OpenStats Canada
Hometown rankings

Canadian Broadband Speeds: Coverage Data vs Lived Speeds

Federal broadband maps report what internet providers say they can deliver to an address, which is a different measurement than what a household actually experiences.

Innovation, Science and Economic Development Canada (ISED) maintains the National Broadband Data (NBD) map, the federal government's primary tool for tracking internet availability across the country. Internet service providers report, twice a year, which services they offer at which locations and the maximum advertised download and upload speeds available there. ISED aggregates these provider submissions into hexagonal grid cells and community-level summaries, which is the data the internet topic site draws from.

This is a coverage and availability dataset, not a performance dataset. It answers the question "what service tier can a provider sell at this location" rather than "what speed does a typical household on this street actually measure." The difference matters more than it might first appear.

What the coverage data measures

Each broadband technology tracked in the NBD map, cable, DSL, fixed wireless, fiber, and satellite, has a different relationship between advertised and delivered speed.

  • Fiber-to-the-premises delivers a dedicated connection to each address, so advertised speeds are usually close to what a household actually gets, aside from normal in-home Wi-Fi losses.
  • Cable (DOCSIS) shares bandwidth across a neighbourhood segment, so advertised maximum speeds can be reasonably accurate off-peak but drop during high-demand evening hours when many households in the same segment are active at once.
  • Fixed wireless speed depends heavily on distance from the tower, line of sight, and how many subscribers share the same tower sector, so advertised maximums often represent a best-case scenario rather than a typical one.
  • Satellite, including newer low-earth-orbit services, can post competitive advertised speeds while carrying higher and more variable latency than terrestrial options, which affects real-time applications like video calls more than simple downloads.
  • DSL speed depends on the copper line's length and condition between the home and the nearest telephone exchange, so advertised maximum speeds decline noticeably in rural or older infrastructure areas even where DSL is nominally available.

Why availability differs from adoption

A location shown as having gigabit-capable service available does not mean every household there subscribes to a gigabit plan. Coverage figures describe what a resident could sign up for if they chose to and could afford it; actual adoption is shaped by price, contract terms, and whether an existing lower-tier plan already meets a household's needs. The internet topic site reports availability at each service tier because that is what the underlying federal data measures, and treats it explicitly as a ceiling on what is offered rather than a measure of what people are using day to day.

How OpenStats internet pages are built

The internet topic site processes the ISED NBD map CSV releases, which include hex-level statistics, community point data, and per-provider service layers covering current deployments, government-supported expansion projects, and private expansion projects. Each city page matches the nearest hex cells and community points to that municipality using geographic proximity, then aggregates the reported maximum available speed tier and provider count for that area.

Where a municipality does not have a hex cell or community point close enough to produce a reliable direct match, the page is marked as using a regional or provincial estimate instead, since broadband infrastructure can vary sharply within a short distance, particularly at the edge of a city's built-up area where fiber build-out often stops.

Reading a city's speed tier

  • Speed tier reflects the fastest advertised service generally available in that area, not an average across all providers.
  • Fiber availability, where flagged, is the strongest indicator that advertised and delivered speeds will be close together.
  • A high headline speed tier with only one provider listed indicates limited competition, which can affect pricing even where speed itself is not a constraint.
  • Government-supported and private expansion project layers indicate planned or in-progress builds, not necessarily service available today.

Why fiber build-out is uneven

Fiber-to-the-premises requires new physical infrastructure, running fiber optic cable to each address, which is capital-intensive and easiest to justify economically in denser urban areas with more potential subscribers per kilometre of cable. Rural and remote communities, where the cost per household to build fiber is far higher, have historically relied more heavily on fixed wireless and satellite. Federal and provincial funding programs, tracked in the NBD map as government-supported expansion projects, exist specifically to offset that economics gap and extend fiber or equivalent service into areas the market alone would not reach as quickly.

This uneven build-out is the main reason broadband speed tiers can differ sharply between a city core and its surrounding rural municipality even when both fall within the same general region, and it is also why the internet topic site reports speed tier at the municipal level rather than only at the provincial or national level.

Putting the data to practical use

Coverage data is most useful for understanding what is theoretically available at an address before shopping for service; it is less useful as a predictor of exact real-world speed on any given evening. Anyone deciding between providers at a specific address should still confirm current plans and pricing directly with providers serving that location, since service availability, pricing, and promotions change more frequently than the twice-yearly NBD map update cycle.

Universal service target and the 50/10 benchmark

Federal broadband policy has long used a universal service objective of 50 megabits per second download and 10 megabits per second upload (often written 50/10), with the option of unlimited data, as a baseline for what counts as adequate fixed broadband. ISED maps and related CRTC monitoring track progress toward that objective and toward higher gigabit-class tiers where networks have been upgraded. A community shown as meeting 50/10 still may lack fibre or cable gigabit options; conversely, a community with fibre available may still have pockets below 50/10 on legacy copper.

Funding programs such as the Universal Broadband Fund support projects in areas where private build-out alone has been slow. Those projects appear in NBD layers as government-supported expansion. Until a project is marked as available service, treat it as planned capacity.

Hex grids, community points, and municipal matching

ISED publishes coverage on a hexagonal grid and as community-level points. Hex cells are a spatial summary of provider-reported service, sized so rural and urban areas can be compared on a common geometry. Community points attach speed and technology attributes to named places. Neither layer is a guarantee for every civic address inside the cell or community boundary.

Edge cases matter. Fibre often stops at the edge of a subdivision. Fixed wireless can cover a hex on paper while a valley floor or tree line blocks the signal. Satellite can cover everywhere in the map sense while remaining a different product on latency and fair-use policy. OpenStats city pages aggregate nearby hex and community evidence to a municipality, then label the match as direct or regional when confidence falls.

Throughput, latency, and what speed tests measure

Advertised download and upload figures describe peak throughput a plan is engineered to offer under the provider's terms. Latency (round-trip time) and jitter matter for video calls, gaming, and some business applications. Cable and fibre usually deliver low latency; geostationary satellite historically delivered high latency; low-earth-orbit satellite narrowed that gap while remaining more variable than fibre. A single speed-test result mixes plan tier, Wi-Fi quality, device limits, and momentary congestion.

CRTC communications monitoring reports provide national and regional context on adoption, prices, and investment. ISED NBD remains the primary open geography for "what can be sold here." Household speed-test datasets from third parties can illustrate lived experience, yet they are volunteer samples with their own biases and are separate from the federal coverage map.

Competition and provider count

Speed tier and provider count answer different questions. A single fibre provider can offer excellent throughput with limited price competition. Several DSL or fixed-wireless providers can create choice without raising the maximum available tier. OpenStats pages report both where the source supports it: the fastest generally available advertised tier, and how many providers appear in the matched area.

Rural build economics and federal funding layers

Cost per premises passed rises quickly when homes sit far apart. Fibre and cable upgrades concentrate first where density supports private returns; fixed wireless and satellite fill many remaining areas until funded projects arrive. ISED's National Broadband Data separates services that are available today from government-supported and private expansion projects that describe planned capacity. The Universal Broadband Fund and related provincial programs exist to close that gap where the market alone moves slowly.

When reading a city or community point near a funded project, check whether the layer is marked as available service or as an expansion project. Ranking a town on a future fibre build before service turns up misstates what a household can buy this month.

Adoption surveys vs coverage maps

Statistics Canada household internet-use surveys ask whether people have a connection at home and how they use it. Those products measure adoption and behaviour. ISED NBD measures what providers report as available to sell. A region can show high advertised tiers in NBD while survey results show many households still on older plans, or the reverse in places where take-up of newly built fibre is fast. OpenStats internet pages follow the coverage map because that is the open geography for municipal comparison; treat StatCan survey releases as national or provincial context on how Canadians use the internet, not as a substitute hex map.

OpenStats rankings and the internet atlas

City pages on the internet atlas summarize the best advertised download tier and related technology flags for the matched hex and community evidence. The broadband series inside City Score uses that advertised-download signal; decade panels may backcast along a documented national speed curve with city-specific fibre-arrival paths, which is a modeling choice stated in the series notes, not a claim that every historical hex map was re-ingested for each calendar year.

Use the atlas for technology and provider context, the City Score pillar for relative standing among Canadian seed cities, and a provider quote at your civic address before you change plans.

CRTC universal service objective and funding

In Telecom Regulatory Policy CRTC 2016-496, the Commission set a universal service objective that Canadians in urban, rural, and remote areas should have access to broadband with at least 50 Mbps download and 10 Mbps upload, with an unlimited data option. CRTC high-speed broadband market reports track progress toward that objective and toward higher gigabit-class tiers. ISED's Universal Broadband Fund and the CRTC Broadband Fund support projects where private build-out alone has been slow. On the National Broadband Data map, funded projects appear as expansion layers until service is marked available. Ranking a town on a future fibre build before service turns up misstates what a household can buy this month.

Why Canadian municipalities differ on advertised tiers

Cost per premises passed rises quickly when homes sit far apart. Fibre and cable upgrades concentrate first where density supports private returns; fixed wireless and satellite fill many remaining areas until funded projects complete. Even inside one metro, fibre often stops at subdivision edges while adjacent rural municipalities remain on older copper or wireless. Provider count and maximum advertised tier answer different questions: a single fibre ISP can offer excellent throughput with limited price rivalry. OpenStats city pages report both where the source supports it, and label direct versus regional match confidence when hex evidence sits far from the built-up area.

Caveats when using OpenStats broadband figures

  • Coverage is an availability ceiling, not lived evening speed.
  • Separate available service from government-supported or private expansion layers.
  • Latency and fair-use policies matter for satellite and some wireless products.
  • StatCan household surveys measure adoption; NBD measures what can be sold.
  • Confirm plans at the civic address with providers before changing service.

Primary sources

Open internet data

Related: Methodology · Data sources · All guides