Every municipality in Canada was already measured. Nobody had put them side by side.
The biggest decision most households make, where to live, gets made with the data of one city, or none. Statistics Canada counted every municipality in the country in 2021 and published the same profile for each of them. But the portal serves them one at a time, so a family in Scarborough can learn everything about Scarborough and nothing about Quispamsis. The comparison that matters most is the one the data was never arranged to make.
I loaded every 2021 census subdivision in Canada, 5,161 of them: every city, town, rural municipality, reserve, northern area and settlement. To the census profile I joined nine other public datasets: monthly labour force numbers, the Crime Severity Index back to 2010, proximity to groceries, parks, clinics and libraries, business openings and closings, migration, OpenStreetMap amenities, wildfire history and federal turnout. That is 131 indicators per place and just under 600,000 values, refreshed every hour. Each indicator is ranked as a percentile against the 3,995 regular municipalities and rolled into a 0 to 100 score across nine dimensions.
Here's the first thing the table shows.
About 3,700 municipalities have enough published data to be scored. The other 1,450, mostly reserves and places under a thousand people, the census itself keeps in the dark.
The second thing is where the big cities land. The national median score is about 51. Calgary is 54.1, Vancouver 48.5, Ottawa 48.0, Montréal 39.1 and Toronto 39.0. Toronto and Vancouver score 1 out of 100 on housing, lower than every other scored place in the country, and 1 on equity. The top of the national list is small Québec and Ontario exurbs and base towns, Petawawa, Lucan Biddulph, Labrador City, Boischatel, and the bottom is rural Alberta counties. The big cities are not the top of the table. They are the middle of it, and on housing they are the floor.
That's the compilation. Here's what it's for.
Take one question a household actually asks: if I can't afford Toronto, how far out do I have to go? With every municipality on one scale it takes a minute to answer. Divide each place's median home value by its median household income, the multiple CMHC uses, and measure the distance from the nine big cities. Not one of the 113 municipalities within 150 km of Toronto, or the 58 within 150 km of Vancouver, is as affordable as the middle of Calgary. Calgary is 4.7 times income; Edmonton, Winnipeg, Québec City and Halifax are 4.2 to 4.3. The lowest ratio within 100 km of Toronto is 6.4, and the exurbs that get there have two to three times the share of hour-plus commuters. Around Montréal and Ottawa the same drive pays off in the first suburb: Sainte-Catherine is 3.9, Gatineau 4.0. That is one query. The table answers thousands.
What a household actually asks
Where can we afford to buy, and what do we give up to get there? How long is the commute? Is the place growing or emptying out? Are there shops opening or closing? How far is a grocery store, a park, a clinic? Has it burned? Every one of those has a column in the table, for every municipality, on the same scale. Two things people ask do not: school quality, because no public dataset measures it at this level, and air quality. I would rather the table say so than dress up adult education rates as a school ranking. So it says so.
The comparison a household needs, this town against that one two provinces over, takes seconds instead of an afternoon of tabs. Search any of the 5,161 names. Drill from the country to a province to a census division and read the top and bottom ten. Open a place and get eight tabs, people, identity, housing, income, work, education, commute, with each number's national rank beside it. Put two places side by side. Switch to the news lens and see what has happened there in the last two weeks.
This isn't a ranking of the best places to live
What I want to be clear about: the score is one lens. The nine dimensions and their weights are editorial choices, stated in full on the methodology page, and I have not validated the score against how satisfied people say they are with where they live. It tells you where a place sits among 5,161 on the things that were measured. It does not tell you where you will be happy.
And the data has limits it did not choose. About 70 percent of it is the 2021 Census, four years old and counted in a pandemic spring. Crime and unemployment are published for metropolitan areas and provinces, not towns, so every municipality inside a metro carries its metro's number. Reserves are kept out of the percentile pool because the census suppresses so much of their data; the methodology page puts it plainly: the score "is not a fair lens for Indigenous communities. Where reserve data is shown, it is partial by design." That is right. The table is for reading the country, not for grading it.
The question isn't which town wins. It's whether a family in Scarborough can see Quispamsis at all.
Until now they couldn't, not without a spreadsheet and a week. The census is paid for by everyone and published for everyone, and read, at the municipal level, by almost no one, because reading it meant one download per town. The table and the API behind it are open, with no login, and rebuild themselves every hour as Statistics Canada releases monthly labour and annual crime figures. The 2026 Census, due in February 2027, adds a second point in time to every one of the 5,161 rows, and the trends start.
Methodology
The unit is the 2021 census subdivision, keyed by its Statistics Canada DGUID, with centroids and polygons from the 2021 Cartographic Boundary File. Sources: Census Profile 2021 (98-316-X2021006) and population counts (98-10-0002); Labour Force Survey 14-10-0459, monthly; Crime Severity Index 35-10-0026, 2010 to 2024; Proximity Measures Database 2020; business openings 33-10-0270; components of population change 17-10-0149; OpenStreetMap; the Canadian National Fire Database 2014 to 2023; Elections Canada 2021. Skewed indicators are log-transformed, then each is ranked as a percentile against the 3,995 regular municipalities, capped at the 1st and 99th, and inverted where lower is better. Dimensions combine their indicators by inverse-variance weight, and the score is score = Σ wᵈ · dimensionᵈ ÷ Σ wᵈ × 100. over the dimensions present, with weights of 14.3 for prosperity and housing, 11.8 for equity, 10.9 for safety, education, mobility and community, and 8 for livability and opportunity. A dimension needs half its indicators and a place needs five dimensions, or it is marked insufficient. The affordability example is PTI = median dwelling value ÷ median household income. for the 732 regular municipalities of 5,000 or more, with straight-line distance to the central municipality of each metro over about 800,000 people plus Halifax. Everything above is reproducible from the public API.
If you're a planner or a demographer and you think I've got this wrong, my email's in the Planada about page. I'd rather be corrected publicly than wrong publicly.