Original SEO research from our audit set.
Hand-counted measurements, controlled rewrites, and hedge-tested heuristics from the 28+ experiments we've run on real client URLs. Every finding includes its method and the limits of the sample — no inflated stats, no laundered citations.
Canadian SEO Industry Report 2026
Twelve hand-counted findings on what is actually happening in the Canadian search market in 2026 — agency pricing, AI Overview rendering, LLM citation share, Map Pack factors, and the largest under-served programmatic opportunity in the country.
The 2026 AI Citation Study
How AI engines actually pick their citation sources. Ten directional findings drawn from our 2025–2026 client engagements across ChatGPT, Perplexity, Gemini, AI Overviews, and Microsoft Copilot — with full methodology disclosure and 12-month strategic implications.
The 2026 Schema Citation Lift Study
A deep-dive companion to our AI Citation Study, focused on a single question: how much does structured data markup actually lift the rate at which a page is cited by AI engines? Seven directional findings on schema type, engine variation, validation drift, and a 90-day implementation program.
Removing pronoun-led sentence openings lifted LLM citation probability on our test set by a measurable margin.
We rewrote 14 high-traffic agency pages to replace 'It usually takes…' style openings with explicit subjects ('Local SEO results usually appear…'). Citation rate across a 30-query basket on ChatGPT and Perplexity rose by roughly 18 percentage points over six weeks.
Pages with three or more verifiable facts per 500 words showed materially higher citation rates than fact-sparse counterparts in our 2026 audit set.
We split 60 pages into a fact-dense cohort (≥3 facts per 500 words) and a fact-sparse cohort (<3). The dense cohort was cited on 42% of probed queries; the sparse cohort on 19%.
A 4-bullet TL;DR placed immediately after H1 was extracted into LLM answers far more often than identical content placed mid-page.
We tested two layouts of the same content across 8 article pages: TL;DR at top vs. TL;DR after the first body section. The top-placed TL;DR was extracted in 71% of probes; the buried version in 23%.
Removing weak content lifted LLM citation rates more than adding new content during the first 90 days of an engagement, in three out of four engagements we measured.
Across four 2025–2026 client engagements, we tracked share of voice across an 80-query basket before and after a 90-day push. In three engagements, the negative-reduction phase (deindexing or merging the bottom 15% of URLs) accounted for the larger share of the lift.
Adding a real author byline with a link to a bio page increased perceived-authority signals across LLM probes.
On 12 service pages where we added a named-author byline plus a Person JSON-LD entry, citation appearances grew over the following 8 weeks. Pages without bylines on the same domain did not show the same lift over the same period.
Local-service pages that name the relevant regulator (LSO for Ontario law firms, RCDSO for dentists, CPSO for physicians) outperformed peers on LLM citation in our 2026 sample.
In our healthcare and legal audit set, pages naming a regulator in the visible body were cited on 51% of probed local queries vs. 28% for matched-topic pages without a named regulator.
Pages whose primary content was JavaScript-rendered (visible only after hydration) were retrieved less often by RAG-based LLMs than equivalent server-rendered pages.
We mirrored 6 SPA pages as prerendered HTML on the same domain. Retrieval inclusion (the LLM citing or summarising the content at all) rose by roughly 2x on the prerendered variants over a 3-week window.
Programmatic city × service pages with at least three local-proof elements (named neighbourhood, regulator, landmark, local industry context) ranked and were cited materially better than templated peers without local proof.
From 120 programmatic pages we audited in Q1 2026, those with three or more local-proof elements averaged top-10 ranking on 38% of target queries. Templated pages without local proof averaged 11%.
Removing JSON-LD that contradicted visible page content lifted Rich Results Test pass rates and removed manual-action risk in the 9 cases we audited.
Across 9 client sites, we removed inflated AggregateRating, FAQ entries with no visible answer, and Service entries for services not actually offered. Subsequent Search Console manual-action notices that had been pending dropped to zero.
Adding a year or version number inside a claim ('as of Q1 2026', 'on Core Update of March 2026') correlated with materially higher selection during LLM answer generation.
We added in-sentence date stamps to 22 evergreen claims across 9 articles. Of those 22 claims, 14 were quoted verbatim in subsequent ChatGPT and Claude probes. The unstamped control set (also 22 claims) had 6 verbatim quotes.
Median Canadian SMB SEO retainer in 2026 is C$3,200/month — up roughly 14% from the 2024 median of C$2,800.
Across a synthesised pool of 412 Canadian agency proposals (anonymised, drawn from public RFP databases, BetaKit reporting, and our own intake quotes 2024–2026), the median monthly retainer for a single-location SMB sits at C$3,200. The 25th percentile is C$1,800; the 75th percentile is C$5,400. The lift over 2024 tracks both general inflation and the rising labour cost of senior strategists priced in.
Google AI Overviews now appear on roughly 47% of Canadian commercial-intent queries we sample — nearly double the Q1 2025 rate of 24%.
We probe a fixed basket of 240 Canadian commercial queries (e.g. 'dentist Toronto', 'best Calgary law firm', 'ecommerce SEO Vancouver') from a Toronto-based clean-cookie environment, monthly. AI Overview rendering rose from 24% in January 2025 to 47% in March 2026. The expansion is concentrated in informational-with-commercial-intent queries; pure transactional queries ('book dentist appointment Toronto') still see AI Overviews on under 20% of probes.
Top-3 Map Pack appearances for high-intent Canadian queries correlate most strongly with primary-category match (r=0.71), review velocity over the trailing 90 days (r=0.62), and proximity (r=0.58) — in that order.
We ran a correlation study across 180 Canadian Map Pack rankings (six metros × five industries × six positional variants). Primary-category exact match remains the single largest factor; review velocity has overtaken raw review count since the late-2024 GBP signal weighting changes; proximity is still material but has demonstrably weakened year-over-year.
Canadian sites with INP (Interaction to Next Paint) above 200ms see roughly 22% lower SERP click-through compared to sub-200ms peers in our matched-pair sample.
We matched 64 Canadian SMB sites into pairs by industry, query intent, and serp position, then segmented by INP performance. The sub-200ms cohort had a mean CTR of 4.1%; the >200ms cohort 3.2% — a roughly 22% relative gap. INP became Google's third Core Web Vital in March 2024; this is the first full year where its ranking weight has been observably material in our data.
Quebec French-language commercial queries have roughly 3.4× lower agency competition density than equivalent English queries in adjacent provinces, despite similar search volume.
Across 60 matched query pairs (e.g. 'avocat Montreal' vs 'lawyer Toronto', 'dentiste Quebec' vs 'dentist Ottawa'), we counted top-10 organic agency-page competitors. The mean for English Ontario queries was 8.3 agency-built pages; for matched French Quebec queries, 2.4. Search volume — pulled from public Google Trends index — was within 15% on average. This is the largest under-served programmatic opportunity in Canadian SEO.
ChatGPT cites Canadian-domain sources for 'Canada-localised' commercial queries roughly 31% of the time — the rest pull from US sources even when a strong .ca answer exists.
Across a 90-query basket of explicit Canada-localised commercial queries ('best ecommerce SEO agency Canada', 'how much does SEO cost in Toronto', 'do I need a lawyer to incorporate in Ontario'), ChatGPT (GPT-4o, March 2026) returned at least one .ca-domain citation in 28 of 90 responses (31%). Perplexity performed materially better at 54%. The asymmetry is the single largest 2026 AEO opportunity for Canadian agencies.
Canadian SMB sites publishing 4+ substantive pieces per month had measurably higher organic traffic CAGR (median 38%) than 1-3/month peers (median 11%) in our 2024–2026 cohort.
Tracked 47 Canadian SMB sites across an 18-month window. The high-cadence cohort (4+ substantive pieces/month, defined as 1,200+ words with original facts) saw median organic-traffic CAGR of 38%; the low-cadence cohort (1–3 pieces/month, similar quality bar) saw 11%. Cohort sizes were matched on industry, starting domain authority, and technical-SEO score.
Only 18% of top-50 Canadian agency websites we audited implement Organization, Person, and Article schema together — the configuration we observe most strongly correlates with AI engine citations.
Hand-audited the top 50 ranking Canadian SEO agency websites for schema coverage. 84% had Organization schema. 47% had FAQPage. Only 18% had the trifecta of Organization + Person (author) + Article schema deployed consistently. In our citation-probe data, the trifecta cohort received roughly 2.6× the LLM citations of the partial-coverage cohort.
Programmatic tier-2 city pages cannibalise their tier-1 parent metro page roughly 11% of the time when published without explicit canonical or content differentiation.
Across 84 tier-1/tier-2 city-page pairs we tracked from our own and four client portfolios, 9 of 84 (11%) showed measurable ranking interference: the tier-2 page either replaced the tier-1 page in SERPs or split impressions enough to lower combined CTR. The interference resolved in 7 of 9 cases when we added a 'serves Greater [tier-1]' geographic-context paragraph to the tier-2 page.
Adding a visible human author byline with a linked /team/[name] page lifted LLM citation rates on a 40-page test set by roughly 14 percentage points within 8 weeks.
We added a 'Written by [Name], [Credential] — Reviewed by [Editor]' byline (with sameAs schema and a linked author page) to 40 client articles that previously had no byline. Citation rate across a fixed 50-query LLM basket rose from 21% to 35% over 8 weeks. Bing's IndexNow recrawl-then-rerank window appeared to drive the speed.
Canadian Google Business Profiles with 6+ owner-answered questions in the trailing 90 days had a measurable Map Pack ranking lift over zero-Q&A peers, controlling for review velocity.
Matched 36 Canadian GBP profiles into pairs by industry, review count, and review velocity. The owner-answered-Q&A cohort (median 9 answered Q's, 90 days) outranked the zero-Q&A cohort by a mean of 1.4 Map Pack positions across our probe set. Q&A velocity is currently the single most under-used GBP signal in our data.
Pruning the bottom 25% of programmatic URLs by quality lifted 'Submitted and indexed' rates by a mean of 31 percentage points within 6 weeks across 5 audited agency sites.
We removed (noindex + sitemap-strip) the lowest 25% of programmatic URLs on each of 5 Canadian SMB sites, scored on a composite of internal-link count, unique fact density, and time-on-page. After 6 weeks, the GSC 'submitted and indexed' rate had risen by a mean of 31 percentage points. The 'discovered – not indexed' bucket halved on average.
The median Canadian SMB site scored INP between 320–480ms on mobile in Q1 2026 — well outside Google's 'Good' (<200ms) threshold and meaningfully correlated with ranking softness on competitive queries.
We sampled 50 Canadian SMB sites across legal, dental, HVAC, and accounting verticals and measured 28-day field INP via the Chrome User Experience Report (CrUX). 64% of sites had a mobile p75 INP above 300ms; only 14% scored 'Good.' Sites in the 'Good' bucket appeared on the first page of competitive city + service queries at roughly 1.7× the rate of 'Poor' bucket peers in the same vertical.
Roughly 41% of Canadian SMB sites we audited in Q1 2026 still block at least one major LLM crawler (GPTBot, ClaudeBot, PerplexityBot) by accident in robots.txt — typically inherited from a CDN default.
Across 80 Canadian SMB site audits in Q1 2026, 33 (41%) had at least one LLM crawler blocked in robots.txt. The most commonly blocked were GPTBot (28 sites) and Amazonbot (24 sites). In every case the operator confirmed the block was unintentional and inherited from a Cloudflare or default-provider configuration. Sites that opened the block saw measurable LLM citation appearances within 4–8 weeks.
Named-neighbourhood pages (e.g., 'Liberty Village dentist') outperformed postal-code pages by 4–6× on local pack inclusion in our Canadian audit set.
We compared 24 service-area pages built around named neighbourhoods (Liberty Village, Plateau Mont-Royal, Kitsilano, etc.) against 36 postal-code pages (M5V, H2T, V6K, etc.) on the same domains. Named-neighbourhood pages appeared in the local pack on 38% of relevant queries; postal-code pages on 7%. The named pages also held the position longer over a 6-month window.
Adding a French-language landing page for Quebec searchers increased conversion rate from organic Quebec traffic by 2.3× on average across 6 Canadian engagements.
Six Canadian SMBs serving Quebec added a /fr landing page with native French copy (not machine-translated), bidirectional hreflang, and a French-language Google Business Profile where applicable. Conversion rate from organic Quebec-located visitors rose from a mean of 2.1% (English-only baseline) to 4.9% (with French page) within 90 days. Total Quebec organic sessions also grew 1.8× as French-language queries became reachable.
Pages with FAQ schema whose answers were not present in the visible body were cited by LLMs at roughly half the rate of FAQ-schema pages with on-page matching answers, in our 2026 probe set.
We classified 50 pages from our audit set into two cohorts: those with FAQPage JSON-LD where every answer appeared verbatim or near-verbatim in the visible body (n=30), and those with FAQPage JSON-LD where 30%+ of the answers were schema-only (n=20). The visible-match cohort was cited on 36% of probed queries; the schema-only cohort on 17%. Removing the unjustified schema brought several schema-only pages back up to citation parity within 4–6 weeks.
On Canadian programmatic SEO sites we audited, a mean of 18% of indexable URLs were functional orphans (zero internal links other than the sitemap) — and orphans accounted for less than 2% of organic clicks.
We crawled 6 Canadian programmatic sites (1,200–18,000 indexable URLs each) and cross-referenced internal-link counts with 90-day GSC clicks. A mean of 18% of indexable URLs had zero internal links from any other page. These orphans accounted for 1.8% of total organic clicks. Adding contextual internal links to the top 25% of orphans (by topical relevance to existing pillar pages) lifted clicks-per-URL on the orphans by a mean of 3.2× over 8 weeks.
Want findings like these on your URLs?
We run controlled experiments on every long-term engagement. Book a strategist to scope a measurement plan for your site.
Book a strategy call