Volledige transcriptie van When the paperwork breaks down the building

00:05:04.240 60 different logins, 30 different departments. 00:05:04.319 --> 00:05:14.000 And it's not unusual for a citizen to have to access many different sites just to change their address as they move, which means they're keeping track of multiple passwords. 00:05:14.079 --> 00:05:19.920 And when I learned this, it sounded pretty bad, but maybe we should feel even worse for public servants. 00:05:20.399 --> 00:05:30.720 I've heard that a single marine inspector at Transport Canada might need 20 different logins to do their job, each with different copies of the same data. 00:05:30.879 --> 00:05:31.519 Is that true? 00:05:31.839 --> 00:05:36.240 Yes, Alistair, that's true, and it's our reality, unfortunately. 00:05:36.399 --> 00:05:42.800 So and and it shows why modernization and more specifically digitization is so important. 00:05:43.439 --> 00:05:47.839 So in Marine, right now, each program has its own system. 00:05:48.000 --> 00:05:51.839 So it means that, yes, different login for each of them. 00:05:52.000 --> 00:06:06.079 So through modernization, uh our goal is really to come up with fewer logins for sure, less duplication, more time focused on the real inspection services uh that we have. 00:06:06.399 --> 00:06:15.360 So of course it's a challenge because we are currently dealing with more than 50 legacy systems that are not talking to each other. 00:06:15.680 --> 00:06:17.680 So that's our reality. 00:06:17.920 --> 00:06:19.920 You are you are absolutely right. 00:06:20.079 --> 00:06:24.720 It's uh it's difficult to work in that kind of legacy environment right now. 00:06:24.959 --> 00:06:34.240 But with modernization, uh, I think we uh we are in a good position to start replacing those old systems with new ones. 00:06:34.480 --> 00:06:40.560 But the challenge is also the fact that we cannot just replace an old system by a new one. 00:06:40.720 --> 00:06:43.439 We will have to do it piece by piece. 00:06:43.600 --> 00:06:59.600 So as soon as we are decommissioning some functionality of a legacy system and building new applications, so this is where we will find finally some light at the end of the tunnel in many, many years, but it's a building block type of approach. 00:06:59.920 --> 00:07:04.079 Yeah, I mean you can't just turn off the ports for a few days while you install the new software, right? 00:07:04.560 --> 00:07:05.199 Absolutely not. 00:07:05.839 --> 00:07:12.959 Can you tell me a little about like these decisions to have many separate systems with different logins, made sense at the time. 00:07:13.120 --> 00:07:16.079 Like the people that did this were making a sensible decision. 00:07:16.240 --> 00:07:18.160 Can you explain how that happened? 00:07:18.480 --> 00:07:22.480 Well, Marine Safety and Security is a long-standing organization. 00:07:22.720 --> 00:07:27.839 Like the Canadian Passport story you shared with us in the previous podcast. 00:07:28.079 --> 00:07:36.079 So, for example, one of our service, such as the

uh vessel registry, have rules that go back many, many years. 00:07:36.319 --> 00:07:51.680 So, 30 years ago, each program developed their own tool to meet their own business needs, reflecting their organizational uh structure and sometimes and also the regulations that they are responsible for. 00:07:51.920 --> 00:08:02.240 So this resulted in dozens and dozens of systems that basically accomplish different goals, but also replicate features, data, and processes. 00:08:02.399 --> 00:08:08.560 So today we have inherited a patchwork of separate systems that are not talking to each other. 00:08:08.639 --> 00:08:15.120 And as a result, yes, we have multiplied the access point for our inspectors, for example. 00:08:15.279 --> 00:08:20.319 So that's the legacy challenge uh we are trying to solve now through modernization. 00:08:20.480 --> 00:08:30.160 So hoping that we will be successful in harmonizing processes, consolidating our data, building shared components as well. 00:08:30.240 --> 00:08:32.399 It's another approach that we are looking at. 00:08:32.559 --> 00:08:38.240 And as a result, we will have a strong foundation which will enable transformation. 00:08:38.559 --> 00:08:45.440 So I guess 30 years ago, it made a lot of sense to like buy a turnkey system for each service or program, right? 00:08:45.600 --> 00:08:47.519 You didn't want a vendor to sell you half of something. 00:08:47.759 --> 00:08:49.279 Like, no, no, give me a thing that works. 00:08:49.440 --> 00:08:52.559 Give me the computer, the operating system, the application. 00:08:52.799 --> 00:08:54.080 This is what it's for. 00:08:54.320 --> 00:08:58.639 But nobody expected the citizen to be accessing their data directly. 00:08:58.799 --> 00:09:06.960 Like we had this idea that you would pick up the phone and call someone who was sitting in front of a computer, or that you would go to a Service Canada kiosk and do something there. 00:09:07.200 --> 00:09:15.600 Nobody envisioned this world where every single Canadian or almost every Canadian is walking around with a device in their pocket that can connect to computers. 00:09:15.759 --> 00:09:23.519 And they also didn't really expect government employees to need a single set of data or to work across systems. 00:09:23.679 --> 00:09:30.320 Like these are things that seem obvious today, but we have to go back to 30 years ago and recognize that we couldn't anticipate that. 00:09:30.399 --> 00:09:37.600 Nobody 30 years ago knew what this world would look like with phones in our pockets and employees working across a citizen's entire portfolios. 00:09:37.759 --> 00:09:42.480 When did Transport Canada start to realize that this approach had become a a liability? 00:09:42.799 --> 00:09:53.759 So uh I would say that in 2019, for the first time we were successful in getting funding, and we decided to launch two major modernization initiatives. 00:09:53.919 --> 00:10:00.159 So, and this is where we saw okay, maybe that's our chance, that's an

opportunity to do things differently. 00:10:00.399 --> 00:10:07.200 So the two initiatives that got funding was the vessel registration and the certification services for seafarers. 00:10:07.519 --> 00:10:11.919 So you talk a bit about the uh the three-phase approach that we have in place. 00:10:12.000 --> 00:10:24.960 We'll talk more later, which means that we start with uh the process and then the data, and later on we are looking at digitization of the interaction with external stakeholders. 00:10:25.200 --> 00:10:35.919 So when we start with the first uh phase, which is the analysis of the business processes, both teams realized that basically they had the same needs. 00:10:36.159 --> 00:10:43.919 They needed to digitize the management of the workload, and they also needed to manage the documents that are submitted by the to the program. 00:10:44.240 --> 00:10:50.720 So it was making no sense that we had two teams building the same tool at the same time. 00:10:50.879 --> 00:10:56.000 So we decided in in uh to split the work between each team. 00:10:56.240 --> 00:11:05.759 So one would take the responsibility of designing and building the workload management application and based more on the common components approach. 00:11:05.919 --> 00:11:15.039 And then the other one would do the same thing, building a common component that is scalable for the document management type of application. 00:11:15.440 --> 00:11:17.679 So you got each you split it up into two teams. 00:11:17.759 --> 00:11:18.000 I have to do that. 00:11:18.240 --> 00:11:24.720 One was dealing with the workload for the software, and the other one was dealing with the the actual management of the documents in that workload. 00:11:25.039 --> 00:11:25.440 Exactly. 00:11:25.519 --> 00:11:29.600 So it took two years to develop both because we developed it from scratch. 00:11:29.759 --> 00:11:51.200 And then two years later in 2021, each team basically were you interchange their component and then they scale it to their specificity of their services, and we were able to work with the same tool essentially, but we had we have adapted it to their business uh profile. 00:11:51.519 --> 00:11:58.559 So if I understand this correctly, it's like saying you and I are gonna build a house and we both need plumbing and we both need electricity. 00:11:58.720 --> 00:12:03.600 So we decide that I'm gonna make my house electric and you're gonna plumb you do the plumbing on your house. 00:12:03.679 --> 00:12:08.240 And then when we've got them working, we go, hey, here's how to plumb a house, here's how to make it electric. 00:12:08.399 --> 00:12:14.480 And now we both have plumbing and electricity in half the time it would have taken if we'd both had to invent it ourselves. 00:12:14.639 --> 00:12:16.960 Plus, the plumbing and electricity is standardized. 00:12:17.039 --> 00:12:19.279 So like I can fix your plumbing, you can fix my plumbing. 00:12:19.360 --> 00:12:21.440 If you make your plumbing better, mine gets better too. 00:12:21.600 --> 00:12:22.639 Is that a fair analogy? 00:12:22.879 --> 00:12:26.799 Yes, we can take this uh because the foundation basically is the same. 00:12:26.960 --> 00:12:40.320 It's just a question of deciding, okay, what what color will be the the fixture or what or the type of uh I don't know that the the the different uh sis electric systems and so on, like you said. 00:12:40.720 --> 00:12:49.279 But of course it it was a sort some sort of a risk because we knew that in doing this it would take longer to deliver. 00:12:49.679 --> 00:13:16.399 But in the long run, it would accelerate our broader transformation effort in marine safety and security because in having those common components that are built to meet a certain business capabilities, then we knew that any other marine services ready for transformation will have to just onboard on this new common component instead of building from scratch again in again. 00:13:16.960 --> 00:13:22.480 Okay, so first benefit, if I want to build a third house, I already know how to do the plumbing and the electricity. 00:13:22.720 --> 00:13:27.360 Second benefit, if I want to build a heating system, now I can install heating in all the houses. 00:13:27.600 --> 00:13:31.840 Third benefit, everyone can buy the same lamp and it'll just work with the electrical system. 00:13:32.000 --> 00:13:35.120 So this is kind of the standardized components that you're talking about. 00:13:35.360 --> 00:13:36.879 Yeah, it's like a plug and plate. 00:13:36.960 --> 00:13:39.279 It's like the building block approach. 00:13:39.519 --> 00:14:01.120 I have a component, so to meet a certain business capability, and then I'm taking it, I'm tailoring it to my specific needs, and then I can also through an API use another one that will do, let's say, document management, and then I can use another one through an API, which is issuing electronic certificates, for example. 00:14:01.360 --> 00:14:09.360 So in this way has many services and marine safety and security as the same type of process in place. 00:14:09.519 --> 00:14:11.519 We are all doing basically the same thing. 00:14:11.679 --> 00:14:20.559 So we are managing service requests, we are managing documents submitted by clients, we are issuing certificates, we are inspecting. 00:14:20.720 --> 00:14:21.919 So it's all the same. 00:14:22.080 --> 00:14:29.600 So why designing and building the same component many, many times for a specific program instead? 00:14:29.840 --> 00:14:39.120 Like look at this in a more holistic way and build a product that fits the needs of a business capability instead. 00:14:39.440 --> 00:14:39.600 Sure. 00:14:39.679 --> 00:14:42.720 And modern software is made out of building blocks like this, right? 00:14:42.799 --> 00:14:52.720 When I make a website, I might embed a Google form, I might add Stripe for payment, I might include some third-party video player in the web page instead of building all those things

from scratch. 00:14:52.960 --> 00:15:00.879 So these kinds of common components are like reusable pieces for invoicing or payment or competency verification. 00:15:01.039 --> 00:15:10.000 And I've heard that in the government of Canada, there are some common components such as forms and notify, which are standards that you can use to build on. 00:15:10.080 --> 00:15:12.080 Have those caught on within government departments? 00:15:12.399 --> 00:15:12.879 Absolutely. 00:15:13.120 --> 00:15:18.559 So yes, we are leveraging the common components built by the Canadian Digital Service. 00:15:18.799 --> 00:15:24.960 So they have the G what they call the GC Notify and also the GC Forms components. 00:15:25.120 --> 00:15:29.360 So instead of reinventing the wheel, we use those components. 00:15:29.519 --> 00:15:39.919 And again, as a plug and play or Lego type of building block, we are just using those components to add it to our end-to-end process. 00:15:40.320 --> 00:15:46.320 We had a speaker, David Eves, who wrote something recently about, you know, commodity components are the key to digital sovereignty. 00:15:46.480 --> 00:15:51.679 This idea that if you have your own stack of these components, it's really important for building government acts. 00:15:51.840 --> 00:15:55.200 I want to pivot back to something a little related to this. 00:15:55.519 --> 00:16:09.440 In 2017, there was this thing called the Service Fees Act, which basically said that every fee that a department charges must be publicly listed, along with the service standards, which say things like if you pay us for this service, we'll do this thing within this many days. 00:16:09.679 --> 00:16:17.360 And importantly for this discussion, the department also has to report on how often it hit or missed those standards and sometimes pay for missing them. 00:16:17.600 --> 00:16:24.799 Most government departments are funded by taxes, not fees, so there's less of a connection between what the citizen gets and whether the department gets its budget. 00:16:24.879 --> 00:16:30.159 But for Transport Canada, many of the fees, the things you do charge fees, right? 00:16:30.320 --> 00:16:35.679 Like many of the things that you charge for, like a permit for driving a boat, for example, you have to charge a fee. 00:16:35.759 --> 00:16:42.879 And as a result, you have to track processing times to measure your performance, you have to track your costs to justify the fees, and you have to issue refunds. 00:16:43.120 --> 00:16:49.440 As a department that collects fees from its citizens, you probably have to report more about what your services are doing. 00:16:49.600 --> 00:16:55.840 So across all your services, did you build common components that made it possible to do this kind of collection? 00:16:56.080 --> 00:16:56.960 Yes, absolutely. 00:16:57.120 --> 00:17:03.360 The department built a common component of what we call the cost recovery service management solution. 00:17:03.600 --> 00:17:15.119 Essentially, is to meet the requirements under the Service Fees Act in terms of reporting on program cost, program revenue, performance of our service standards. 00:17:15.359 --> 00:17:22.160 And also there's a piece under the Service Fees Act which talk about the remission policy, essentially. 00:17:22.400 --> 00:17:34.400 So what that means is if we are missing a service standard, we will have to remit a portion of the fee paid by the client to the client, essentially. 00:17:36.400 --> 00:17:43.519 So just so I I'm clear, for a fee, government departments have to refund you if they don't deliver the service they said they would in the time they said they would. 00:17:43.759 --> 00:17:44.240 Correct. 00:17:44.400 --> 00:17:49.119 So this is what we call the remission policy attached to the Service Fees Act. 00:17:49.200 --> 00:17:55.440 So we have to remit a portion of the fee paid by the client if we are not meeting our service standards. 00:17:55.599 --> 00:18:03.599 But this has to be tracked and reported on an annual basis to Parliament through a fee report that is tabled annually. 00:18:03.839 --> 00:18:12.880 And then from there we need to track those data with regard to our cost recovery services or all the services with fees. 00:18:13.119 --> 00:18:14.880 That's an obligation that we have. 00:18:15.039 --> 00:18:25.759 And for for doing so, there's a common component, yes, that was developed by our group here in Service and Digital to do to perform those specific functions. 00:18:25.920 --> 00:18:41.759 And again, it's a it's a common capability that uh any service with fees in Transport Canada can use and plug and play to, in my case, the workload management application that we build up front because it will feed. 00:18:42.000 --> 00:18:44.960 So it's part of the end-to-end process essentially. 00:18:45.119 --> 00:19:03.039 When I'm getting a request for delivering a service, if if there's a fee associated to this service, I will have to plug with the cost recovery service management solution uh up to the invoicing of the service, but also the reporting to parliament. 00:19:03.359 --> 00:19:03.599 Wow. 00:19:03.839 --> 00:19:10.880 So if if you didn't have this service in place, then humans would have to track all this data, and that's a lot of extra work, right? 00:19:11.039 --> 00:19:13.039 So having software do it makes more sense. 00:19:13.200 --> 00:19:27.920 But instead of just getting each service that you offer to have its own tracking, you said to each service, if you use this component, we will automatically track everything from the moment the service is requested to the moment it's delivered, how much we charge, and all that kind of stuff. 00:19:28.079 --> 00:19:35.039 And then your system just rolls that up at the end of the year and puts it into a document to parliament that makes you accountable to the citizens who use the service. 00:19:35.359 --> 00:19:35.759 Exactly. 00:19:36.319 --> 00:19:37.200 That sounds amazing.

00:19:37.519 --> 00:19:38.400 It is amazing. 00:19:38.559 --> 00:19:43.440 So uh our team in IT did amazing work in developing this component. 00:19:43.519 --> 00:19:46.319 And the beauty of this is it's all automated. 00:19:46.559 --> 00:20:07.599 So as soon as I have the data into my workload management application, so I can feed the cost recovery service management solution right away, and all those calculations and tracking of spent on doing XYZ help me assess well uh reporting on the program cost, for example. 00:20:07.759 --> 00:20:17.920 And then since I I am invoicing my clients or collecting the fees up front, I can have right away the revenue generated by this service. 00:20:18.079 --> 00:20:31.440 And then boom, it's uh move into a platform that is uh also has the capacity the ability to report back to Parliament and prepare this uh beautiful report uh for our uh taxpayers. 00:20:31.759 --> 00:20:35.359 And it sounds like, you know, this is the government says you must report this once a year. 00:20:35.599 --> 00:20:40.480 But if I understand your technology, in theory, you can just go at any time if you had the right query. 00:20:40.559 --> 00:20:42.720 You could just go say, How did we do today? 00:20:42.880 --> 00:20:47.839 Or, you know, you can you can get more granular because you've collected the data in a proper structured way. 00:20:48.160 --> 00:20:48.960 Yes, it's true. 00:20:49.119 --> 00:20:52.000 So uh we can have the information in real time. 00:20:52.160 --> 00:20:54.880 So this is why it's a management tool as well. 00:20:55.039 --> 00:21:00.400 So we can create Power BI reports for on a monthly basis to see progress made. 00:21:00.640 --> 00:21:03.440 Are we on track to meet our service standards? 00:21:03.519 --> 00:21:10.720 That's information that is useful for management as well to make to make some correction in the delivery of the service if needed. 00:21:10.880 --> 00:21:16.160 So uh of course we have uh this information at our finger uh tips right now. 00:21:16.319 --> 00:21:18.240 So it's a huge improvement. 00:21:18.559 --> 00:21:22.880 I mean, I don't want to sound too excited, but it sounds like what digital government's supposed to do. 00:21:23.119 --> 00:21:32.319 Let me ask you one more question about the motivations of why you changed, because obviously simplifying these systems was important, complying with the act so you could collect fees was important. 00:21:32.559 --> 00:21:38.960 I guess digital services are also more efficient, not just in collecting stuff, but simply like digitizing things. 00:21:39.119 --> 00:21:45.759 You mentioned 60,000 physical files stacking up in a building that you told me about before. 00:21:46.000 --> 00:21:52.160 Can you tell me what it feels like to have 60,000 files in a building and what that does to your will to modernize things? 00:21:52.480 --> 00:21:55.279 Well, I'm seeing this as an opportunity, actually. 00:21:55.519 --> 00:21:59.680 So uh of course, we had some issue with the infrastructure. 00:22:00.160 --> 00:22:01.680 Of the building at some point. 00:22:01.920 --> 00:22:04.960 There was no room anymore to stack more files. 00:22:05.359 --> 00:22:07.680 So it's a physical building full of documents. 00:22:07.839 --> 00:22:08.079 Exactly. 00:22:08.319 --> 00:22:10.400 And it was just literally the building is full. 00:22:10.799 --> 00:22:18.880 The floor was too heavy with all those files, so the building infrastructure was unable to support it anymore. 00:22:19.039 --> 00:22:32.720 Which means that we were we were looking for another floor or even another area, another room to store all the medical files that we had accumulated over the years. 00:22:32.880 --> 00:22:41.279 But the beauty of it is at the same time we got the funding, as I mentioned earlier, to modernize the Seafarer services. 00:22:41.440 --> 00:22:58.400 And this is where we start our journey of modernizing all those paper-based processes, which means that we had to take one by one those 60,000 files, scan it, and move it into our workload management application. 00:22:58.559 --> 00:23:12.160 So it's it's incredible all the work that was done over time because uh we are not imaginating so that kind of effort that is required to digitize. 00:23:12.319 --> 00:23:14.160 But again, it's our reality. 00:23:14.240 --> 00:23:17.359 It's not like you said, building a flashy portal. 00:23:17.440 --> 00:23:24.960 It's really get rid of the paper base and make sure that our staff can work in a digital environment. 00:23:25.039 --> 00:23:27.359 And it's starting by scanning files. 00:23:27.680 --> 00:23:29.359 Yeah, it does seem to be the boring stuff. 00:23:29.440 --> 00:23:32.799 Like Lisa Fast talked about she spent six months getting the form right. 00:23:32.960 --> 00:23:35.200 But these boring things can have tremendous impact. 00:23:35.279 --> 00:23:43.759 But you know, when you go from I need another floor because the the floor is about to collapse from the weight to oh, you can have the floor back, it's all digital now. 00:23:43.920 --> 00:23:45.839 Um Yeah, now it's empty. 00:23:46.240 --> 00:23:48.720 What's it like to scan 60,000 files? 00:23:49.039 --> 00:24:03.519 Well, it's well, at the beginning, and when you start a transformation process, so I remember when we start this one, so I was entering in the room and there was piles of files on desk with post-its right up on top of it. 00:24:03.680 --> 00:24:16.000 It was okay, that's the pile for uh that we just received, that's the pile for approval, that's the pile for incomplete files that we need to do some back and forth with the client, and so on. 00:24:16.160 --> 00:24:18.720 There was files all over the place in the room. 00:24:18.880 --> 00:24:25.279 So, and then on top of that, we have the vertical cabinet with all the medical files, the 60,000. 00:24:25.599 --> 00:24:37.680 So it was quite overwhelming at the beginning, but we said since we start by the big by the

beginning with the UX designers, we look at process step by step. 00:24:37.839 --> 00:24:40.559 Okay, so it was all on post-it, easy. 00:24:40.720 --> 00:24:43.599 So, but now where are the duplications? 00:24:43.759 --> 00:24:47.119 Is there approvals that are not necessary anymore? 00:24:47.279 --> 00:24:54.000 Is there any way where we can simplify the process to better align with a digital environment? 00:24:54.160 --> 00:25:05.839 So all those questions and pain points were identified at that at that time to come up with a recommendation of a simplified process for digitization. 00:25:06.000 --> 00:25:47.920 And this is where we build the famous workload management application, where it was necessary if we wanted to move the scanned files into a system, we needed that because as soon as you have a CFARE who apply for a medical uh certificate, we took the opportunity to pull the files from the uh physical cabinet, scan it, included in the workload management application with the service request uh of the CFAR, and then boom, slowly it took two years to scan all the 60,000 files, but we did we did it based on each request coming in. 00:25:48.079 --> 00:25:51.200 So, because you cannot boil the ocean all at the same time. 00:25:51.359 --> 00:25:58.559 So we took it one by one, and finally at the end uh uh of the journey, we were able to be fully digital. 00:25:59.039 --> 00:26:03.279 I love that I got someone from the Marine Division telling me we can't boil the ocean, that seems perfect. 00:26:03.519 --> 00:26:12.319 So if I'm clear here, you went in there and looked at the post-it notes and the process, and you recognized, oh, there's a process here called the form is incomplete. 00:26:12.400 --> 00:26:17.440 I have to talk to someone, or there's a process here that says this has to be renewed or this has expired or whatever. 00:26:17.680 --> 00:26:23.680 And you looked at the whole process and you said, what things in this process don't make sense in the digital world anymore? 00:26:24.240 --> 00:26:26.960 Like, okay, there's a physical signature here, how do we change that? 00:26:27.039 --> 00:26:28.240 Or like what needs to be changed, right? 00:26:28.319 --> 00:26:29.680 So you have the process, right? 00:26:29.920 --> 00:26:32.319 Because you may already know what the process is or whatever. 00:26:32.559 --> 00:26:35.359 Then you go, okay, now we have this new process. 00:26:36.240 --> 00:26:39.119 You didn't just stop and scan 60,000 files. 00:26:39.279 --> 00:26:48.160 You went, I'm going to come up with a way of migrating from the analog world to the digital world that doesn't necessarily create more work. 00:26:48.319 --> 00:26:54.079 It's just that the next time I need to touch a physical thing, that's when I will enter it into the digital thing. 00:26:54.240 --> 00:26:57.680 Instead of kind of having everybody come in on weekends until you've scanned everything. 00:26:57.759 --> 00:26:57.839 Right. 00:26:58.079 --> 00:26:58.319 Correct. 00:26:58.480 --> 00:26:58.640 Correct. 00:26:58.799 --> 00:27:04.079 So even part of the process is the process for gradually migrating from analog to digital. 00:27:04.160 --> 00:27:07.440 And at the end of two years, you've kind of gone through all the files at least once. 00:27:07.519 --> 00:27:09.759 And so now you'd incorporated them, right? 00:27:10.000 --> 00:27:11.039 Yeah, that's correct. 00:27:11.279 --> 00:27:19.680 Because also given the fact that we had 60,000 files, we have some C Fairs in there that was probably 120 years old. 00:27:19.920 --> 00:27:20.240 Wow. 00:27:20.480 --> 00:27:22.720 So why scanning those files? 00:27:23.039 --> 00:27:23.519 What happened? 00:27:23.839 --> 00:27:28.480 I mean, you've got these historical records that sound like they should be in a museum, but they're just sitting in a warehouse. 00:27:29.039 --> 00:27:29.680 Exactly. 00:27:29.839 --> 00:27:37.839 So it becomes like more uh archives than physical files that are necessary to deliver a service as of today. 00:27:38.000 --> 00:27:47.119 This is why we decided, okay, as soon as we uh we are adding a request, we will do the search of the physical file and then digitize everything. 00:27:47.279 --> 00:27:51.359 Otherwise, we wouldn't we are probably today still scanning files. 00:27:51.759 --> 00:27:57.759 So at the end of it, you can kind of look around and all the files you have left are like retired mariners and retired seafarers. 00:27:58.160 --> 00:28:05.279 We dispose of them in a secure way because there's uh, as you know, life a lifespan of those files. 00:28:05.359 --> 00:28:13.359 Yeah and but uh there's a what there was a lot of uh files in there uh that were not relevant anymore. 00:28:13.759 --> 00:28:17.119 Wow, I didn't know this was gonna be like a maritime history session, but this is fascinating. 00:28:17.200 --> 00:28:21.039 Okay, so you mentioned earlier process, then data, then implementation. 00:28:21.119 --> 00:28:26.160 I think I understand you start with the process because you've got to understand how to modify it for digital. 00:28:26.319 --> 00:28:32.559 Then you start with the data because the data happens in the context of the process, and then you start with the actual implementation. 00:28:32.720 --> 00:28:34.400 What happens if you do those in the wrong order? 00:28:34.640 --> 00:28:35.920 Well, wrong order. 00:28:36.000 --> 00:28:40.319 It's really any group can decide which order they would like to take. 00:28:40.559 --> 00:28:43.440 So in our case, we built a strategy. 00:28:43.599 --> 00:28:53.039 I set the vision for marine safety and security, and we decided that it was making better sense for us to digitize processes first. 00:28:53.440 --> 00:28:56.000 Because let's say I'm starting the opposite. 00:28:56.160 --> 00:28:58.559 I'm starting with building a flashy portal. 00:28:58.720 --> 00:29:05.680 So your client will submit their files electronically through my portal, but what happened in on my

side? 00:29:05.839 --> 00:29:15.200 I will print the files, I will probably put it in a f in a cabinet somewhere, and then I will do my processing all manual. 00:29:15.680 --> 00:29:18.079 So we said that doesn't make any sense. 00:29:18.319 --> 00:29:20.880 We need to equip our staff first. 00:29:21.440 --> 00:29:36.640 We need to improve our digital literacy in working in a digital environment, and this also is taking a lot of time, and this is why we have the change manager working very closely with those teams to help those people. 00:29:36.799 --> 00:29:43.920 Well, they it's managing resistance because we were changing their way of working from analog to digital. 00:29:44.079 --> 00:29:56.319 So that being said, uh we prefer we prefer the approach of equipping our staff to work in a digital environment and then look at the data, as I mentioned. 00:29:56.480 --> 00:30:06.079 Once we have those tools in place, we are able to extract information from the workload management application, but also from the other systems. 00:30:06.400 --> 00:30:18.400 And before using it for, let's say, uh an online platform or even just visualization type of products, we have to make sure that we trust those data as well. 00:30:18.559 --> 00:30:25.039 And we trust the infrastructure that we have built with all the cybersecurity uh requirements and so on. 00:30:25.359 --> 00:30:28.480 So there's a this is why data is second. 00:30:28.640 --> 00:30:33.680 So as soon as we are in a position to to produce data, we want to trust them. 00:30:33.839 --> 00:30:36.960 So we need to make sure that we understand what they are. 00:30:37.200 --> 00:30:43.279 So we built a dictionary, like uh uh the La Rosse dictionary or Le Petit Robert. 00:30:43.680 --> 00:30:47.599 We have to define each of those data to understand it. 00:30:47.759 --> 00:30:50.960 And then also data is is an asset, right? 00:30:51.119 --> 00:30:55.279 It's not specific to the a program uh needs necessarily. 00:30:55.359 --> 00:31:03.359 So it has to be available and accessible to any program in Transport Canada in need of this information to do their own work. 00:31:03.519 --> 00:31:11.119 So, which means that we move all the data into a centralized repository that we call the sort of data lake. 00:31:11.279 --> 00:31:29.119 And this is where we are uh able to manipulate or transform the data to produce algorithm, a Power BI report, uh some application or whatever you want, or even uh AI, uh artificial intelligence. 00:31:29.279 --> 00:31:31.759 So, like uh let's say a chat bot. 00:31:31.920 --> 00:31:37.519 So, but if I have no trust in my data, I will be facing garbage in, garbage out. 00:31:38.000 --> 00:31:50.079 And just so people understand about the the shared stuff, if I know who the seafarer is and their medical history, and then there's the boat coming into a port, I might want to link to here comes the boat, and this is the guy steering it. 00:31:50.240 --> 00:31:58.079 So I now connect two things that were different records, but they're much more valuable when they're together because you can pick the health of the guy steering the boat, for example. 00:31:58.559 --> 00:32:00.720 Um it sounds pretty esoteric. 00:32:00.799 --> 00:32:06.160 And before I get further, I want to point out that the term data lake is actually a term from data science. 00:32:06.240 --> 00:32:10.400 It's not the transport candidates naming data structures after itself, right? 00:32:10.799 --> 00:32:11.839 From the marine world. 00:32:12.000 --> 00:32:12.079 Yeah. 00:32:12.400 --> 00:32:12.640 Yeah. 00:32:12.799 --> 00:32:14.400 Just not everything's marine world. 00:32:14.559 --> 00:32:18.000 Data lake was a data scientist thing, not a transport canada thing. 00:32:18.079 --> 00:32:19.279 You're happy to be using the lake. 00:32:19.599 --> 00:32:19.759 Okay. 00:32:20.720 --> 00:32:21.599 Let's make this concrete. 00:32:21.680 --> 00:32:24.720 Give me a concrete example, something like vessel registration. 00:32:24.880 --> 00:32:28.319 How did you transform a process like that with this approach? 00:32:28.720 --> 00:32:33.599 Remember that I mentioned the piles of uh files in the room and so on. 00:32:33.759 --> 00:32:38.400 So the first thing we are doing, so we set up a uh a multidisciplinary team. 00:32:38.559 --> 00:32:51.920 So maybe our family, the audience will be familiar uh with a scrum team essentially, uh including uh developers, a product owner, a technical advisor, maybe a specialist in data. 00:32:52.160 --> 00:33:04.400 But on top of that, in marine safety and security, we decided to add two other roles to that team, which is the UX designer and researcher as well as a change manager. 00:33:04.559 --> 00:33:14.240 And that's the the first time I'm not really aware of any other group doing that, having a change manager at the table with the developers. 00:33:14.480 --> 00:33:15.759 So that's the first step. 00:33:15.920 --> 00:33:28.799 And then, as mentioned, based our on our vision, which is the three-step approach, we have we start with the process first, second is the data, and third, we are developing the uh the tools. 00:33:29.039 --> 00:33:48.240 So, what's that mean is in the case of the marine medical unit, we start reviewing all the processes with the UX researchers to identify where were our pain points duplication in our processes in order to first introduce the workload management application. 00:33:48.640 --> 00:34:02.559 And from there, once the staff with the support of the change manager is used to work with the new workload management application, we are in a position now to extract data from it. 00:34:02.720 --> 00:34:08.880 And the data we are extra extracting from the workload management application is twofold. 00:34:09.119 --> 00:34:15.360 The first one is will help the program better understand the performance of the program itself. 00:34:15.519 --> 00:34:19.599 So for decision making, it's uh absolutely helpful.

00:34:19.679 --> 00:34:49.440 And the second thing is also the fact that the the workload management application provides us the ability to see the process and phases, which means that I can collect data on each phase of the process, and this allows me to see where are the bottlenecks in my process, which means that if I have, let's say, uh 3,000 files uh stacked on at the approval level, maybe have a problem with my approval process. 00:34:49.519 --> 00:35:09.199 So may uh so I will nail down and dig a little bit further with the UX researcher, understand what's exactly the problem over there, and put the resources accordingly, maybe adjust my roadmap accordingly and focus really on that specific bottleneck to bring some more efficiency uh in my process. 00:35:09.360 --> 00:35:16.639 So this comes back to what what Lisa Fast was saying, that like you can't know what the world is like when you start trying to fix it. 00:35:16.800 --> 00:35:22.320 And so the only thing you can do is get information very quickly to learn what the next big priority is. 00:35:22.480 --> 00:35:27.039 With passports, it turns out it was the form, and they went down that problem, and then it was like what documents to bring. 00:35:27.280 --> 00:35:29.440 In your case, you're like, okay, I go look at the process. 00:35:29.599 --> 00:35:33.280 It turns out it's a bottleneck at the approval, so I should change my priorities there. 00:35:33.440 --> 00:35:53.760 That seems in very stark contrast to the kind of waterfall planning model where people go, this is what I'm gonna do, and I'm gonna give you the full five-year plan when I start, as opposed to your case where you're like, I know what my goal is, but I'm gonna use the data I'm collecting during the process to change my priorities and my project plan to make sure it works properly. 00:35:54.079 --> 00:35:54.800 Yes, exactly. 00:35:54.880 --> 00:36:02.239 It's sometimes the um the analogy could be uh when you do renovation, you open a wall and oops, there's a pipe. 00:36:02.559 --> 00:36:05.440 So it's same, it's the same thing in transformation. 00:36:05.599 --> 00:36:15.360 So in this case, we realize that there was a bottleneck there, but when we dig a little bit further, we realize, oops, that's a regulatory change that will be required. 00:36:15.519 --> 00:36:22.559 And this could take two, three years to happen because we have to go through the gazetteing process and so on. 00:36:22.800 --> 00:36:36.239 So sometimes we are underestimating the time it could take because we are finding some little things like this throughout the delivery of our transformation project. 00:36:36.480 --> 00:36:38.079 So we have to be patient. 00:36:38.400 --> 00:36:38.960 Yeah. 00:36:39.199 --> 00:36:47.760 Um, so what things can a marine inspector or a boat owner or a seafarer do today that they couldn't do five years ago when you started on this project? 00:36:48.079 --> 00:36:50.480 Well, a lot of things has changed in five years. 00:36:50.639 --> 00:36:54.960 Today, since let's we started this journey in 2019. 00:36:55.039 --> 00:37:01.199 Uh since then, we have the cargo inspection program that now is fully digital from end to end. 00:37:01.280 --> 00:37:14.400 So inspectors are uh managing the service request for inspection up to the payment of fee and the delivery of the certificates all online, all digital. 00:37:15.039 --> 00:37:27.599 Uh vessel owners can now uh receive their registration and insurance certificates electronically because we have a common component that supports that specific uh common capability. 00:37:27.840 --> 00:37:33.360 Uh, we have also the Canadian voters who have the Pleasure Craft Operator card. 00:37:33.920 --> 00:37:43.199 Uh they can use our online tool to search for a course provider if they have losed their card or need a replacement. 00:37:43.360 --> 00:37:53.760 So uh on the marine medical side of the house, we have now the marine medical examiners who are able to submit their medical reports on seafare to through an online portal. 00:37:53.920 --> 00:37:56.000 This uh was not possible in the past. 00:37:56.159 --> 00:37:57.599 And I can continue now. 00:37:57.679 --> 00:38:01.920 But all those changes mean that we can provide the service faster. 00:38:02.159 --> 00:38:05.840 We have less paperwork to deal with, uh for sure. 00:38:06.000 --> 00:38:13.840 And it's more time for our inspector and staff to really focus on what matter the most, safety and security of the transportation system. 00:38:14.159 --> 00:38:23.119 It sounds like you're like if I didn't know I was talking to Transport Canada, I would think I was talking to one of the other countries that I'm always talking to about how they fixed it. 00:38:23.199 --> 00:38:27.360 Like these sound like the kind of things that that we hear from other countries. 00:38:27.440 --> 00:38:30.639 And it's it really honestly feels quite nice to know that you're doing this well. 00:38:30.880 --> 00:38:42.159 It sounds like you're pretty far ahead, not just in terms of the services, but also with like the baseline reporting that your department needs to get to comply with the Service Fees Act and things like that. 00:38:42.400 --> 00:38:47.679 What I really want to focus on uh as we wrap this up, you've said a couple of times at a change manager. 00:38:48.000 --> 00:38:52.480 Overcoming resistance is always the organizational work of digital transformation. 00:38:52.880 --> 00:38:56.400 What does it take to get people on board with the resistance to this process? 00:38:56.719 --> 00:39:05.039 I know some of the people in Transport Canada, and I'm especially thinking of someone like Julie Lees, who used to be CIO, have traditionally been very digitally literate. 00:39:05.280 --> 00:39:14.719 So what role has your team and past executives changed taken in making digital transformation and change happen within the organization? 00:39:15.039 --> 00:39:21.760 Well, I'm glad that you have this

question because it's uh I would say that it's one of our trademark in marine safety and security. 00:39:21.920 --> 00:39:28.320 So we have dedicated change managers associated with each of our transformation initiatives. 00:39:28.400 --> 00:39:29.840 And the reason is simple. 00:39:30.079 --> 00:39:34.000 Technology alone doesn't deliver transformation. 00:39:34.239 --> 00:39:39.360 We need to focus on people because it's people who are using those tools. 00:39:39.519 --> 00:39:43.039 If if we are not supporting them, they will not use them. 00:39:43.280 --> 00:40:00.639 You can have your IT department developing the best tool in the world, but if I have no one to tell them or train them or provide them with standing operator procedures to operate in in the digital world, they will not use it. 00:40:00.800 --> 00:40:05.519 They will resist because they will not understand the new way of working. 00:40:05.679 --> 00:40:07.599 So this is why it is critical. 00:40:07.679 --> 00:40:10.320 And I I think the the statistics are there. 00:40:10.480 --> 00:40:17.440 So it's about 70% of transformation projects are failing because of a lack of change management. 00:40:17.679 --> 00:40:49.039 So uh this is why we want to make sure that any service line going through transformation will be supported by a change manager to make sure that people understand why we are doing those changes and also have the necessary uh tools and support, such as, as I mentioned, operating procedures, training, uh a SharePoint a SharePoint site where to uh find the information. 00:40:49.119 --> 00:40:52.559 It could be you need someone who is doing that work. 00:40:52.639 --> 00:40:54.239 It won't happen by magic. 00:40:54.400 --> 00:40:58.079 So it's just not a question, okay, here's the tool, deal with it. 00:40:58.239 --> 00:40:59.599 It's not happening this way. 00:40:59.840 --> 00:41:01.840 Uh you have a master's in public administration. 00:41:01.920 --> 00:41:07.519 You're not a technologist, although a lot of the words you've been saying about APIs and shared components sound pretty digital. 00:41:07.760 --> 00:41:15.760 Do you think having a background in public administration and not tech helped you or hindered you when you were trying to make this transformation happen? 00:41:16.000 --> 00:41:27.360 I would say that it helped me a lot because it it's helping me explaining the business side of the house to my IT folks that I'm working in very close partnership with them. 00:41:27.599 --> 00:41:35.440 So my master's degree essentially gave me all the courses that I needed and that and then that that I apply today essentially. 00:41:35.519 --> 00:41:48.000 I have courses in change management and business process analysis in org design, uh applying uh a Meinsberg type of approach at the time, a long time ago. 00:41:48.400 --> 00:41:51.519 Course in results-based management, data analytics. 00:41:51.760 --> 00:41:57.360 Yes, we were using Excel in 2000, but still the basics, the foundation is there. 00:41:57.679 --> 00:42:07.519 So absolutely, I think I am really on at the right place to lead this team towards transformation. 00:42:07.760 --> 00:42:31.519 But of course, all those skill sets is uh through a good partnership with the program people on the marine side, but uh also with my counterpart in service and digital group, where I have to share the vision uh of the R3 phase approach, but also to bridge the gap between the technical and the our operational realities. 00:42:31.760 --> 00:42:38.079 So I think that's the idea of service modernization and data sounds like a digital job, but you're so right. 00:42:38.400 --> 00:42:40.159 Very seldom are there technology problems. 00:42:40.239 --> 00:42:43.840 There are always people problems, process problems, sometimes data problems. 00:42:44.000 --> 00:42:47.440 I want to wrap up with a question that's a little more profound. 00:42:47.760 --> 00:42:49.039 What's still broken? 00:42:49.199 --> 00:42:50.559 Like what keeps you up at night? 00:42:50.800 --> 00:42:51.760 What's still broken? 00:42:51.840 --> 00:42:59.920 I would say that unfortunately each director, I think, and this is not only within Transport Canada, you can take it more globally. 00:43:00.239 --> 00:43:08.079 We are still designing and building the same tools that is basically doing all the same things. 00:43:08.320 --> 00:43:20.800 Many of us, not just in transport, as I mentioned before, we need to manage our workload, we need to issue certificates, we need to manage documents, all those things. 00:43:21.039 --> 00:43:25.920 We need to communicate with our um with our external clients. 00:43:26.159 --> 00:43:29.599 We need to it's all the same functionality essentially. 00:43:29.760 --> 00:43:36.400 So we have to stop, I would say, this nonsense of building the same tools over and over. 00:43:36.559 --> 00:43:39.360 Just take the cost recovery service management solution. 00:43:39.519 --> 00:43:42.719 We are not the only department having service fees. 00:43:42.960 --> 00:43:44.639 There's a lot of other departments. 00:43:44.880 --> 00:43:46.000 Health Canada has fees. 00:43:47.440 --> 00:43:51.599 Yeah, they probably built a ser a cost recovery service solution as well. 00:43:51.840 --> 00:43:57.039 But how many cost recovery service solutions we have in the federal government right now? 00:43:57.519 --> 00:43:59.519 And if we had one, then when there was one change. 00:44:00.159 --> 00:44:01.360 It everyone would benefit. 00:44:01.440 --> 00:44:08.159 Like if we added a new language or we changed the process with the regulation, then it would be one change and everybody would benefit, right? 00:44:08.400 --> 00:44:13.280 Yeah, just make it scalable to adapt it to the specific needs of a service line. 00:44:13.440 --> 00:44:15.280 The good thing is we are on the right track. 00:44:15.440 --> 00:44:25.199 Canadian digital service is starting to provide us with some common components that are reusable, that are meeting some of our needs. 00:44:25.360 --> 00:44:28.400 So I I'm confident that we are on the right track for sure. 00:44:28.719 --> 00:44:29.199 What's next? 00:44:29.280 --> 00:44:31.760 So what what's keeping me up at night? 00:44:32.159 --> 00:44:36.800 Um I would say that it's making sure we can keep moving our initiative. 00:44:37.199 --> 00:44:44.000 So modernizing marine services or any type of services is really a long-term commitment. 00:44:44.480 --> 00:44:55.920 And on an annual basis, I need to convince my senior management that it's worth to continue our effort on uh on an annual basis. 00:44:56.079 --> 00:45:00.880 So my pressure is I have to deliver and I have to show the results. 00:45:01.119 --> 00:45:06.800 But I again I have a wonderful team, very brilliant people working with me. 00:45:06.960 --> 00:45:14.000 And I feel that my team, myself, with my master degree, like you said, uh I am at the right place. 00:45:14.239 --> 00:45:20.800 But moreover, my entire team is at the right place to deliver very successful results. 00:45:21.039 --> 00:45:21.519 That's amazing. 00:45:21.679 --> 00:45:23.360 Lucy, thank you so much for your service. 00:45:23.519 --> 00:45:28.159 I'm I I kind of wish I had a boat just so I could use all these amazing services. 00:45:28.320 --> 00:45:44.400 But uh no, it's it's really nice to see someone thinking about this from first principles, from the process, from the data, from shared components and economies of scale, from thinking about the change, from uh recognizing that you don't know what the plumbing's like until you pull out the wall. 00:45:44.480 --> 00:45:45.440 So you can't make plans. 00:45:45.679 --> 00:45:50.079 You have to kind of earn the trust by making gradual improvements and being transparent about it. 00:45:50.159 --> 00:45:53.840 And so many of the things that I've heard work for other countries. 00:45:54.000 --> 00:45:57.440 Uh it's really nice to hear that they're happening in some places in the Canadian government, too. 00:45:57.760 --> 00:45:58.559 Thank you, Alistair. 00:45:58.719 --> 00:46:01.599 Thank you for this opportunity to share uh our work. 00:46:01.679 --> 00:46:03.199 Uh, I really, really appreciate. 00:46:03.760 --> 00:46:06.880 Thanks for joining us on the Functional Government Podcast.