Screaming Frog SEO Spider is the technical audit workhorse Canadian practitioners rely on for crawling sites up to hundreds of thousands of URLs. This review covers licensing in CAD, feature depth for bilingual sites, and how to extract maximum value from both free and paid versions in a Canadian SEO context.
The free version caps at 500 URLs per crawl. For single-location service businesses, small e-commerce catalogues, or competitor homepage-and-category scans, that threshold works. You get full access to response codes, redirect chains, duplicate title detection, and XML sitemap generation. The constraint becomes binding when auditing multi-location franchises, content-heavy publishers, or any site with parameterized URLs that push past a few hundred indexed pages.
The paid licence—currently £149/year, roughly CAD 260 at recent exchange rates—removes the crawl cap and adds JavaScript rendering via integrated Chromium, custom extraction with XPath and regex, scheduled crawls, and API connections. If you audit more than two medium-sized sites per month, the licence pays for itself in time saved manually checking pagination, faceted navigation, or dynamically loaded content. The single-user licence is non-transferable but can be moved between your own machines. Multi-user licences exist for agencies running parallel audits.
Bilingual sites serving both English and French content require careful hreflang implementation and language-specific meta validation. Screaming Frog handles this natively—configure the crawl to follow all internal links, then use the Hreflang tab to surface missing reciprocal tags, incorrect language codes, or self-referential errors. Custom extraction lets you pull lang attributes from the HTML element or specific meta fields to verify consistency across /en/ and /fr/ subdirectories or domain pairs like example.ca and example.com.
For multi-regional setups targeting provincial audiences or US cross-border traffic, use the custom extraction and bulk export features to audit geo-targeting signals: rel=canonical pointing to the correct regional variant, Open Graph locale tags, and structured data with addressCountry set to CA. The bulk export combines crawl data with Google Analytics API traffic by page, letting you identify high-traffic pages with misconfigured regional signals that dilute relevance.
Paid licence holders get Chromium-based JavaScript rendering, critical for React, Vue, or Angular sites where content loads client-side. Toggle between 'Text Only' and 'JavaScript' modes in the configuration. The JavaScript crawl is slower—expect 1–3 seconds per URL instead of milliseconds—so use it selectively. Run a text-only baseline crawl first, then compare rendered HTML on suspect pages to identify discrepancies.
Common finds: product descriptions or review widgets invisible to the text crawler, navigation links in JS bundles that break internal linking structure, or canonicals rewritten dynamically. The rendering tab shows before-and-after HTML, so you can pinpoint where JavaScript execution changes indexable content. For sites that mix static CMS pages with JavaScript-heavy user dashboards, configure crawl limits and include/exclude URL patterns to render only the public-facing tree.
Default memory allocation is often too low for large crawls. On a 100,000-URL site, Screaming Frog may stall or crash halfway through if limited to 2 GB. Open System > Advanced and allocate 4–8 GB depending on available RAM. Crawl speed settings default to conservative politeness—5 URLs per second with 0.2-second intervals. For staging environments or local dev servers, increase to 10–20 URLs/second to cut audit time from hours to minutes.
Thread count affects parallel processing. More threads (up to your CPU core count) speed up crawling but can trigger rate-limiting on live production servers or shared hosting. Monitor the response time graph in real-time; if you see spikes or 503 errors, dial back speed. For large Canadian e-commerce sites on Shopify or BigCommerce, start conservative and ramp up after confirming the server handles load without throttling.
Connect Google Search Console to overlay impressions, clicks, and average position per URL. This surfaces high-impression pages with poor CTR that need title or meta description rewrites, and indexed URLs with zero impressions signaling orphaned or low-authority content. The Analytics integration pulls sessions, bounce rate, and goal completions, letting you prioritize technical fixes on pages that already drive traffic.
PageSpeed Insights API integration runs Lighthouse audits on each URL, appending Core Web Vitals and performance scores directly into the crawl export. Filter for pages with LCP above 2.5 seconds or CLS above 0.1, then cross-reference with server response time and image file size columns to diagnose root causes. This is especially valuable for Canadian retail sites competing in seasonal niches where page speed affects ad Quality Score and organic conversion alike.
Custom extraction uses XPath, CSS selectors, or regex to pull non-standard elements: schema.org markup, custom meta fields, or inline JSON-LD. Set up extraction rules under Configuration > Custom, then export the combined dataset with URL, title, H1, and your extracted fields in a single CSV. This replaces manual spreadsheet merges and reduces audit report prep time.
Common workflows: extract author names from byline markup to audit E-E-A-T signals on editorial sites, pull product SKUs and price from structured data to verify consistency, or scrape internal link anchor text to identify over-optimized exact-match patterns. Combine with filters—show only 200 OK pages with missing Open Graph images, or canonical chains longer than two hops. The bulk export becomes your single source for technical prioritization, feeding directly into developer tickets or content team briefs.
Screaming Frog Ltd. is UK-based, so invoices are in GBP and VAT does not apply to Canadian purchases. Pay via credit card or PayPal; the charge converts at your card's exchange rate, typically landing between CAD 210–240 for the annual licence depending on fluctuations. Renewal is manual, not auto-billed, and the licence key works across Windows, macOS, and Linux.
For Canadian agencies buying multiple seats, no volume discount exists below five licences, but multi-user packs (5, 10, 25 users) offer per-seat savings. Treat the expense as software subscription for CRA purposes—fully deductible in the year incurred. The company issues a PDF invoice immediately on purchase, sufficient for bookkeeping. No separate Canadian entity or reseller exists, so all support and billing go through the UK office, typically responsive within one business day via email.
Yes, if the site has fewer than 500 indexable URLs. You get complete access to crawl data, redirect chains, duplicate content detection, and sitemap generation. The main limitation is the URL cap—if the site has faceted navigation, paginated archives, or dynamic parameters that push URL count higher, you'll need the paid licence to see the full picture.
No. All invoices are in GBP (British pounds) from the UK-based company. The annual licence is £149, which converts to approximately CAD 210–240 depending on your credit card's exchange rate at time of purchase. There is no separate Canadian pricing tier or regional discount.
It crawls both language versions without special configuration, following internal links across /en/ and /fr/ subdirectories or separate domains. Use the Hreflang tab to validate reciprocal language tags, and set up custom extraction to pull lang attributes or meta fields. This ensures your French content properly signals to Google and your English pages don't accidentally canonical to the wrong language variant.
Text-only mode fetches raw HTML as Googlebot's initial request sees it—fast but misses client-side rendered content. JavaScript mode uses Chromium to execute scripts and render the final DOM, capturing React or Vue content but running slower. Use text-only for broad audits, then switch to JavaScript on specific page types where you suspect rendering discrepancies.
Yes. Connect via API under Configuration > API Access. You can pull Search Console impressions, clicks, and positions, and Analytics sessions and conversions, then export a combined CSV with crawl data and performance metrics. This works for any property you have access to, regardless of geographic location, including .ca domains and regional GSC properties.
For sites with 10,000–50,000 URLs, allocate 4 GB. Beyond 50,000 URLs, use 6–8 GB to prevent mid-crawl crashes. Open System > Advanced to adjust memory. If you're on a laptop with limited RAM, consider running overnight crawls with other applications closed, or use cloud-based alternatives like Sitebulb for exceptionally large crawls exceeding 100,000 URLs.