Pdf To Md Review: Strong PDF conversion (85/100)

Pdf To Md is a freemium PDF conversion service for Obsidian and Notion users.

Fact	Detail
Domain	pdf-to-md.app
Category	PDF conversion for Obsidian/Notion users
Pricing	Freemium; 5.75–7.9 USD
Pages crawled	40
Crawl date	2026-09-30

Pdf To Md presents a focused product for Obsidian and Notion users, and its strongest results come from clarity and execution. First impressions score 91/100, audience and messaging 92/100, usability 95/100, accessibility 100/100, and design execution 94/100. Together, those scores describe a site that explains its purpose plainly and provides a strong baseline experience.

The weaker results are narrower. Performance scores 78/100, with mobile LCP at 2.0s and INP at 300ms; the hero image is the main stated opportunity. Writing quality scores 82/100, while risk and stability, editorial QA, and technical SEO each score 78/100. The clearest buying friction is decision support at 45/100: the three-plan pricing grid does not explain which option to choose.

Across 40 crawled pages, the public surface was broadly crawlable and server-rendered, with 40 URLs returning 200.

PDF to MD makes its product and audience clear immediately: it turns PDFs into editable Markdown notes for Obsidian and Notion. The hero states a concrete outcome, while pricing tiers, usage limits, sample conversions, and feature descriptions supply proof close to the claim. The brand name matches the pdf-to-md.app domain, keeping recognition consistent. The main opportunity is competitive context: the site does not provide comparison pages, so visitors must infer how this focused workflow differs from alternatives. Maintain the explicit audience language and add carefully scoped comparison content if the product needs stronger differentiation. The positioning already has a clear, falsifiable promise; additional proof should reinforce it rather than broaden the category.

The messaging aligns strongly with a defined audience: Obsidian and Notion users see the product category and outcome in plain language above the fold. “PDF to Markdown” gives visitors a usable mental model, while sample input/output sections and pricing tiers connect the promise to practical use. The site avoids a jargon mismatch and answers critical questions within 2 clicks. The remaining opportunity is proof from customers: the current evidence is product-led rather than supported by named testimonials or company logos. Keep the direct vocabulary and audience-specific guides, then add customer evidence only where it can be verified. Any added messaging should preserve the distinction between free local conversion and paid cloud or OCR usage.

The site is broadly functional, but several key tasks lack a direct path. On the homepage, “Get Started For Free” leads to /auth/login and does not clearly signal a trial or signup action. The pricing page marks Pro as “Popular” without explaining who should choose it, the contact page provides a form without visible phone or email details, and the guides index has no search or filters. These are fixable path-clarity issues rather than broad functional failure. Link the primary homepage action directly to signup with matching copy, add plan-specific guidance, make support options visible, and provide guide discovery controls. The fixes should be tested against the exact tasks visitors are trying to complete.

The static accessibility audit gives the site a perfect 100/100: forms are labeled, images have appropriate alt attributes, contrast meets WCAG 2.1 AA thresholds, and keyboard navigation has no reported focus traps or untabbable controls. The audit also found no ARIA misuse, redundant roles, or hidden interactive elements. No remediation priority is indicated by the supplied checks. The remaining qualification is scope: automated static analysis did not verify dynamic content behavior, including live announcements or screen-reader interactions driven by JavaScript. Preserve the current labeling, contrast, focus, and keyboard patterns, and supplement them with manual checks for dynamic states as the interface changes.

Design execution is strong across hierarchy, tokens, and mobile behavior. The site uses a 16px base font, a 4px radius convention, one H1 per page, consistent heading levels, visible focus states, and no reported horizontal scroll or tap-target failures in the sampled mobile checks. Contrast also remains AA-compliant. The remaining defects are small consistency drifts: Free pricing uses 24px card padding while Pro and OCR Pack use 16px, some H2 elements use weight 500 while others use 400, and Local/OCR buttons use 4px radii while Get Started buttons use 8px. Standardize those component tokens to make the system more disciplined without changing the underlying layout.

Performance is strong on desktop but weaker on mobile, where LCP is 2.0s and INP is 300ms. The LCP element is the 85KB hero image, rendered at 606px wide from a 1040px source; it has no preload or fetchpriority hint in the supplied finding. Serve a responsive image size, prioritize the actual LCP asset deliberately, defer non-critical third-party scripts, and remove unused route code through tree-shaking or code-splitting. Validate the result on mobile after each change so the optimization does not trade away the clear hero experience.

PDF to MD uses concise, utility-focused copy that states capability, audience, and privacy boundaries quickly. The hero explains that users can turn PDFs into editable notes for Obsidian and Notion, start free in the browser, and choose paid OCR for scans. Marketing buzzwords are minimal, which keeps the promise concrete. Template quality is less consistent: /word-to-markdown, /image-to-markdown, and /html-to-markdown repeat the H2 “Questions people ask before converting” twice consecutively. The image-to-markdown page also presents an unverified sample-output message, and the guides index combines titles and descriptions inside H2 links. Remove duplicate heading wrappers, separate guide titles from previews, and replace placeholder output text with verified rendered content.

The pricing grid lists Free, Pro, and OCR Pack accurately enough to compare features, but it does not help a buyer choose. There is no “Best for” guidance, recommendation, or reasoning tied to user needs; the Pro “Popular” signal is unsupported. Several rows show checkmarks for every plan, and terminology varies between OCR pages and standard cloud conversions. Cost presentation also needs alignment: Pro displays $69 billed yearly while homepage schema lists $7.90 monthly, and the billing toggle does not clearly state the monthly-versus-annual tradeoff. Add segmented recommendations, normalize the feature axes, show the key differentiators, label billing states, and disclose the $69 annual total. This is the clearest buying-friction fix in the review.

The available evidence does not show a broad crawl or rendering failure: sampled public pages return 200, robots.txt permits the public surface, the sitemap is present, the nonexistent URL returns 404, and raw and rendered content match on sampled templates. The clearest stability and trust risk is protocol duplication: the HTTP homepage returns a full 200 instead of redirecting to HTTPS. The developers template adds a narrower mobile risk, reporting 1,911px scroll width at a 390px viewport across its localized equivalents. Enforce one permanent HTTP-to-HTTPS redirect, correct the developer layout at the narrow viewport, then connect GSC and analytics before attributing any traffic movement. No traffic diagnosis is supported because no symptom date or measurement data was supplied.

Editorial quality is strong in the core guides, API references, and privacy terms: the content is practical, avoids formulaic filler, and generally follows a usable heading sequence. The template layer confirms the content-quality issues identified in Writing quality: shared tool-page heading structure needs deduplication, and sample result boxes need verified rendered content. The German word-to-markdown path also contains English UI copy. Complete translation coverage for localized paths. These changes would improve trust and structural consistency without changing the editorial voice.

The sampled public surface is broadly crawlable and server-rendered: all 40 requested URLs returned 200, robots.txt allows public paths, the sitemap declares 96 URLs, the nonexistent URL returns a genuine 404, and /llms.txt returns 200. Raw and rendered word counts also match on sampled templates, including 1,269 words on the homepage, so primary content is available without a rendering dependency. The main technical issue is protocol consolidation: the HTTP root returns 200 instead of redirecting to HTTPS. The /pricing/ variant adds a 307 hop to /pricing, and developer pages report 1,911px scroll width at 390px across four locales. Redirect HTTP consistently, normalize non-slash URLs, and make the developer template responsive. These findings represent sampled evidence from 40 pages.

Pdf To Md earns an 85/100 because its audience, positioning, and core presentation are unusually clear. It is a good fit for Obsidian and Notion users who want editable Markdown notes from PDFs and prefer a focused, plainly explained service.

The main remaining weakness is pricing-page decision support. The unexplained “Popular” treatment is not enough decision support. Two secondary opportunities are improving the hero image’s loading priority and tightening the remaining writing, editorial QA, and technical SEO gaps identified in the review.

The product does not need a broader promise. It needs clearer buying guidance and a small set of targeted refinements around mobile performance and content quality.

This is a 13-dimension review based on the public dimension score table and a crawl of 40 pages on 2026-09-30. The review combines observed page content and experience checks across positioning, messaging, usability, accessibility, design, performance, writing, decision support, risk and stability, editorial QA, and technical SEO.

Google Search Console was not connected, so no Search Console enrichment was available. Risk and stability is therefore a vulnerability scan rather than a traffic-loss diagnosis.

Read How SiteList scores for the scoring method and data notes. Byline: a 13-dimension review of the sampled public surface.