Voices from the Field

Abstract

At first glance, Manitoba appeared to be making steady progress toward its commitment to create 23,000 new child care spaces. But when we looked beyond provincial totals, a more complex picture emerged. Child care spaces were opening, and many more had been announced. Yet a closer look at the data highlighted gaps in access, suggesting that families in some communities may face greater challenges finding care than others, and raising questions about whether current progress was enough to meet not just a target, but the real demand.

This article explores how data analytics supported our performance audit work on Child Care Access and Inclusivity in Manitoba. More specifically, it reflects how analytics helped us move from broad reporting on child care spaces to a clearer understanding of access, planning pressures, and future risk. The analysis strengthened the evidence for our findings by showing where the available data was sufficient, where it was limited, and where important questions remained unanswered.

Introduction

Performance auditors are often asked to assess whether plans, processes, and results are adequate and achieving the goals set out for a program. In many cases, organizations can report what they are doing, but the information available is not always enough to show whether those efforts are reaching the right places, responding to the right pressures, or keeping pace with future needs. That was one of the central issues in our audit of Child Care Access and Inclusivity in Manitoba.

During the initial stages of our audit, we noted the department had data on open and committed child care spaces, along with some planning tools intended to support decision-making. But as the audit progressed, it became clear that these sources did not fully explain where child care access pressures were the greatest, how those pressures varied across Manitoba, or whether current planning efforts were likely to meet Manitoba’s commitment to create 23,000 additional spaces under the interactive dashboard. The detailed results of this analysis are available in our audit report and accompanying interactive dashboard.

Data analytics gave us a way to examine the issue from a different perspective. One lesson that runs through this work: analytics helped us see not only what the data showed, but what was missing.

From reported spaces to real-world access

We started off by looking beyond the total provincial numbers, breaking down available child care spaces in each community. While province-wide numbers support reporting, they do not reflect whether families and communities have reasonable access to child care. Access is shaped by geography, local population, and how close facilities are to where children live. It is also influenced by the fact that a single facility may serve more than one area, especially near geographic boundaries.

To better understand this, we used population estimates, facility locations, and geographic groupings to examine access at the community level. We also used Forward Sortation Areas (FSAs), geographic areas defined by the first three characters of a postal code, as a geographic frame to look at how licensed and committed spaces relate to the population of children within and around those areas.

For this analysis, we converted Manitoba's 23,000-space commitment into a coverage-based benchmark of approximately 3.8 spaces per 10 children, based on the population of children under age seven at the start of the agreement. As shown in Figure 1, projected child care coverage varies considerably across Manitoba communities. While some areas were projected to approach Manitoba's coverage-based equivalent of the 23,000-space target, many communities remained well below this level.

A practical note on methodology

Understanding child care access is not as simple as counting spaces within a geographic area. Families do not necessarily use the child care centre closest to their home, and many facilities serve children from neighbouring communities, particularly in areas near municipal or postal boundaries.

To better reflect how families access child care, we considered the location of child care spaces, the population of children, and how far families might reasonably travel for care. Rather than assigning a facility's spaces only to the area in which it is located, the analysis recognized that child care facilities may serve families from multiple surrounding communities (additional details on the methodology, assumptions, and calculations are available in our audit report).

The analysis considered both distance and population. We estimated the distance between child care facilities and nearby geographic areas and applied weighting factors based on proximity and size of the child population in each area. This allowed facility capacity to be distributed across neighbouring communities rather than attributed to a single location.

In practical terms, a child care centre located near the boundary of an area was treated as potentially serving families from several nearby communities. This provides a more nuanced estimate of geographic accessibility than assigning every space exclusively to the area in which the facility was located. Figure 2 illustrates this concept. Rather than assigning all spaces to a single geographic area, the accessibility model distributes capacity across neighbouring communities based on distance and population, providing a more realistic estimate of access.

This method gave us a reasonable way to assess accessibility with the available data and helped to identify communities where access pressures may persist despite planned growth in child care spaces.

This approach provided a useful estimate of geographic accessibility, not a measure of utilization. It did not incorporate all factors that may influence child care utilization, such as parental preferences, commuting patterns, transportation availability, or waitlist information.

The analysis also highlighted limitations in available data. Key information such as province-wide waitlists, facility closures, and geographic demand indicators were not consistently tracked or available, making it difficult to fully assess child care access and demand across Manitoba. While analytics could not compensate for these data gaps, it helped make them visible and strengthened the evidence supporting our audit findings.

Connecting access to implementation risk

Once we had a better understanding of access at the local level, the next step was to connect that to Manitoba’s commitment under the CWELCC Agreement. The department was reporting annually on the number of spaces that were open and committed, but our audit question was broader: Would the current pace of progress be enough to achieve 23,000 spaces within the required timeframe?

Using data on open and committed projects, we examined the department’s progress against Manitoba’s target of 23,000 additional spaces. Between 2021 and 2025, the department committed to an average of 2,309 spaces per year. Our analysis showed that these committed spaces typically take 18 months to four years to open and become operational.

Using historical commitment rates and observed timelines between commitment and operation, we projected that Manitoba would achieve the 23,000-space target by May 2031, with full operation not expected until sometime between 2032 and 2035. This meant Manitoba was unlikely to achieve its 23,000-space commitment by the March 31, 2026 deadline (the underlying assumptions and forecasting approach are discussed further in our audit report).

As shown in Figure 3, Manitoba had opened 5,419 spaces and committed a further 6,127 spaces by November 2025, leaving a substantial gap to the 23,000-space target. The visualization highlights both the number of spaces opened and committed to date, and the additional commitments that would be required to meet the original deadline.

This analysis was particularly useful in connecting implementation progress with future delivery risk. The report found that the department did not maintain province-wide waitlist information, did not continuously monitor space closures, and did not have a model that could estimate the number of spaces needed by area. It also showed how much more planning and analysis could be done with the information the department already has.

Analytics did not fill all those gaps, but it made them much harder to ignore. In that way, it strengthened the audit by showing not only where evidence existed, but also where management lacked the information needed for stronger planning.

Applying a coverage lens

Another helpful part of this work was the use of a coverage rate as a benchmark for comparison. We converted the 23,000 additional spaces from Manitoba’s CWELCC Agreement into a coverage-based target (38%). This allowed us to move beyond space counts and assess how well the planned number of spaces aligned with the distribution of Manitoba's population of children under seven.

We then developed a publicly available interactive dashboard that allowed users to explore coverage across Manitoba by postal code and community. Our analysis compared the number of open child care spaces as of November 30, 2025, and projected open spaces in 2031, with the population of children in each area. We estimated the number of additional spaces needed to meet Manitoba’s CWELCC commitment of 23,000 spaces (38%) and a coverage benchmark of 59% of the population under age seven. Although a 59% coverage-based target is used in other provinces, Manitoba has not committed to a coverage-based target under CWELCC. We used a 59% coverage rate for illustrative purposes only.

While Manitoba's commitment was expressed as a target number of spaces rather than a coverage rate, the benchmark provided a practical way to put the target into context and examine how access varied across communities. It also helped illustrate where projected spaces may still fall short of achieving comparable levels of coverage.

Figure 4 shows the interactive dashboard we used to compare current and projected coverage rates across Manitoba communities and to estimate the number of additional spaces required under different planning scenarios.

From current gaps to future pressure

The next step was to ask what the projections meant for the future. If access was already uneven across communities, how might those pressures evolve as population and space commitments changed?

To help answer those questions, we used forecasting based on historical commitment patterns and observed timelines for projects to become operational. This allowed us to estimate when Manitoba would likely reach the 23,000-space target based on the pattern of commitments recorded up to November 2025, as well as typical timelines for those spaces to become operational. As with any forecast, the analysis assumes that historical patterns provide a reasonable basis for estimating future commitment rates, project completion timelines, population growth, and program implementation. Actual results may differ due to changes in funding, construction timelines, policy decisions, population growth, or other factors affecting child care expansion.

The analysis indicated that Manitoba was unlikely to achieve the 23,000-space target by the March 31, 2026 deadline and full operation of the required spaces would occur several years later. This provided a clearer picture of implementation risk and highlighted the scale of additional commitments that would be required to meet the target on time.

Figure 5 illustrates the relationship between projected space openings, population growth, Manitoba's 23,000-space commitment, and higher coverage benchmarks used elsewhere in Canada. The analysis suggests that projected growth in child care spaces is unlikely to close the gap within the timelines established by the agreement.

Forecasting also helped reinforce an important audit point: planning cannot rely only on present counts. The audit found that the department had not conducted this type of forward-looking analysis as part of its own planning. This mattered because forecasting is not simply a technical exercise; it is a management tool. It helps decision-makers understand whether current activity is aligned with future commitments and whether risks are building quietly in the background.

The area-level analysis supported this further. While additional spaces were expected to open in future years, projected coverage remained uneven across Manitoba. Some communities were expected to continue experiencing lower levels of access, even after committed projects became operational. This helped explain why accessibility remained a planning challenge despite significant investments and ongoing space creation efforts.

Limitations

As with any audit analytics, the results should be interpreted in light of the data and methodological limitations. Several important indicators of child care demand were not available. As noted in our report, the department did not maintain province-wide waitlist information, did not continuously monitor child care space closures, and did not have geographic estimates of the number of additional spaces needed by community. As a result, our analysis relied primarily on the relationship between child populations and the supply of licensed and committed child care spaces rather than direct measures of unmet demand.

Our analysis considered the location of child care facilities, child population estimates, and travel distance between communities and facilities. This approach assumes that proximity influences access and that facilities may serve families from multiple neighbouring communities rather than only the community in which they are located. However, actual child care choices may also be influenced by factors that were not included in the model, such as parental work locations, transportation options, program preferences, operating hours, and socioeconomic circumstances.

The forecasting analysis relied on historical population trends, departmental information on committed spaces, and observed timelines between commitments and facility openings. The projections assume continuity in historical trends and do not account for future policy changes, funding decisions, economic conditions, significant demographic shifts, or facility closures. Therefore, the forecasts should be viewed as illustrations of potential implementation risk rather than precise predictions of future outcomes.

Why analytics matter in audit work

One of the clearest lessons from this audit is that data analytics can play a meaningful role in performance audit work when they are used with care. They do not eliminate the need for professional judgment or a thorough understanding of the context, policy intent, or operational realities. What they do provide is another way to test whether the evidence supports the conclusions being drawn.

When used well, analytics can strengthen findings by reducing reliance on anecdote, better linking evidence to conclusions, and making gaps in information easier to identify. They can help auditors turn broad commitments into measurable questions, connect activity to future risk, and highlight whether management has the information needed to plan effectively.

What stood out most in this work was how much more became visible once the data was viewed differently. The underlying information already existed, but patterns in accessibility, community-level pressures, and emerging risks were not immediately apparent. Mapping and visualizing the data did more than illustrate the findings; it helped reveal questions we did not previously know to ask. Analytics showed that when strong questions guide the work, the same data can tell a more complete and meaningful story.

Ultimately, the value of analytics in audit work is that they help auditors better understand the issues they are examining and provide clearer, more persuasive evidence to support their findings.

References

[1] Sun, S.; Sun, Q.; Zhang, F.; Ma, J. A Spatial Accessibility Study of Public Hospitals: A Multi-Mode Gravity-Based Two-Step Floating Catchment Area Method. Appl. Sci. 2024, 14, 7713. https://doi.org/10.3390/app14177713

[2] Liu, L.; Lyu, H.; Zhao, Y.; Zhou, D. An Improved Two-Step Floating Catchment Area (2SFCA) Method for Measuring Spatial Accessibility to Elderly Care Facilities in Xi’an, China. Int. J. Environ. Res. Public Health 2022, 19, 11465. https://doi.org/10.3390/ijerph191811465

[3] Liao, Y.; Furuya, K. A Case Study on Children’s Accessibility in Urban Parks in Changsha City, China: Developing an Improved 2SFCA Method. Land 2024, 13, 1522. https://doi.org/10.3390/land13091522