Heyroger Review: strong writing, slow experience (72/100)

Heyroger is a freemium product site reviewed from 38 crawled pages on 2026-10-10.

Fact	Value
Domain	heyroger.ai
Category	Unknown from supplied enrichment
Pricing	Freemium; 69–299 USD
Pages crawled	38
Crawl date	2026-10-10

Heyroger earns a 72/100 overall score from a mixed but clear evidence profile. Its strongest work is in writing quality (82/100), technical SEO (85/100), and risk and stability (88/100): the supplied review describes practitioner-led copy, a clean technical foundation, and no detected catastrophic traffic-loss mechanisms. Usability (78/100), editorial QA (78/100), accessibility (72/100), and design execution (74/100) add a solid base. The main constraint is performance, which scores 38/100. PSI records a 12.9 MB homepage load, 12.6-second mobile LCP, and 15.5-second time to interactive. Decision-support surfaces score 68/100, while review-content integrity also scores 68/100 because the comparison evidence is useful but lacks visible methodology, date, or provenance. The result is a site with credible content and foundations whose mobile experience needs focused improvement.

Heyroger offers clear routes for Pricing, Get started, Claude Code vs Roger, Blog, and About, but mobile performance delays the moment those routes become usable. Mobile PSI records 5.3 s FCP, 12.6 s LCP, and 15.5 s time to interactive. The homepage also presents Agencies, Pricing, MCP, Skills, multiple signup links, and Sign in, which creates competing destinations for a first-time visitor. Give one signup action clear priority, group secondary routes by intent, and complete the standalone label “Free signup with code:”. The usability evidence comes from simulated walkthroughs and a crawl of 38 of 40 pages, not live user testing, so the route structure is clear but the interaction verdict remains medium-confidence.

Heyroger has a solid semantic baseline, including header, navigation, main, and footer landmarks, and the homepage PSI accessibility score is 95. The verified gaps are specific: PSI reports a homepage color-contrast failure, the pricing outline jumps from h1 to h3 and later to h4, and 3 of 7 pricing inputs have no label or ARIA name in the crawl facts. The sampled homepage, pricing, and blog pages also report no skip link. Add a first-focusable skip link, restore sequential headings, and give each reported control an explicit accessible name in the rendered DOM. Then resolve the measured contrast pairs against WCAG AA. Keyboard order, focus traps, announcements, and error association were not supplied, so they remain verification work rather than findings.

Heyroger has a recognizable visual system: the font inventory reports one Roger family, with a 38px H1, 30px H2, and 16px body hierarchy. Sampled mobile layouts report a 390px scroll width, so horizontal overflow is not the dominant issue. The design debt is delivery and control sizing. Mobile PSI records 12,576.736 ms LCP and 15,473.645 ms interactive, while desktop and mobile runs report contrast failures. The mobile inventory also includes 42px menu buttons and 38–40px header links against a 44px target guideline. Prioritize render-blocking work, measured contrast fixes, sequential footer headings, and 44px hit areas before changing the visual language that is already distinctive.

Heyroger’s mobile lab performance is the main constraint: the homepage transfers 12,929 KB across 103 requests, reaches LCP at 12,576.7 ms, and becomes interactive at 15,473.6 ms. PSI estimates 2,700 ms of render-blocking savings and 344 KiB of unused JavaScript. Third-party transfers are material, including 2,145 KB from the testimonial embed and 1,817 KB from fonts. Desktop LCP is healthier at 0.9 s, but desktop total blocking time is still 805.7 ms. Identify the mobile LCP request first, defer testimonial and tracking work, reduce non-critical font work, split non-critical JavaScript, and serve slot-sized responsive images. No CrUX field metrics were returned, so this is a lab diagnosis rather than a 28-day real-user verdict.

Heyroger’s writing quality is strong because it names concrete paid-ad failure modes instead of relying on broad claims. The About page cites “Search Partners on,” “Advantage+ expanding the audience,” and “Display Network leaking into Search,” while the homepage states that Roger finds wasted spend, makes the ads, and waits for approval. Main marketing pages average under 15 words per sentence, supporting a direct rhythm. The cleanup is mainly metadata and image context: the same description is used on /, /pricing/, and /agency/, the homepage title is duplicated on /agency/, and 61 homepage plus 86 agency images lack alt text. Keep the specific voice, write page-specific metadata, and describe informative product screenshots while leaving decorative icons empty.

Heyroger’s Claude Code vs Roger page is a useful guided comparison, not a bare feature table: it frames “15 hard steps, or 3 easy ones,” connects the choice to Google and Meta paid-ad work, and provides a Get started CTA. It is not yet fully defensible because the supplied evidence shows no scenario where Claude Code is preferable, no dated methodology, and no normalized axes for cost, ownership, integrations, limits, support, security, or approval controls. Add an explicit choose-Claude-Code/choose-Roger block with tradeoffs, then use 8–12 buyer-relevant axes and a source date. Pricing also needs reconciliation: the page shows $99/month for up to 3 businesses, while Offer data contains Solo $69, Team $179, and Agency $299 tiers.

The Claude Code versus Roger comparison is specific enough to help, but the supplied evidence does not establish independent hands-on testing, a dated methodology, or source provenance for every comparative claim. “Honest comparison” therefore needs a visible evidence basis. The page cites “15 hard steps, or 3 easy ones” and “Same intelligence,” yet the crawl supplies no methodology or publication date for those claims. Add a compact block that separates observed page facts from vendor-provided statements, names the comparison date, links or qualifies material claims, and lists what remains unverified. Keep the Get started CTA, but do not imply independent testing unless it can be substantiated. This is a credibility upgrade, not a request for a large review apparatus.

Heyroger’s technical risk profile is strong: the supplied Mode V analysis found no sitewide noindex, public robots disallow, redirect loop, or client-side-only rendering dependency. Public marketing URLs return HTTP 200 with self-canonicals and sitemap inclusion; authentication routes carry noindex without affecting marketing pages. HTTP and WWW hosts redirect directly to the canonical HTTPS host, and the site serves /llms.txt at 3,443 bytes and /llms-full.txt at 8,830 bytes. The remaining risks are optimization risks: the homepage and /agency/ share an exact title, and choice-diy.png is 2,510 KB. Differentiate commercial-page metadata and optimize non-hero images. The analysis identifies stable foundations, not a guarantee against future regressions.

Heyroger’s editorial quality is strong: the supplied copy has attributed authors, attributed authors and legal-entity context, and no detectable generic AI phrasing. Regional blog indexes state plainly that posts are in English rather than presenting weak auto-translations. The highest-priority QA defect is factual consistency in structured data. Homepage FAQPage JSON-LD describes Solo at $69, Studio at $99, and Team at $179, while the visible interactive slider starts at $99/month for up to 3 businesses. The internal link graph also repeats generic anchors 25 times for “get started,” 25 for “commencer,” 25 for “empezar,” and 66 for “roger.” Sync the FAQ answer with live pricing, then vary anchors with contextual destinations such as agency pricing or Roger skills.

Heyroger has a clean technical SEO foundation: 100% of visible text across tested templates is present in raw HTML, HTTPS and host consolidation use direct 301 redirects, robots.txt references the sitemap index, and hreflang covers en, fr, and es. The main technical debt is asset weight and metadata duplication. choice-diy.png is 2,510 KB at a 280px display width, while step-configure.png is 1,823 KB at 192px; the homepage and /agency/ also share an exact title, and /, /pricing/, and /agency/ share a meta description. Convert heavy PNGs to WebP or AVIF with responsive sizing, then write distinct titles and descriptions for agency and pricing pages. Add descriptive alt text to informative images while keeping decorative icons empty.

Heyroger is a credible product site with unusually concrete writing, a recognizable visual system, and strong technical SEO evidence. The fixable weaknesses are concentrated: reduce the 12.9 MB homepage load and address the 12.6-second mobile LCP and 15.5-second time to interactive; make the comparison evidence more transparent by adding visible methodology, date, and provenance; and strengthen the decision-support surface beyond its current guided comparison. Heyroger is best suited to readers who value practitioner-led explanation and want a clear route through pricing, setup, and product context. The supplied enrichment does not establish a more specific audience.

This 13-dimension review combines the supplied crawl, PSI measurements, site facts, and public dimension score table. It covers 38 pages crawled on 2026-10-10 and includes 13 · Review-content integrity at 68/100. Supplied enrichment is pending for locale, audience, vertical, site shape, and business model; GSC is not connected. Read How SiteList scores for the methodology.