I've seen, is focusing on the initiative or at the project level, and that's good. Anything is good, because we're learning.