SEO Audit Checklist 2026: 47 Points Every Indian Website Must Pass

Most SEO audit checklists circulating online were written for Western markets, on desktop connections, with English-language content in mind. Indian websites face a distinct set of challenges: shared hosting performance on 4G networks, multilingual content without hreflang, citation inconsistencies on JustDial and IndiaMart, and the emerging pressure to optimise for AI Overviews. This 47-point checklist addresses all of these while covering the universal technical and content fundamentals that apply regardless of geography.

Work through this systematically. Each category has a clear scope and the points are ordered from highest to lowest impact within that category. Flag each item as Pass, Fail, or Not Applicable, then prioritise the failures by the estimated traffic impact before beginning fixes.

Category 1: Technical SEO (10 Points)

1. Crawlability verified in GSC: Check Google Search Console > Coverage report. Submitted URLs should exceed indexed URLs by less than 10%. If you have large numbers of excluded or error pages, diagnose before any other work.

2. XML sitemap submitted and valid: Your sitemap.xml must list only indexable URLs. Common failure: sitemaps include paginated URLs, 301-redirect destinations, or noindex pages. Validate with Screaming Frog's sitemap mode or xml-sitemaps.com validator.

3. robots.txt does not block critical resources: Check that your robots.txt does not accidentally block CSS files, JavaScript, or page sections that Google needs to render your content. Use Google's robots.txt tester in GSC. India-specific note: Many Indian shared hosting providers install default robots.txt files that block staging subdirectory paths — verify after any hosting change.

4. Core Web Vitals pass in India field data: Check PageSpeed Insights using your actual Indian users' field data tab, not the lab data. LCP under 2.5 seconds, INP (Interaction to Next Paint) under 200ms, CLS under 0.1. Indian 4G on mid-range Android devices makes LCP the most common failure point.

5. Mobile usability — no issues in GSC: GSC's Mobile Usability report should show zero errors. Text too small, clickable elements too close, and content wider than screen are the three most common issues on Indian SMB websites built on older WordPress themes.

6. HTTPS configured correctly site-wide: No mixed content warnings. No internal links using HTTP. Canonical tags pointing to HTTPS versions. HTTP-to-HTTPS redirect properly configured without redirect chains.

7. Page speed — server response time under 600ms: Use GTmetrix or WebPageTest from a Singapore server (the closest major Google data centre to India). Indian shared hosting frequently has TTFB (Time to First Byte) above 1 second, which penalises LCP disproportionately.

8. Canonical tags present and self-referencing on all key pages: Every indexable page should have a canonical pointing to its own URL. Missing canonicals on paginated pages and filtered product pages are frequent issues on Indian e-commerce sites built on Magento or older OpenCart versions.

9. Hreflang implemented correctly for multilingual content: If your site has Hindi, Tamil, Malayalam, or other Indian regional language versions alongside English, hreflang tags must correctly link the equivalent pages. Missing hreflang causes Google to flag content as duplicate and suppress one version from ranking.

10. No crawl errors for key commercial pages: 404 errors on service pages, product pages, or location pages directly forfeit ranking. Run a Screaming Frog crawl monthly and resolve 404s and 500 errors before other improvements.

Category 2: On-Page SEO (10 Points)

11. Title tags unique, under 60 characters, describe the page: Duplicate title tags across multiple pages are common on Indian WordPress sites that use the same post template without unique titles. Screaming Frog's "Duplicate" filter finds these immediately.

12. Meta descriptions unique, under 155 characters, complete sentences: A meta description truncated mid-sentence signals low editorial care to Google's quality systems. Every page should have a meta description that functions as a standalone sentence.

13. One H1 per page, describing page content: Multiple H1 tags on a page is a common issue on Indian sites using page-builder tools like Elementor or Beaver Builder, where the theme adds an H1 separately from the builder's heading widget.

14. Heading hierarchy correct (H1 > H2 > H3): Skip no heading levels. H3 should not appear before H2. This matters for accessibility and for Google's ability to parse your content's structure into featured snippets and knowledge panels.

15. Target keyword present in URL, H1, and first 100 words: The URL slug should use hyphens (not underscores), reflect the page's topic, and not contain dates that become stale. Keyword in the first 100 words of body text is a straightforward on-page relevance signal.

16. No keyword density above 3% for any single term: More than three instances of any exact phrase per 100 words of body text triggers Google's spam quality classifier. Use semantic variants and related terms to maintain topical relevance without repetition.

17. Internal links — key commercial pages receive links from content: Blog posts and informational pages should link to the relevant service or product pages where the topic connects naturally. Orphan pages with no internal links receive diluted crawl budget and ranking support.

18. Image alt attributes — descriptive, not keyword-stuffed: Every image on an indexed page needs a descriptive alt attribute. "web-development-kerala.jpg" as an alt is acceptable; "best web development company Kerala Trivandrum cheap" is keyword stuffing. Describe what the image shows.

19. URL structure — clean, logical, no session IDs or parameters: Product and service URLs should not contain dynamic query parameters that create URL variants. Indian e-commerce sites on older Magento or OsCommerce installations frequently have this issue.

20. Content length — key pages above 800 words of substantive content: Pages with fewer than 300 words of unique text are classified as thin content by Google's crawlers. Service pages, location pages, and product category pages should all have substantive descriptive text.

Category 3: Schema Markup (8 Points)

21. LocalBusiness or Organisation schema on homepage: Every Indian business website should have either LocalBusiness (with address, phone, openingHours) or Organisation schema on the homepage. Validate with Google's Rich Results Test.

22. Article schema on all blog posts: Article schema with datePublished, dateModified, author (Person type), and image is required for eligibility for Top Stories placement and Google Discover. Missing on most Indian blog posts using generic WordPress themes.

23. FAQPage schema on pages with FAQ sections: If a page has a FAQ section, mark it up with FAQPage schema. Each Question and acceptedAnswer must be specific to that page's topic — generic answers copied across pages violate quality guidelines.

24. BreadcrumbList schema on all pages below the homepage: Breadcrumb schema helps Google understand site structure and replaces the URL with a readable breadcrumb in search results. Present on every blog post, service page, and location page.

25. Review/AggregateRating schema — only where real reviews exist: Never add review schema with placeholder data. If your site legitimately aggregates reviews from verified customers, mark them up. Inflated or fictitious review schema is a spam policy violation.

26. Product schema with Indian Rupee pricing for e-commerce pages: Product schema must include priceCurrency: "INR" for Indian markets. Missing currency specification causes Google to suppress price-rich results in India. Availability must reflect real stock status.

27. SpeakableSpecification on FAQ and introductory content sections: Mark sections suitable for voice answers using SpeakableSpecification. Most relevant for news and informational content — business sites benefit indirectly by adopting the short-sentence format the spec requires.

28. Schema validation — zero errors in Rich Results Test: Run every page type through Google's Rich Results Test at search.google.com/test/rich-results. Warnings are acceptable; errors block rich result eligibility entirely.

Category 4: Content Quality (8 Points)

29. No thin content pages under 300 words that are indexed: Either noindex thin pages or expand them. Location pages for every district in Kerala that say only "We serve [City]. Contact us." are thin content and likely suppressing the whole domain's quality assessment.

30. No duplicate content across pages: Use Copyscape or Siteliner to identify internal duplication. Boilerplate paragraphs shared across multiple location pages are a common issue for Indian IT service companies that created city pages by cloning and finding-and-replacing the city name.

31. Content freshness — key pages updated within 12 months: Update the dateModified schema and add fresh data or statistics to key pages at least once a year. Outdated prices, case studies from 2019, or product features that no longer exist all weaken E-E-A-T signals.

32. E-E-A-T signals present on service and health/finance pages: Author bios with credentials, About pages with verifiable business history, team pages with real people, and external mentions in credible publications all contribute to Experience, Expertise, Authoritativeness, and Trustworthiness signals.

33. FAQ sections answer the actual question specifically: FAQ answers must directly answer the question in the first sentence. Answers that begin with "Great question! There are many factors..." before reaching the actual answer are a quality signal failure. Every answer should be usable as a standalone voice search response.

34. Featured snippet readiness — answer-first paragraphs on key informational pages: For queries where your page ranks in positions 2-10, check whether a featured snippet exists. If one does, rewrite your page's relevant section to place the direct answer in the first 40-60 words of the section, after a clear H2 or H3.

35. No banned phrase patterns that signal AI-generated content: Generic template phrases like "In today's digital landscape," "It's worth noting that," and "plays a crucial role" are pattern-matched by Google's March 2026 Spam Update quality classifiers. Review all content published after September 2024 for these patterns.

36. Word count on blog posts — 1,500+ words of unique substantive content: Informational blog posts under 800 words rarely compete for featured snippets or top-3 positions in 2026. Length must correlate with depth — 2,000 words of padding performs worse than 1,500 words of genuine insight.

Category 5: Local SEO (6 Points)

37. Google Business Profile claimed, verified, and fully completed: Profile completeness score should be 100%. All sections filled: hours, attributes, services/products, description (750 characters), primary and secondary categories, website link, and phone number.

38. NAP consistency across website, GBP, JustDial, Sulekha, and IndiaMart: Your Name, Address, and Phone number must be identical across every directory and citation. Even minor variations ("+91" vs "0" prefix, "Pvt Ltd" vs "Private Limited") create inconsistency signals that suppress local rankings. India note: JustDial frequently auto-creates listings from old data — check and claim your JustDial profile if you have not already.

39. Local citations on minimum 15 Indian directories: Beyond JustDial and Sulekha, ensure presence on Indiamart, Tradeindia, Grotal, LocalBanya, MakeMyTrip (for hospitality), Practo (for healthcare), and the Kerala-specific Yellow Pages Kerala directory.

40. Google reviews — minimum 20 reviews at 4.0+ average: Fewer than 10 reviews is a significant local pack ranking weakness in competitive Indian cities. An active review generation process (QR codes, WhatsApp follow-up links) is now a standard local SEO requirement, not an advanced tactic.

41. LocalBusiness schema — addressLocality matches GBP service area: The city in your LocalBusiness schema's addressLocality field must match your GBP's listed city exactly. Mismatches between schema and GBP data weaken local relevance signals.

42. Location-specific pages — unique content per city served: If you serve multiple cities, create a dedicated page for each with genuinely unique content: city-specific case studies, local landmark references, industry context relevant to that city's economy. Pages that differ only in the city name substitution are doorway pages and violate Google's spam policies.

Category 6: AEO Readiness (5 Points)

43. Answer-first content structure on key informational pages: Each H2 section should open with a direct, self-contained answer to the implied question in its heading — within the first two sentences. Background context and supporting evidence follow the answer, not precede it.

44. AI Overview eligibility assessed for primary keywords: Search your target keywords in Google (using an incognito Indian IP if possible) and note whether an AI Overview appears. If one does and your site is not cited, your content structure is the most likely reason. Improve answer-first formatting and add specific data points that AI systems can extract as citations.

45. Entity markup — your business/brand as a distinct entity in schema: Use sameAs in your Organisation or Person schema to link to your LinkedIn profile, Wikidata entry (if one exists), and authoritative directory listings. This helps Google's Knowledge Graph identify your business as a distinct entity rather than an ambiguous name.

46. Citation sources from authoritative Indian publications: Mentions in The Hindu, Economic Times, YourStory, Inc42, or industry-specific publications like Ayurveda Journal India establish external E-E-A-T signals that AI Overview systems use to assess source authority. Seek earned media placements, not paid-link insertions.

47. LLM visibility — your site accessible and not blocked to AI crawlers: Check your robots.txt for disallow rules blocking GPTBot, ClaudeBot, PerplexityBot, or GoogleOther (the crawler for AI Overviews). If these crawlers are blocked, your content cannot be cited in AI-generated answers, regardless of its quality. Unblock them deliberately unless you have a specific reason to exclude AI crawlers.

Frequently Asked Questions