Evaluation of Data Analytics Services
Evaluation Report
May 2025
How the report is structured
- The report in short
- Acronyms and abbreviations
- What is covered
- What we learned
- How to improve the program
- Management response and action plan
- Appendix A: Visual depiction of Data Analytics Services
- Appendix B: Key Data Analytics Services projects
- Appendix C: Evaluation questions and indicators
- Appendix D: Interview quantification scale
- Appendix E: Status of Data Analytics Services objectives
The report in short
Through Budget 2018, funding was announced to support Statistics Canada’s (StatCan’s) modernization agenda and enhance its technological statistical infrastructure through the development of Data Analytics Services (DAS). With the exponential growth in data and their use, in addition to the corresponding demands for both storage and processing, new and innovative approaches to infrastructure delivery were required. DAS is a cloud-based platform designed and developed internally to provide users with access to StatCan data, analytical tools, software and the necessary computing power to complete various analyses. DAS is intended for a wide range of external users, such as researchers, data analysts, data scientists and professionals in the public and private sectors.
Overall, DAS provides several functionalities:
- secure personal or collaborative workspaces for high-capacity computing
- high-quality, timely and trusted StatCan data uploaded directly to users’ workspaces
- rich metadata and search infrastructure to ensure that data are findable, accessible, interoperable and reusable
- state-of-the-art tools—from statistical software for familiarity and convenience to open-source software for greater agility and flexibility
- advanced capacities like artificial intelligence, machine learning techniques and high-performance data processing.
Of note, one project that uses DAS is the Virtual Data Lab (VDL). It has a wide user base and provides similar services to other DAS environments. However, the architecture is not as technologically advanced and relies on non-cloud native infrastructure, so modernization efforts are required.
DAS governance extends across the Digital Solutions Field (Field 9) and the Strategic Data Management, Methods and Analysis Field (Field 6). In 2023, under the leadership of the Chief Data Officer (Assistant Chief Statistician [ACS] of Field 6), Field 6 took on business ownership for DAS and is responsible for oversight of all DAS processes and program management. Field 9 continues as the executive budget holder and remains responsible for technology expertise. There is also the DAS Advisory Council, jointly chaired by the directors general of both fields, which reviews issue escalations, the scope, the budget and the strategic alignment of activities.
The objective of the evaluation is to provide credible and neutral information on the relevance and performance of DAS. The scope of this evaluation focused on the relevance of DAS and the achievement of its intended results, as well as considerations for continued improvement and sustainability of the platform. Value for money was also evaluated at a high level.
Key findings and recommendations
There is a continued need for DAS. It is important for federal data modernization and secure collaboration within a Protected B environment focusing on cloud infrastructure, as well as real-time access, addressing the unique needs of researchers, data scientists and policy makers. DAS is not considered duplicative of other services and has potential, with improvements in functionality and usability, to continue to meet the evolving needs of users.
However, DAS has had mixed success in achieving its intended results. While it has successfully met some objectives, such as enabling access to data and supporting collaboration, challenges such as lengthy onboarding, access delays and alignment issues with StatCan’s mandate have hindered its full potential. Internal users report higher satisfaction compared with external users, who have struggled with inefficiencies that impact project initiation and the overall user experience. While DAS has supported various projects aimed at secure data sharing and policy support, many projects are still in progress or have been discontinued, limiting the assessment of their full impacts at the time of the evaluation.
Further, DAS faces significant sustainability challenges because of funding deficits, information technology (IT) capacity limitations and declining user satisfaction. While recent improvements show promise, their impact is not yet measurable, and ongoing concerns could lead to underuse and high operational costs. More time and monitoring are needed to determine the long-term viability of DAS.
In light of these findings, the following recommendations are proposed.
Recommendation 1
The ACS of Strategic Data Management, Methods and Analysis (Field 6), in collaboration with the ACS of Digital Solutions (Field 9), should ensure that the modernization efforts for DAS and VDL are aligned and not redundant. This will support alignment with the long-term vision for DAS while also making efficient use of StatCan’s limited IT resources.
Recommendation 2
The ACS of Field 6, in collaboration with the ACS of Field 9, should seek ways to improve the experience of external users to sustain the uptake of DAS and promote sustainable costs. Based on the evaluation findings, areas of improvement should include, but not be limited to, onboarding, service cataloguing, costing models and timely data access.
Recommendation 3
The ACS of Field 6, in collaboration with the ACS of Field 9, should ensure that effective monitoring of the program is carried out and, more specifically, that
- processes are in place to track and monitor direct and indirect DAS clients to better understand the DAS client base and uptake over time
- performance indicators for DAS, such as client use and satisfaction, are established and monitored regularly
- ongoing assessments of overall program costs, efficiency and duplication of services (i.e., VDL, Advanced Analytics Workspace, Collaborative Analytics Environment), and impact for users are carried out.
Recommendation 4
The ACS of Field 6, in collaboration with the ACS of Field 9, should explore possibilities to make the functionality and technology of the DAS platform more available to a broader audience of users.
Recommendation 5
The ACSs of Field 6 and Field 9 should review the current budget arrangement to ensure that it is efficient and effective, and that it aligns with organizational policies and practices.
Acronyms and abbreviations
- AAW
- Advanced Analytics Workspace
- ACS
- Assistant Chief Statistician
- AI
- Artificial intelligence
- API
- Application programming interface
- CAE
- Collaborative Analytics Environment
- DAS
- Data Analytics Services
- DAS BO
- Data Analytics Services business owner
- FAIR
- Findable, Accessible, Interoperable and Reusable
- FRDC
- Federal Research Data Centre
- FSDH
- Federal Science DataHub
- GAE
- Geospatial Analytics Environment
- GPU
- Graphics processing unit
- IT
- Information technology
- LoA
- Letter of agreement
- PPE
- Personal protective equipment
- RDC
- Research data centre
- SDMX
- Statistical Data and Metadata eXchange
- SSC
- Shared Services Canada
- StatCan
- Statistics Canada
- VDL
- Virtual Data Lab
What is covered
1. Background
Through Budget 2018, funding was announced to support Statistics Canada’s (StatCan’s) modernization agenda and enhance its technological statistical infrastructure through the development of Data Analytics Services (DAS). With the exponential growth in data and their use, in addition to the corresponding demands for both storage and processing, new and innovative approaches to infrastructure delivery were required. DAS is a cloud-based platform designed and developed internally to provide users with access to StatCan data, analytical tools, software and the necessary computing power to complete various analyses. DAS is intended for a wide range of users, such as researchers, data analysts, data scientists and professionals in the public and private sectors.
DAS allows users to combine StatCan data with external datasets (i.e., client-owned or publicly available data) to create more detailed datasets, while maintaining strict screening and security protocols. These datasets can be stored securely in the cloud for users to access remotely and conduct various analyses collaboratively to generate outputs like tables, charts and data visualizations. The overall goal of DAS is to foster collaboration on data-driven projects, enhance the user experience when accessing StatCan data and increase the relevance of StatCan to key users.
The platform is partially funded through a cost-recovery modelFootnote 1 in which fees are based on the scope, complexity and size of a given project. These fees cover costs such as salaries, licences and cloud services. Cloud costs are paid to Microsoft Azure, and they vary depending on how much a client uses the platform.
The DAS platform offers four distinct environments to users, depending on their needs and expertise (Figure 1). Users can also combine functionalities across the different DAS environments.

Description - Figure 1 Available Data Analytics Services environments
Figure 1 provides an overview of four distinct environments designed to support various levels of data analytics and statistical data management. Each environment is tailored to meet the needs of different user groups, ranging from beginners to advanced users.
- Collaborative Analytics Environment (CAE):
- Provides a drag-and-drop experience that enables users at all skill levels to develop quick insights.
- Offers familiar Microsoft suite of analytics products (e.g., MS Power BI, Data Bricks, Azure Machine Learning, Synapse, and DevOps).
- Advance Analytics Workspace (AAW):
- Increases power and flexibility to process, analyze and visualize data for more advanced users.
- Offers a leading free open-source suite of analytics products such as JupyterLabs, R, Python, R Shiny, Kibana, and Kubeflow.
- Geospatial Analytics Environment (GAE):
- Enables users to integrate geospatial components in analysis and visualizations.
- Offers leading free open source and proprietary geospatial analytics products such as ESRI, ARCGIS, and Notebooks.
- Statistical Data and Metadata eXchange (.Stat SDMX):
- Provides an open-source platform for the efficient production and dissemination of high-quality statistical data.
Overall, DAS provides several functionalities:
- secure personal or collaborative workspaces for high-capacity computing
- high-quality, timely and trusted StatCan data uploaded directly to users’ workspaces
- rich metadata and search infrastructure to ensure that data meet the Findable, Accessible, Interoperable and Reusable (FAIR) principlesFootnote 2
- state-of-the-art tools—from statistical software for familiarity and convenience to open-source software for greater agility and flexibility
- advanced capacities like artificial intelligence (AI), machine learning techniques and high-performance data processing.
DAS also provides various support services, such as operations and monitoring, data ingestion, access to StatCan data (e.g., microdata, protected data), solution engineering, onboarding coaches, and sandbox services, to aid users in accessing and using DAS. For more information about DAS, please refer to Appendix A for a depiction of its various environments, inputs, services and outputs.
Various projects have leveraged DAS environments to achieve different objectives, such as increased collaboration and access to data. Appendix B shows the variation through a list of sample projects. Of note, one project that uses DAS is the Virtual Data Lab (VDL), which is an early legacy version of the Collaborative Analytics Environment (CAE) launched in 2021. It provides users with the infrastructure and tools they need to remotely and securely leverage StatCan confidential microdata, an alternative to research data centres (RDCs) or the Federal Research Data Centre (FRDC). It has a wide user base of 38 sponsoring organizations and 375 users, as well as efficient onboarding and access because of its stricter governance processes and more straightforward service offerings and use cases. While VDL provides advanced security and a locked-down environment, the architecture is not as technologically advanced as other DAS environments and relies on non-cloud native infrastructure.
The onboarding process
Clients begin the onboarding process by engaging with the DAS team to provide a description of their needs. A solution, costs and timelines are agreed upon, and then the project is initiated on the platform. While the timeline for onboarding differs depending on project complexity, Figure 2 outlines the typical journey for a DAS user from project initiation to decommissioning. Various changes have been made to the onboarding journey since the initiation of DAS—Figure 2 depicts its current state.

Description - Figure 2 Data Analytics Services onboarding process
Figure 2 illustrates the four stages of the DAS onboarding process, detailing the steps and interactions involved from initiation to decommissioning.
- Initiation: In the initiation stage, clients access the DAS portal to submit an application. The Intake Crew engages with the client, triages the request, and gathers the necessary requirements.
- Administration and Governance: During the administration and governance stage, clients are provided with a proposed solution, cost estimate, and timeline. The project undergoes a comprehensive review of data governance, privacy, and ethics in collaboration with the Office of Privacy Management and Information Coordination and the Data Ethics Secretariat.
- Onboarding Support: Once a service agreement is signed, the solution is developed and implemented. Clients receive project management as well as ongoing IT support and maintenance. StatCan subject matter experts are also available to clients for methodological support.
- Decommissioning: In the decommissioning stage, following the end of the project, the data and environments are deleted, accounts are revoked, and a client satisfaction survey is sent.
Governance
DAS governance extends across the Digital Solutions Field (Field 9) and the Strategic Data Management, Methods and Analysis Field (Field 6). In 2023, under the leadership of the Chief Data Officer (Assistant Chief Statistician of Field 6), Field 6 took on business ownership for DAS, and it now resides under the Centre for Statistical and Data Standards. Field 6 is responsible for oversight of all DAS processes and program management under the DAS business owner (DAS BO) team. Field 9 continues as the executive budget holder and remains responsible for technology expertise.
There is also the DAS Advisory Council, jointly chaired by the directors general of both fields. Members of the council include various directors from Field 6 (including from the Office of Privacy Management and Information Coordination, and the Data Access Division), Field 9, subject-matter divisions, the Statistical Information Service and the DAS BO team. The council reviews issue escalations, the scope, the budget and the strategic alignment of activities.
Evolution of Data Analytics Services
Since its initiation in 2018, DAS has undergone several important changes. Below is a brief overview, focusing on key developments that are relevant to the evaluation.
- 2018: DAS initiation
DAS was initiated in 2018, at first to address the data analytics needs of the StatCan data science community and, later, to partner on pilot projects with other federal departments. Field 9 was both the business owner and the executive budget holder responsible for building the platform. - 2020: Impact of COVID-19
The pandemic significantly impacted DAS, accelerating the use of the Advanced Analytics Workspace (AAW) and CAE. For example, one project enabled collaborative analysis to inform decisions about the availability of personal protective equipment (PPE). - 2023: Project closeout and business ownership transfer
As the DAS project concluded, an important gap was identified: the lack of a strategic plan for managing DAS as a program moving forward. In May 2023, the Chief Statistician appointed Field 6 as the new business owner for DAS. Field 9 continued to be the executive budget holder and remained responsible for technology expertise. - 2024: Strategic planning and DAS refinement
After the transfer of business ownership, several key changes were made to enhance the effectiveness, efficiency and sustainability of DAS:- Plans were made to transition StatCan users to more efficient internal platforms (i.e., The Zone and DAS alternative), with DAS focusing on external users (e.g., other federal departments; provincial, territorial and municipal governments; and private sector).
- Service offerings were streamlined by reducing customization for users, easing pressures on StatCan information technology (IT) specialists.
- Users whose needs did not align with StatCan’s mandate were precluded from the DAS platform.Footnote 3
- A new letter of agreement (LoA) was created between Field 6 and Field 9, and it outlines the roles and responsibilities for both fields, including budgetary allotments.
The evaluation acknowledges the evolving nature of DAS and has taken recent changes into consideration for the recommendation themes.
2. About the evaluation
Authority
The evaluation was conducted in accordance with the Treasury Board Policy on Results and StatCan’s Integrated Risk-Based Audit and Evaluation Plan 2024/2025 to 2028/2029.
Objective and scope
The objective of the evaluation is to provide credible and neutral information on the relevance and performance of DAS.
The scope of this evaluation, determined in collaboration with various key stakeholders (i.e., Field 6, Field 9 and senior management), focused on the relevance of DAS and the achievement of its intended results, as well as considerations for continued improvement and sustainability of the platform. Value for money was also evaluated at a high level by assessing the extent to which DAS demonstrated program relevance and performance, and through user perceptions related to the cost of the service and of the platform.
The evaluation work was conducted from September 2024 to January 2025.
Approach and methodology
The following three evaluation questions were identified:
- To what extent is there a continued need for DAS?
- To what extent has DAS achieved its intended results?
- To what extent is DAS sustainable in its current state?
More information about the evaluation questions and related indicators can be found in Appendix C.
The data collection methods outlined in Figure 3 were used. The findings outlined in this report are based on the triangulation of these data collection methods.

Description - Figure 3 Data collection methods
Figure 3 outlines the methods used by the evaluation for data collection.
- Interviews with DAS Users: Semi-structured interviews were conducted with 15 external DAS users, including individuals from other federal government departments, researchers, and policy analysts. Additionally, 14 internal DAS users, who are employees of Statistics Canada, were interviewed.
- Interviews with Program Representatives: Semi-structured interviews were also conducted with 15 program representatives and partners within Statistics Canada.
- Document Review: A review of Statistics Canada's documents was carried out, including the summary survey data provided by the program.
Four main limitations were identified, and mitigation strategies were employed, as outlined in Table 1.
| Limitations | Mitigation strategies |
|---|---|
| Self-report bias can occur in interviews, where individuals reporting on their own activities may portray them in a more positive light. | To the extent possible, feedback and reflections on activities were sought from a range of perspectives. A review of program documents also supported a balanced perspective. |
| Given that the timing of this evaluation coincided with significant program planning and restructuring, it was challenging to evaluate these efforts because not enough time had passed. | A review of current successes and challenges, as well as program efforts to leverage and address them, was conducted. Recommendations centre around outstanding or additional efforts needed to address ongoing gaps and limitations. |
| It was challenging to identify external users. When they were identified, most who were interviewed had not accessed or used Data Analytics Services (DAS) or declined to be interviewed. This made it challenging to fully evaluate user impacts. | Several external projects had internal subject-matter leads who were able to be interviewed. Internal and external users’ access, satisfaction and impacts were examined to the extent possible. However, ongoing performance measurement and evaluation will be needed to assess success and impacts on external users moving forward. |
| There were highly technical and complex financial aspects of DAS. These, in addition to other uncertainties, were challenging to contextualize and required clarification. | Several informal meetings were held with Field 6, Field 9 and other Statistics Canada stakeholders throughout the evaluation to clarify various technical, financial and management-related components of DAS and provide important context for the findings. |
What we learned
1. Relevance: Continued need
To what extent is there a continued need for DAS?
There is a continued need for DAS. It is important for federal data modernization and secure collaboration within a Protected B environment focusing on cloud infrastructure, as well as real-time access, addressing the unique needs of researchers, data scientists and policy makers. DAS is not considered duplicative of other services and has potential, with improvements in functionality and usability, to continue to meet the evolving needs of users.
DAS aligns with federal- and agency-level priorities with respect to data modernization, accessibility and collaboration. To further align with agency-level priorities, user eligibility was recently updated to exclude external users whose projects fall outside StatCan’s mandate.
At the federal level, DAS aligns with key priorities set out in the Data Strategy for the Federal Public Service, including creating whole-of-government data initiatives, facilitating secure data sharing, and supporting digital transformation and cloud computing.
At the agency level, DAS aligns with the StatCan Data Strategy and modernization agenda, particularly in regard to the following:
- digital transformation and IT modernization (e.g., cloud enablement; leveraging of AI and machine learning; digital workplaces; cutting-edge tools for acquiring, processing, integrating and analyzing data)
- collaborative data management and partnerships with diverse internal and external stakeholders, such as federal departments, provincial and territorial governments, academia, Indigenous organizations, and others (e.g., sharing data, strengthening national statistical systems, developing integrated approaches for data collection and analysis, breaking silos)
- broad data access (e.g., access to Protected B data and anonymized microdata) while maintaining a rigorous, transparent process that upholds privacy, ethics and legislative requirements (e.g., DAS Advisory Council and oversight for data acquisition and management)
- administrative-data-first approach and leveraging of data ecosystems (e.g., open data).
However, alignment issues with StatCan’s mandate were identified, which changed the way that DAS could be used. According to interviewees’ interpretation of the Statistics Act, which sets out StatCan’s mandate, DAS should not be used solely as an IT infrastructure or data server for external users. Instead, DAS should be used to enable external collaboration to enrich or add value to collective statistical outputs, support the production of official statistics and outputs, and support StatCan’s role in national coordination of data. External users whose needs did not align with this interpretation of StatCan’s mandate were recently precluded from the DAS platform. For example, external users who did not need to leverage StatCan data were no longer able to use the DAS platform for their analytical needs.
Some program interviewees noted that Shared Services Canada (SSC) may be better suited to host a broader IT infrastructure or data server for external users because this fits its mandate of delivering enterprise-wide digital programs and services. This would allow for the precluded projects to leverage the technology that has been developed and avoid duplication of efforts.
There is increasing demand for secure, collaborative environments for data analytics, with a particular focus on cloud infrastructure, real-time access and data governance. While DAS aimed to meet these needs, many users faced significant challenges that led to delays or the search for alternative solutions. Despite the perceived future potential of DAS, concerns over its functionality and usability persisted among users.
Based on a review of federal and agency strategies, as well as interviews, there is a demand among researchers, data scientists and policy makers for secure, collaborative and efficient environments for handling data, with a particular focus on cloud infrastructure, real-time access, and data governance and security. This is particularly relevant given the rapid rise of cloud computing and cloud-based analytical platforms globally. Generally, DAS is viewed as a powerful and innovative tool that can support these needs because of cloud enablement; collaborative workspaces; access to StatCan data and expertise; and strict security, privacy and access protocols.
However, program documents and interviews showed that users faced technical and governance-related hurdles (e.g., lengthy onboarding, costing issues, platform instability, lack of beginner-friendly guidance), which led to project delays or the search for alternative solutions (these challenges are outlined further in the Performance subsection). Overall, while the theoretical aspects of the platform were seen as strong among most interviewees, functionality and usability were still in question for most of them. This was particularly true for interviewees who were external users and for smaller-budget projects or projects not requiring complex capabilities (e.g., producing basic descriptive statistics).
While several other services offer similar analytical capabilities to users, the unique features of DAS demonstrate that these services are not duplicative. DAS provides added value for specific use cases, such as supporting federal employees who require external collaboration within a Protected B environment or assistance from StatCan experts, as well as non-federal employees who require access to data, support and collaboration.
Several other services were identified during the interviews that offer analytical capabilities that are similar to those of DAS, including the following:
- SSC’s Federal Science DataHub (FSDH): This collaborative cloud-based platform for federal scientists has data infrastructure and analytical solutions using a self-service model. The FSDH was identified by most federal user interviewees as an alternative platform that they considered.
- StatCan’s The Zone and DAS alternative: This cloud-based platform for internal StatCan employees offers services similar to DAS’s AAW and CAE platforms, with some improvements. As a result, internal StatCan employees will no longer use DAS for generic or routine functions. Program interviewees identified that these platforms were modelled on DAS, noting that development and implementation were more efficient as a result.
- Private companies: These are cloud-based data infrastructure and analytical solutions through Microsoft, Amazon, Google, etc.
However, when comparing these options with DAS, it was determined that these services were not duplicative because of several unique value-added features, as depicted in Figure 4.

Description - Figure 4 Value added of Data Analytics Services
Figure 4 compares the value-add of DAS with other platforms, including Shared Services Canada’s Federal Science DataHub (SSC FSDH), StatCan The Zone and DAS Alternative, and private companies, across four criteria: external collaboration, protected B, StatCan data access, and StatCan support.
- External collaboration
- SSC FSDH: No
- StatCan The Zone and DAS Alternative: No
- Private companies: Yes
- DAS: Yes
- Protected B (Protected B is a security level for sensitive information and assets in Canada. It refers to information that, if compromised, could cause serious injury to an individual, an organization or a government)
- SSC FSDH: No
- StatCan The Zone and DAS Alternative: Coming soon
- Private companies: Yes
- DAS: Yes
- StatCan data access
- SSC FSDH: No
- StatCan The Zone and DAS Alternative: No
- Private companies: No
- DAS: Yes
- StatCan support
- SSC FSDH: No
- StatCan The Zone and DAS Alternative: No
- Private companies: No
- DAS: Yes
More specifically, DAS provides the following value added:
- External collaboration: DAS is accessible by researchers, data scientists and policy makers outside the federal government. No other federal service allows access for and collaboration with these types of external users.
- Protected BFootnote 4 environment: DAS is a Protected B environment, which allows access to a greater volume and type of protected data for external collaboration. No private company can offer external users this type of environment.
- StatCan data access: DAS provides users with remote access to confidential StatCan microdata and, in some cases, access to pre-release StatCan data. It gives access to a greater volume of data and to more types of data. No other federal service or private company can offer users these types of StatCan data.
- StatCan support: Users have access to subject-matter experts, methodological resources and the DAS team to support their projects. Some user interviewees indicated that this was very helpful for answering methodological questions and identifying platform customization needs. Comparatively, the FSDH will be self-serve.
A few other options were noted by external user interviewees to support their analytical and data needs, such as legacy platforms within other federal departments and RDCs and the FRDC, but it was acknowledged that these systems were not comparable to DAS in terms of analytical capabilities, access and collaboration. A few interviewees also mentioned the unique geospatial capabilities of the Geospatial Analytics Environment (GAE), compared with other services. However, GAE is a newer component of DAS, compared with CAE and AAW, so fewer interviewees were able to speak to this environment.
2. Performance: Achievement of intended results
To what extent has DAS achieved its intended results?
DAS has had mixed success in achieving its intended results. While it has successfully met some objectives, such as enabling access to data and supporting collaboration, challenges such as lengthy onboarding, access delays and alignment issues with StatCan’s mandate have hindered its full potential. Internal users report higher satisfaction compared with external users, who have struggled with inefficiencies that impact project initiation and the overall user experience. While DAS has supported various projects aimed at secure data sharing and policy support, many projects are still in progress or have been discontinued, limiting the assessment of their full impacts at the time of the evaluation.
While DAS achieved most of its original objectives, some are still in progress and one was not realized because it was deemed out of scope.
There were 25 objectives outlined for DAS when it was initiated. Most of the objectives were achieved, while some are still in progress and one was not realized because it was deemed out of scope. The following summarizes the key takeaways:
- Fully achieved (15): DAS provided a comprehensive platform that integrated the user experience, advanced data management and analytics. It supported AI and machine learning, enabled big data analysis through advanced computing, and leveraged virtualization for efficient data delivery and advanced storage. DAS provided open-access metadata, facilitated business workflows and ensured secure access through strong authentication. It improved access to StatCan data through pipeline engineering and offered remote capabilities. The platform also supported analysis with open-source tools and provided scalable, agile infrastructure for diverse user needs.
- Partially achieved (9): DAS fostered collaboration in algorithm and data sharing with Git-based features and AI chatbots but is still working on external user version control via Azure DevOps. Auditing and reporting goals were partially met, with completion expected by the next fiscal year. A data discovery function was deployed, and there is a need for work to continue on developing a data discovery navigator. Data access for the broader community is ongoing, with the data catalogue needing further development. DAS supported machine-to-machine data exchange but faced challenges with integration and a steep learning curve. Geospatial data capabilities are enhanced, though full integration with the Federal Geospatial Platform is pending. While governance improvements are underway, platform infrastructure and analysis support have been partially achieved, with cost constraints and budget issues preventing full realization. Information governance is expected to be realized in 2025/2026. Finally, DAS had secure data lake architecture in place, with ongoing work to improve secure sharing.
- Not realized (1): Provenance and lineage management was deemed out of scope for DAS because it has been included as a component of the Target Enterprise Architecture.
A list of these original objectives and their status can be found in Appendix E.
Since its inception, a variety of external and internal users have accessed or requested access to the DAS platform, with most external users using VDL. About one new intake form for DAS is submitted per month by external users (not including VDL). However, the disparate systems within DAS made it challenging to fully understand user uptake.
According to the summary data provided by the program, there were 769 DAS users (Figure 5). About half (55%) of these users were external to StatCan, primarily including those who accessed VDL. There were also DAS users from three other federal government organizations (i.e., the Treasury Board of Canada Secretariat, Health Canada and the Public Health Agency of Canada), one municipality and one university.
Internal users (45%) included two StatCan areas (i.e., the Artificial Intelligence (AI) Methods Division [formerly named the Data Science and Innovation Division], and the Centre for Population Health Data) and seven StatCan programs (i.e., VDL, the Census of Population, the Census of Environment, the Oral Health Statistics Program, internal trade, the Canadian Centre for Energy Information and the Statistical Geomatics Centre). At the time of the evaluation, these internal users had begun to be transitioned to StatCan’s The Zone.

Description - Figure 5 Number of external and internal users of Data Analytics Services and the Virtual Data Lab (n=769)
Figure 5 presents the number of external and internal users of DAS and VDL in a column chart.
- External Users are comprised of 45 DAS users and 375 VDL users
- Internal Users are comprised of 349 DAS users
In addition, summary data provided by the program suggest that there are ongoing interest in and demand for DAS. For example, over a six-month period in 2024/2025, seven new intake forms were received from external users for DAS (averaging about one new intake form per month for CAE, AAW or GAE). VDL demand also increased and is expected to continue growing, with a 30% gain over three years.
More importantly, there is currently no reliable method to monitor demand and uptake. Information comes from a variety of sources, causing difficulties in understanding access and use. The DAS team is currently working on developing linkages to better understand who is accessing the platform.
There were mixed satisfaction levels among DAS platform users (not including VDL users), with internal users generally reporting higher satisfaction than external users. However, assessing overall satisfaction was challenging, and there is a need to collect better user experience data moving forward.
Internal and external user interviewees were asked how satisfied they were with the DAS platform on a scale from very dissatisfied (1) to very satisfied (5). Five user interviewees did not provide a response, and the remaining 24 interviewees—12 internal users and 12 external users—provided the following average ratings regarding their satisfaction with DAS.

Description - Figure 6 Data Analytics Services user satisfaction rating
Figure 6 presents the user satisfaction ratings for both DAS internal users and external users in a doughnut chart. The Internal User Satisfaction Rating is 4.3 out of 5 whereas the External User Satisfaction Rating is 3.3 out of 5.
User interviewees who gave positive satisfaction ratings were primarily internal users, but a few were external users (early adopters of DAS, in particular). These users appreciated the improved onboarding process, effective communication from the DAS team and that DAS met its intended purpose. However, almost all user interviewees noted issues with cost, system bugs and delays in data access. External user interviewees involved in the PPE project during the COVID-19 pandemic expressed high satisfaction with DAS. However, it was also noted that after the pandemic, there were significant barriers to accessing DAS because of a change in requirements (i.e., projects not requiring access to StatCan data were precluded from DAS because of misalignment with StatCan’s mandate).
External user interviewees were the most dissatisfied, often seeking alternative solutions for their projects. More information about the gaps and limitations experienced by external users is provided in the section below, but generally, there were delays in being able to use DAS (e.g., onboarding delays, delays in accessing data).
VDL users were not included as interviewees in the evaluation because a separate evaluation was previously conducted for this component. Overall, program interviewees noted that there was higher satisfaction among VDL users. This is because VDL can provide more efficient onboarding and access as a result of stricter governance processes and more straightforward service offerings and use cases.
Overall, assessing user satisfaction with DAS was challenging because most external user interviewees either had not yet gained access to DAS or were waiting for more data before proceeding with their projects, so their experiences were limited. Unfortunately, external users were the most important group to assess because they will be the focus of the platform moving forward (internal users will use The Zone and DAS alternative). Ongoing work is needed to assess the satisfaction of these users.
Further, the program provided summary survey data, which highlighted mixed satisfaction levels among users in both 2022 and 2024. However, these data were not comparable because the 2024 data did not include responses from external users and there were significantly fewer responses, compared with the data from 2022. Moving forward, there is an opportunity to collect more diverse survey data from users to assess satisfaction with DAS.
Several limitations of the DAS platform were highlighted, with the most significant being the lengthy onboarding process. This delay hindered project initiation and impacted users’ planning and efficiency, particularly for external users. There were also challenges with access and use.
Program documentation and interview data highlighted several gaps and limitations of the DAS platform, including challenges with onboarding, access and use, precluded projects, and data stewardship. The following is a summary of the key gaps and limitations:
- Onboarding: Program documentation and interview data frequently cited onboarding delays as one of the biggest challenges in accessing DAS, especially for external users. According to summary survey data from 2024, about half of users reported being neutral, unsatisfied or very unsatisfied with the onboarding process. Program interviewees noted that onboarding was dependent on the complexity of the project (e.g., six to eight months for complex projects, three to six weeks for less complex projects). In one case, users for a very complex project had been waiting for over two years to finalize their LoA. Onboarding issues not only delayed project initiation for new users but also affected their planning and efficiency, with some users deciding not to continue with DAS. Importantly, program interviewees noted that when the program has streamlined its service offerings, onboarding times should improve because it will be clearer which projects can proceed and which require further assessment. It is also worth noting that VDL had a separate onboarding process that was identified by interviewees as being faster than DAS because of VDL’s stricter governance processes and straightforward service offerings and use cases.
- Access and use: Program documentation and interview data also identified several challenges when using the DAS platform. Some of the more commonly identified issues included platform instability (e.g., bugs, downtime, crashes, server disconnect), access problems (e.g., username and password issues, onerous authentication processes, delays in getting data), insufficient support for technical questions (for some users), steep learning curves and a lack of beginner-friendly guidance and documentation, issues with costing estimates, and graphics processing unit space issues (e.g., data-heavy or geospatial analysis) and space issues for new clients to access VDL (i.e., VDL is reaching its capacity). Further, it was noted in program documents and interview data that DAS (notably the AAW environment) may be overly complex for the average use case (e.g., running descriptive statistics).
- Precluded projects: As noted earlier, in 2024/2025, program and user key interviewees identified that several DAS projects were precluded because they did not align with StatCan’s mandate. User interviewees indicated that they are now trying to recreate a similar analytical platform within their own federal department and highlighted the inefficiency of this process for the Government of Canada. Others expressed frustration with the time and effort lost during planning and onboarding.
- Data stewardship: Some interviewees noted that DAS successfully addressed data access, sharing, security and privacy and is now progressing with standards since moving to Field 6. However, some other interviewees noted limitations to data stewardship because of implementation changes and budget constraints, such as lacking optimization for cloud use, promoting data silos and not addressing data classifications.
User projects had a range of objectives, including creating new ways to cooperate, enabling secure data ingestion pipelines for data sharing, providing access to more data, supporting policy decisions through multisector data sources and producing official statistics. However, many projects were ongoing or in progress at the time of the evaluation, and several others were discontinued by the user or cancelled by the program, making it difficult to assess the impacts of using DAS.
Based on program documents and interviews, projects using DAS that were carried out by external users had a range of different objectives. Figure 7 outlines these key objectives, along with example projects and their status at the time of the evaluation (i.e., not onboarded or awaiting access, currently in progress, or completed). This is a non-exhaustive list of projects that have used or are using DAS; they were chosen to highlight the themes of the key objectives.

Description - Figure 7 Objectives of Data Analytics Services projects
Figure 7 outlines five key objectives of DAS along with an example of DAS project for each objective and their status at the time of the evaluation (i.e., project not yet onboarded, project in progress, or project completed).
- Creating new ways of cooperation
- The SafeTO project brings together multiple stakeholders to leverage multi-sectoral data sources and will help expand the definition of community safety beyond crime statistics and/or enforcement to include prevention and well-being.
- Project not yet onboarded
- Enabling data ingestion pipelines
- The Vehicle Registration project and the Price Analytics Environment project enable the ingestion of large data sets for processing and analysis.
- Project in progress
- Providing access to more data
- The Business Data Lab project offers real time data and interactive tools to help Canadian businesses effectively navigate the business market.
- Project in progress
- Supporting policy decisions
- The PPE project created a dashboard at the beginning of the pandemic to ensure that essential supplies were allocated to areas in greatest need.
- Project completed
- Producing official statistics
- The AgZero project has Agriculture and Data Science Divisions of StatCan as well as external partners using CAE, AAW, and GAE to enable Data Science to eventually develop machine learning models and produce official statistics.
- Project not yet onboarded
It is important to note that many of the projects examined were in progress at the time of the evaluation, making it difficult to assess the impacts of using DAS. Further, several other projects were discontinued by the user or cancelled by the program and, therefore, had no reported impacts. Ongoing monitoring of projects and an assessment of their impact will be needed.
3. Efficiency and sustainability: Current state
To what extent is DAS sustainable in its current state?
DAS faces significant sustainability challenges because of funding deficits, IT capacity limitations and declining user satisfaction. While recent improvements show promise, their impact is not yet measurable, and ongoing concerns could lead to underuse and high operational costs. More time and monitoring are needed to determine the long-term viability of DAS.
With the transfer of business ownership in 2023/2024, DAS implemented changes to address key issues concerning the onboarding process, a lack of strong program management and governance, and technical capacity issues. Moving forward, the effectiveness of these efforts should be assessed on an ongoing basis.
Program documentation and interview data revealed several key issues and challenges with DAS prior to the transfer of business ownership in 2023/2024, such as onboarding delays, a lack of strong program management and governance, and technical capacity issues. Most of these were identified as stemming from the quick rollout of DAS during the pandemic, with a lack of business ownership involvement. This led to insufficient preparation for managing a program and providing services to clients effectively and efficiently. Table 2 outlines the challenges that DAS has experienced and recent improvements that have been implemented to address them.
| Key issue | Challenges | Recent improvements |
|---|---|---|
| Onboarding |
|
|
| Program management |
|
|
| Governance |
|
|
| Capacity |
|
|
Most of the improvements to DAS were completed within the past year and are too recent for an assessment of their effectiveness. While the improvements seem to be aligned with the challenges, they should be assessed on an ongoing basis moving forward to determine the impact on the overall efficiency and sustainability of DAS.
At project closeout in 2023, DAS had a deficit of $1.8 million. Since then, funding issues have persisted, with Field 6 being unfunded for its work on DAS. However, because of recent efforts, the program is now projecting financial sustainability for 2024/2025, and funding has been allocated to Field 6 through a new LoA. A review will be necessary to determine whether the new financial arrangements are sustainable in the long term.
Through federal resources and signed LoAs with users, DAS had funding of approximately just over $40 million from 2018/2019 to 2022/2023. According to the project closeout report in February 2023, the DAS project finished with a $1.8 million program deficit, including
- a $1.14 million deficit generated by the cloud run costs
- a $0.5 million deficit in salary for on-strength employees working for unfunded core DAS activities
- a $163,000 deficit for the unfunded Enterprise Information and Data Management Project.
The report also highlighted key lessons learned, including the lack of planning for the DAS solution’s operation in production, particularly in terms of funding. It also emphasized the need for improved financial forecasting with a longer-term planning horizon to ensure DAS’s sustainability.
Following the transfer of business ownership, expenditures for DAS were significantly reduced. As of mid-October 2024, the program projected financial sustainability for 2024/2025 based on revised operating expenses and expected revenue of approximately $4 million.Footnote 5 However, some DAS projects considered in the financial projections will not be onboarded (because they do not align with StatCan’s mandate), and this will impact potential revenue. In addition, several projects that are currently onboarded will be completed and decommissioned by the end of 2024/2025, and this will affect the accuracy of the projections moving forward.
Overall, the budget will require ongoing monitoring. Program interviewees noted that certain projects, like the City of Toronto SafeTO project, may generate further requests from other municipalities in Ontario (because each municipality must have a community safety plan and demonstrate progress). However, at the time of the evaluation, no official requests from other municipalities had been received from within Ontario or across Canada, perhaps because the SafeTO project has not been onboarded yet and there are no demonstrable benefits at this time.
Finally, while business ownership of DAS was transferred to Field 6, executive budget authority remained with Field 9.
Field 9 indicated that because DAS was initially an IT-led initiative, it was given budget authority, and it has remained there because it aligns with Chief Information Officer accountability for multipurpose computing platforms and the adoption of product management. Further, it was noted that, given the nature of the infrastructure of DAS, with capital expenditures and associated operations and maintenance costs, there is a risk that Field 6 could underestimate IT-related costs.
Field 6 explained that typically at StatCan, the business owner sets priorities and, as the budget holder, distributes funds when services are delivered. Additionally, Field 6 noted that because it does not have budget authority, its work on DAS was not funded until the new LoA was recently signed, and it has had to absorb these costs.
Despite the recent changes to the DAS program, there are still significant risks to its sustainability. Continuous technological investment, limited IT capacity for modernization, maintenance and support, along with low user satisfaction, could result in underuse and persistently high costs.
Despite efforts to improve the efficiency and sustainability of DAS, program documents and interview data suggest several ongoing risks to its sustainability. Figure 8 outlines these key risks, which include challenges with IT capacity, external user satisfaction and uptake, and technological advancements. These risks will need to be considered by the program moving forward, and appropriate actions must be taken to mitigate them.

Description - Figure 8 Ongoing risks for Data Analytics Services
Figure 8 outlines the ongoing risks associated with DAS, detailing the challenges and potential impacts.
- IT Capacity
- There are limited StatCan IT specialists available with the specialized skillset needed for ongoing modernization, maintenance, and support for new users.
- Other competing priorities have led to IT resource reallocation (e.g., Business Transformation) with a need for higher-level guidance around prioritization of programs/services and resources.
- Without ongoing modernization, there is a risk of the platforms becoming outdated and underutilized.
- External User Satisfaction and Uptake
- Issues with onboarding, communication and support, service interruptions, data access, costs, etc. pose an ongoing risk to external user satisfaction and uptake (some of these issues are also contingent on IT capacity).
- A lack of understanding around impacts for external users makes it challenging to understand the value DAS will provide users and whether they will continue using it and/or recommend it to others.
- Underutilization of DAS and reduction of its offered services poses a threat to cost recovery efforts and sustainability of the program.
- Technological Advancements
- Technology is advancing with respect to data analytics and cloud computing, and as a result, DAS will require ongoing modernization/innovation investment to stay relevant to external users.
- As technology progresses, there are some components of DAS that are outdated (e.g., outdated environments of the cloud, older security infrastructure), and efforts are underway towards modernizing these older components.
- It is expected that further investments in modernization will be needed in the future.
In addition to the above risks, VDL’s reliance on non-cloud-native infrastructure has led to challenges such as limited storage capacity, necessitating actions to accommodate growing user demand. As this takes place, it will be important to examine how VDL and the broader DAS environments can evolve together (i.e., opportunities for joint development) to ensure efficient resource allocation and avoid redundancy and duplication.
Given several uncertainties at the time of the evaluation, assessing the value for money provided by DAS was challenging. While DAS offers a relevant and in-demand service, there are performance issues that need to be addressed. Further, recent efforts to streamline offerings and implement stronger governance have narrowed the program’s scope and audience, limiting value and access for some external users.
Because of the recent changes to the scope and the potential for additional changes, it is challenging to fully assess value for money at the time of this evaluation. The Treasury Board of Canada Secretariat defines value for money as “the extent to which programs demonstrate relevance and performance.” There remains a clear need for a DAS-like solution to provide a secure, collaborative environment using cloud infrastructure. While DAS has had mixed success in achieving its goals to date, upcoming changes to the program have the potential to address current challenges and help better meet the growing demand for this kind of product.
However, DAS was initially established with a significant investment to deliver a cloud-based analytics platform that would provide data, analytical tools, software and computing power to a wide range of external users. StatCan operates under the authority of the Statistics Act, which establishes the mandate of the agency. Recent changes have strengthened governance and introduced more stringent processes and procedures. While these improvements have brought greater oversight, they have also narrowed the program’s scope and audience (e.g., customization was reduced for users, focus will be on external users of the platform moving forward, projects must align with the StatCan mandate). As a result, DAS is currently offering less value than originally anticipated, not fully providing the broad impact and accessibility that were initially envisioned.
When interviewees were asked about value for money, at a user level, some agreed that DAS was worth the cost (especially for projects requiring advanced analytical capabilities). Some others expressed concerns over high costs related to storage, data processing and cloud resources. This was particularly an issue for those working on less complex or budget-constrained projects, leading some to perceive the service as too expensive for its value. A few internal user interviewees also identified more affordable alternatives, such as SSC’s FSDH.
At a program level, some interviewees agreed that the cost for DAS is currently high for the number of users. However, it was suggested that with improved management and more users, overall financial performance could be improved, potentially offering a clearer value proposition, though it will take time to fully assess this.
How to improve the program
Recommendation 1
The ACS of Field 6, in collaboration with the ACS of Field 9, should ensure that the modernization efforts for DAS and VDL are aligned and not redundant. This will support alignment with the long-term vision for DAS while also making efficient use of StatCan’s limited IT resources.
Recommendation 2
The ACS of Field 6, in collaboration with the ACS of Field 9, should seek ways to improve the experience of external users to sustain the uptake of DAS and promote sustainable costs. Based on the evaluation findings, areas of improvement should include, but not be limited to, onboarding, service cataloguing, costing models and timely data access.
Recommendation 3
The ACS of Field 6, in collaboration with the ACS of Field 9, should ensure that effective monitoring of the program is carried out and, more specifically, that
- processes are in place to track and monitor direct and indirect DAS clients to better understand the DAS client base and uptake over time
- performance indicators for DAS, such as client use and satisfaction, are established and monitored regularly
- ongoing assessments of overall program costs, efficiency and duplication of services (i.e., VDL, AAW, CAE), and impact for users are carried out.
Recommendation 4
The ACS of Field 6, in collaboration with the ACS of Field 9, should explore possibilities to make the functionality and technology of the DAS platform more available to a broader audience of users.
Recommendation 5
The ACSs of Field 6 and Field 9 should review the current budget arrangement to ensure that it is efficient and effective, and that it aligns with organizational policies and practices.
Management response and action plan
Recommendation 1
The ACS of Field 6, in collaboration with the ACS of Field 9, should ensure that the modernization efforts for DAS and VDL are aligned and not redundant. This will support alignment with the long-term vision for DAS while also making efficient use of StatCan’s limited IT resources.
Management response
Management agrees with the recommendation.
The existing DAS Advisory Council, jointly chaired by the directors general of both fields, reviews issue escalations, the scope, the budget and the strategic alignment of activities. There is also an existing ACS-led DAS governance table. These existing governance structures will transition to the new DAS Steering Committee.
To effectively realize the modernization potential of the DAS platform, the new DAS Steering Committee will provide oversight on DAS’s strategic plan moving forward, aligned with various horizontal efforts, including VDL modernization efforts. The steering committee will review and approve annual workplans, in alignment with overall budgets, as well as be consulted on new work that is outside the existing platforms (defined product offering). The steering committee will review issue escalations, the scope, the budget and the strategic alignment of activities.
Deliverables and timelines
A senior management steering committee with representation from Field 6, Field 9, and key program senior executives and managers from other fields will be established by October 2025.
Terms of reference designating responsibilities and specific accountabilities will be approved by December 2025.
Recommendation 2
The ACS of Field 6, in collaboration with the ACS of Field 9, should seek ways to improve the experience of external users to sustain the uptake of DAS and promote sustainable costs. Based on the evaluation findings, areas of improvement should include, but not be limited to, onboarding, service cataloguing, costing models and timely data access.
Management response
Management agrees with the recommendation.
Field 6, in collaboration with Field 9, will review and articulate a strategy that will include, but not be limited to, the areas of improvement that were identified by the evaluation (i.e., onboarding, service cataloguing, costing models and timely data access). When the strategy is approved, a roadmap and timelines will be developed and approved by the DAS Steering Committee.
Deliverables and timelines
A strategy will be approved by April 2026. The roadmap and timelines will be approved by September 2026.
Recommendation 3
The ACS of Field 6, in collaboration with the ACS of Field 9, should ensure that effective monitoring of the program is carried out and, more specifically, that
- processes are in place to track and monitor direct and indirect DAS clients to better understand the DAS client base and uptake over time
- performance indicators for DAS, such as client use and satisfaction, are established and monitored regularly
- ongoing assessments of overall program costs, efficiency and duplication of services (i.e., VDL, AAW, CAE), and impact for users are carried out.
Management response
Management agrees with the recommendation.
Field 6, in collaboration with Field 9, will articulate a plan for effective monitoring of the DAS program, which will include processes to track and monitor direct and indirect DAS clients, the establishment of performance indicators, and assessment of overall program costs and impact for users. When approved, the program will implement the plan through the establishment of a roadmap and timelines.
Deliverables and timelines
The plan, which will include a roadmap and timelines for implementation, will be approved by the DAS Steering Committee by April 2026.
Recommendation 4
The ACS of Field 6, in collaboration with the ACS of Field 9, should explore possibilities to make the functionality and technology of the DAS platform more available to a broader audience of users.
Management response
Management agrees with the recommendation.
Field 6, in collaboration with Field 9, will explore possibilities to expand the availability of the DAS platform to include a broader set of users.
Deliverables and timelines
A business case analyzing options to expand DAS platform availability, including risks and associated requirements for the options, will be presented to the Strategic Management Committee by September 2026.
Recommendation 5
The ACSs of Field 6 and Field 9 should review the current budget arrangement to ensure that it is efficient and effective, and that it aligns with organizational policies and practices.
Management response
Management agrees with the recommendation.
Field 6, in collaboration with Field 9 and Field 3, will review the budget arrangement and the organizational policies and practices, as well as make recommendations in terms of the long-term budget arrangement.
Deliverables and timelines
A review of organizational policies and practices in the context of DAS will be completed by December 2025. The recommendations to the Strategic Management Committee on long-term budget arrangements for efficient and effective management of DAS will be provided by April 2026.
Appendix A: Visual depiction of Data Analytics Services

Description - Visual depiction of Data Analytics Services
The figure in Annex A provides a detailed diagram of DAS, illustrating various components and processes involved.
Data Analytics Services (DAS) sit on the StatCan Cloud and provide secure and flexible project storage accounts. DAS include the DAS platform itself as well as the support services.
The DAS platform provides the infrastructure and tools needed for data analysis and visualization. It consists of four main environments:
- CAE (Collaborative Analytics Environment)
- Project examples under CAE include:
- Public Service Data Challenge: A project aimed at leveraging data analytics to address public service challenges.
- Price Analytics Environment: A project focused on analyzing pricing data to gain insights and inform decision-making.
- Project examples under CAE include:
- AAW (Advanced Analytics Workspace)
- Project examples under AAW include:
- Census – Machine Learning Coding: A project involving machine learning coding for census data analysis.
- Producer Prices Division Isolated Post: A project focused on analyzing isolated posts within the Producer Prices Division.
- Oral Health Workbench: A project aimed at analyzing oral health data to improve public health outcomes.
- Project examples under AAW include:
- GAE (Geospatial Analytics Environment)
- Project examples under GAE include:
- HR Viewer: A project that involves creating a viewer for human resources data.
- Infrastructure Project Planning Support Tool: A tool designed to support the planning of infrastructure projects through data analysis.
- Project examples under GAE include:
- .Stat SDM
- .Stat SDM is an open-source platform for the efficient production and dissemination of high-quality statistical data. It supports the exchange of statistical data and metadata using the SDMX (Statistical Data and Metadata Exchange) standard.
- Other environments can be created by combining the functionalities from CAE, AAE, and/or GAE:
- Hybrid project examples include:
- Business Data Lab, which combine CAE and AAW functionalities
- SafeTO, which combine CAE and GAE functionalities
- AgZero, which combine CAE, AAW and GAE functionalities
- Hybrid project examples include:
- VDL (Virtual Data Lab) is an early CAE legacy version with some services and legacy tools. It provides a virtual environment for data analysis and experimentation.
The DAS support services include:
- Operations & Monitoring: Continuous monitoring and operational support for the DAS platform ensure that the platform runs smoothly, and any issues are promptly addressed.
- Data Ingestion Services: Services that support the ingestion of data into the DAS platform. This includes the processes and tools required to import data from various sources into the system.
- Approved Access to StatCan Data: Ensures that users have the necessary permissions to access data from Statistics Canada. This access is crucial for conducting data analysis and generating insights.
- Solution Engineering, Onboarding Coaches, Sandbox Services: These services are designed to facilitate the integration and utilization of DAS. Solution engineering involves the technical setup and customization of the DAS platform. Onboarding coaches assist new users in getting started, while sandbox services provide a safe environment for testing and experimentation.
The figure also depicts The Zone, which also sits on the StatCan Cloud but resides outside of DAS. Internally, users access The Zone, which is similar to AAW. Because, there is no collaboration with external partners in The Zone, and it is not Internet abled, it provides a secure space for internal data analysis and experimentation.
Data input to DAS stem from various sources:
- StatCan Data: Data from Statistics Canada.
- External Data: Data from external sources.
- Open Data: Publicly available data.
- Client-owned Data: Data owned by clients.
These data can be uploaded into DAS via:
- Azure Data Factory: A cloud-based data integration service.
- Secure EFT: Secure electronic file transfer.
- APIs: Application programming interfaces for data exchange.
- Direct Connection: Direct data connections to various sources.
- Azure SQL: A managed cloud database service.
- Azure Blob: Object storage for unstructured data.
- Azure Data Lake: A scalable data storage and analytics service.
Once uploaded, the data are stored in secure and flexible project storage accounts. Data are stored and loaded into Azure SQL, Azure Blob Storage on Datalake, or Azure Fileshares. These storage solutions provide secure and scalable options for managing data.
Data outputs can be downloaded from DAS in the following products:
- Tables: Structured data tables.
- Charts: Graphical representations of data.
- Data Visualization: Visual tools and dashboards for data analysis.
Appendix B: Key Data Analytics Services projects
| Project name | Description | Status |
|---|---|---|
| Personal Protective Equipment (PPE) | Enabling collaborative analysis to provide insights and facilitate decision making about availability of PPE during the pandemic. | Complete |
| Business Data Lab | Working with clients to combine datasets; process, model and produce visualizations; and share economics insights with the business community. | Ongoing |
| Price Analytics Environment | Ingesting big administrative datasets to produce statistics. | Ongoing |
| Labour Force Survey Pre-release | Providing clients with secure data visualization of pre-release data. | Ongoing |
| Virtual Data Lab | Providing researchers across the country with the tools they need to securely leverage Statistics Canada data. | Ongoing |
| Canadian COVID-19 Antibody and Health Survey | Providing an advanced analytical workspace for end-to-end data ingestion and integration with external researchers. | Ongoing |
| Vehicle Registration Files | Using secure data ingestion pipelines to manage large datasets. | Ongoing |
| AgZero | Agriculture Division and AI Methods Division (formerly named the Data Science and Innovation Division) of Statistics Canada and external partners using the Collaborative Analytics Environment, the Advanced Analytics Workspace and the Geospatial Analytics Environment platforms to enable AI Methods to develop machine learning models to produce official statistics. | Ongoing |
| Statistics Canada’s Oral Health Statistics Program | Working with clients to provide access to pre-release survey data. | Ongoing |
| Canadian Internal Trade Data and Information Hub | Providing access to data on internal trade using Statistical Data and Metadata eXchange standards. | Ongoing |
| Canadian Centre for Energy Information | Providing a convenient one-stop virtual shop for independent and trusted information on energy in Canada. | Ongoing |
| SafeTO | Leveraging multisector data sources to help expand the definition of community safety beyond crime statistics and enforcement to include prevention and well-being. | Not yet onboarded |
Appendix C: Evaluation questions and indicators
| Evaluation questions | Evaluation indicators |
|---|---|
| To what extent is there a continued need for Data Analytics Services (DAS)? |
|
| To what extent has DAS achieved its intended results? |
|
| To what extent is DAS sustainable in its current state? |
|
Appendix D: Interview quantification scale
Interview responses are quantified and categorized in this report using the scale shown in the table below.
| Term | Definition |
|---|---|
| One | One is used when one participant provided the answer. |
| Few | Few is used when 4% to 15% of participants responded with similar answers. The sentiment of the response was articulated by these participants but not by other participants. |
| Some | Some is used when 16% to 45% of participants responded with similar answers. |
| About half | About half is used when 46% to 55% of participants responded with similar answers. |
| Most or a majority | Most, or a majority, is used when 56% to 89% of participants responded with similar answers. |
| Almost all | Almost all is used when 90% to 99% of participants responded with similar answers. |
| All | All is used when 100% of participants responded with similar answers. |
Appendix E: Status of Data Analytics Services objectives
| Data Analytics Services (DAS) objective | Description | Status |
|---|---|---|
| 1. DAS platform product | This capability focuses on creating a cohesive user experience, designing persona and journey maps, and packaging solutions. | Fully achieved |
| 2. Data science exploration and integration, including artificial intelligence (AI) and machine learning | Using both open-source and commercial tools, data scientists and analysts leverage AI and machine learning to cleanse, label and classify data; detect patterns; and create predictive models. This capability ensures that the platform supports the integration of emerging algorithms and their effective use, with the necessary capacity to achieve research goals. | Fully achieved |
| 3. Big data analytics and processing | Big data typically have properties like volume (large size), velocity (fast processing) and variability (mixed formats). While other capabilities address velocity and variability, conducting complex analysis and value extraction requires powerful platforms that leverage cloud computing services. This capability supports big data capacity through parallel processing, in-memory processing, graphics processing units (GPUs) (for fast computation) and more. | Fully achieved |
| 4. Cognitive and knowledge services | DAS leverages advanced techniques to capture and use the knowledge of researchers, users and internal experts. By mining queries, searches, AI, machine learning and published results, DAS aims to build a knowledge web that enhances its analytic efforts. | Fully achieved |
| 5. Data aging and archiving | Managing data throughout their lifecycle is crucial for complying with information management policies and optimizing resource use. Effective tiered storage and data aging strategies ensure cost optimization. The DAS platform will not centralize all data but will use a combination of consolidation and cataloguing with links to data at other locations. | Fully achieved |
| 6. Data virtualization | The platform hosts complex, diverse data in various forms and structures. While standardizing data into warehouses has been complex, especially for diverse use, modern data virtualization techniques allow data to be stored and managed without full normalization, delivering them “on demand” in the required format. Data provisioning supports comprehensive presentation data, maximizing user value and optimizing management. This capability offers features to address these needs. | Fully achieved |
| 7. Data visualization | Visualization and compelling storytelling are essential for using data to provide evidence for policy, measure results, and offer insights and predictions. This area has seen significant growth in self-serve approaches, allowing users to create customized visual insights from analytic results. Data scientists need visual outputs to advance their work. Capabilities must accommodate the diverse personas and backgrounds of users and researchers, enabling researchers to publish their results efficiently across various platforms (publications, websites, etc.). | Fully achieved |
| 8. Identity and access management | Strong digital identity, authentication and authorization services are essential for security. DAS will integrate with current and future identity schemes, including the Government of Canada digital identity. Directory services and access management linked with data management services ensure well-managed and consistent access. The program will follow the Treasury Board’s Directive on Identity Management and relevant policies. DAS is designing for the use of Government of Canada digital identity services. | Fully achieved |
| 9. Metadata management | Rich metadata are essential for supporting researchers and users throughout their activities. They offer descriptive and statistical context for data, classification and taxonomy tools, record layouts, and more. The platform will provide open access to metadata through various mechanisms (user experience, application programming interfaces [APIs]) and will support the collaborative creation and evolution of shared metadata (e.g., co-developed classifications with other departments and agencies). | Fully achieved |
| 10. Orchestration and workflows | Modern businesses benefit from process and workflow automation, where collections of services are integrated into automated processes to deliver business value (e.g., Netflix services). This capability will provide the means to integrate and automate business workflows composed of services, delivering analytic business value. | Fully achieved |
| 11. Registers and reference data services | Statistics Canada holds extensive master, reference and statistical register data. Through the DAS platform work and in alignment with the Government of Canada data strategy and the Digital Experience Platform, DAS will provide controlled access to these data via APIs and other means, ensuring appropriate credentials and controls. | Fully achieved |
| 12. Remote researcher access | This serves as the starting point for researcher access to the platform, delivering key functions such as
|
Fully achieved |
| 13. Rich analytics workbench | Researchers and users conduct extensive mathematical analysis with their data using various “packaged” workbenches that support popular tools like R, Python, TensorFlow, SAS, SPSS and Stata. A crucial component is the support for external libraries and functions from both internal and external communities (see the collaboration capability). DAS is also transitioning from traditional tool approaches to “analytic notebook” approaches with built-in documentation creation capabilities. | Fully achieved |
| 14. Scalable infrastructure services | The DAS platform needs agile, flexible and scalable infrastructure to meet user needs. Data are growing exponentially, including web scraping, sensor, satellite and Earth observation data. Users need to cleanse, integrate and provision these data for processing and analysis. Critical enablers include scalable storage, computing, memory, GPU access and network capacity. This capability ensures sourcing from powerful and secure public cloud vendors. | Fully achieved |
| 15. Data pipeline engineering | Data are collected from various sources (ingestion, web scraping and APIs) and then processed, cleansed and prepared for use. They flow through diverse teams, from initial access points to statistical infrastructure and subject-specific areas. To optimize this, DAS is focused on building pipelines—engineered with automation and other capabilities—for a streamlined, efficient data flow that maximizes enterprise resources. | Fully achieved |
| 16. Data capability publishing | As researchers and users interact with the platform, and as internal experts and producers create more content, it is important to be able to publish for the broader community via the platform. The goal is to have a one-stop catalogue and means of accessing the data. A key feature is supporting the curation of published data to ensure they remain current and relevant. | Partially achieved |
| 17. Data services and APIs and interoperability and integration services | Data exchange and use extend beyond user–software interactions; systems and solutions across the stakeholder space must also connect machine to machine. DAS will provide data services and APIs, enabling solutions in user departments and businesses to retrieve reference, register, aggregate and other data. Leveraging the Government of Canada’s Digital Exchange Platform and API store, DAS will publish and integrate with other stakeholders. All API exchanges will comply with relevant policies and directives, ensuring that data access respects privacy and statistical use restrictions. In line with the Data Strategy for the Federal Public Service, opportunities will be explored for API access to public reference data. | Partially achieved |
| 18. Analyst and algorithm collaboration | Researchers and users benefit greatly from internal and external collaboration to develop and share algorithms and data. Effective version control and configuration management are crucial, especially for machine learning activities. This capability supports user, team and community code and data collaboration, adhering to open-source industry standards. | Partially achieved |
| 19. Auditing and reporting | DAS builds trust through transparent communication about what is accessed, by whom, for what purpose and when. To ensure appropriate use, detect anomalies, address potential issues and enhance platform value, robust and secure auditing, logging and reporting capabilities are essential. These processes span all levels of the DAS platform—covering infrastructure, data access, solutions and users—and are integrated with identity management and access controls. | Partially achieved |
| 20. Geospatial services | Geospatial data are crucial for research and analysis, serving either as the lens through which researchers engage with data or as a means to present analysis results effectively. The Federal Geospatial Platform offers a significant government-wide service and DAS’s capability will integrate with it. | Partially achieved |
| 21. Information governance | Governance and stewardship are crucial for an effective “data marketplace,” both in the modernization program and national data strategies. Open digital workflows are needed to support the governance community in defining quality, information models and other standards. A key component is creating access control mechanisms for various sensitivity levels and managing them efficiently. | Partially achieved |
| 22. Scalable platform services | Effective use of platform capabilities for complex analysis requires agile, flexible and cost-effective platform and infrastructure capacity. This includes databases, data stores and analytic platform components provided by cloud vendors. | Partially achieved |
| 23. Secure multi-party computation (shared trust data collaboration) | User interviews and stakeholder engagements frequently highlight the need for partners to collaborate using sensitive data without disclosing the entire dataset (for example, deriving a non-sensitive linked record result from sensitive data that neither party wishes to fully disclose). Emerging technology and software approaches can achieve this. DAS is working with academics, international partners and commercial vendors to identify, evaluate, select and deploy solutions that meet strict privacy and security requirements. This collaboration involves stakeholders like the Office of the Privacy Commissioner of Canada and other government departments. | Partially achieved |
| 24. Data discovery | Researchers and users need advanced search tools to locate algorithms and data for building models, visualizations and insights. Discovery features go beyond basic search, using inference and other techniques to uncover necessary resources, including hidden or overlooked data assets within the broader data marketplace. | Partially achieved |
| 25. Provenance and lineage management | Effective research and quality outputs rely on knowing the provenance of data entering a processing or analysis step and understanding their lineage over time. These services offer coarse- and fine-grained provenance and lineage, often as part of broader solutions. | Not realized; deemed out of scope |
Download PDF version
Download PDF version