Technical SEO naspram on-page SEO: ključne razlike

Definicije: šta obuhvata svaka disciplina

Technical SEO i on-page SEO imaju isti cilj — vidljivost u pretrazi — ali deluju u različitim fazama pipeline-a pretraživača. Koristite precizan jezik kad dijagnostikujete probleme: crawling se odnosi na otkrivanje i preuzimanje, indexing se odnosi na to šta pretraživači skladište, a ranking se tiče redosleda rezultata u SERP-u.

Technical SEO se fokusira na infrastrukturu sajta i servera koja omogućava pretraživačima da otkriju, renderuju, razumeju i indexuju stranice. Primeri: pravila u robots.txt, canonical tagovi, odgovori servera, sitemaps, structured data i faktori Page Experience koji utiču na rendering.

On-page SEO se fokusira na sadržaj i HTML signale koji prenose temu i nameru stranice korisnicima i pretraživačima: title tagovi, headingovi, glavni tekst, internal linking, meta descriptions, semantic markup, i usklađenost sadržaja sa search intent-om.

How the two interact across crawl, index, and rank

Problemi se često javljaju na preseku tehničkog i on-page rada. Dobro optimizovana stranica koja nikad nije indexovana zbog ograničenja crawling-a ili pogrešnih canonical signala neće se pojaviti u rezultatima pretrage; obrnuto, stranica koja se može indexovati ali ne odgovara search intent-u može loše rangirati uprkos savršenoj tehničkoj postavci.

Crawlability

Tehničke kontrole (robots.txt, meta robots HTTP headers, server responses) određuju da li crawleri mogu da dohvaćaju URL. Koristite server logove i zahteve sa user-agent-om koje crawler specificira da verifikujete ponašanje. Zapamtite: Google koristi mobilnu verziju kao primarnu osnovu za crawling and indexing. Since July 2024, Google crawls sites for Search with Googlebot Smartphone by default.

Indexation

Indexation odluke zavise i od tehničkih signala (canonical tags, noindex directives, sitemaps) i on-page signala (jedinstvenost sadržaja, relevantnost). Stranica blokirana noindex-om ili canonicalizovana ka drugom URL-u obično će biti isključena bez obzira na kvalitet sadržaja.

Ranking

Ranking je multi-signal. Tehnički faktori koji utiču na rendering i user experience (Core Web Vitals) i on-page signali relevantnosti oboje utiču na rank. Ali izbegavajte deterministički jezik: tehnički popravci omogućavaju da se stranice uzmu u obzir; relevantnost i autoritet oblikuju njihovu poziciju u rezultatima.

Mechanics: concrete areas to own and why they matter

Technical SEO mechanics

Ključne tehničke oblasti koje biste trebali upravljati ili proveriti:

• Robots.txt and HTTP headers: control crawler access and understand server responses.
• Canonical tags and redirect rules: consolidate duplicate content and prevent index confusion.
• Sitemaps and crawl budgets: surface priority pages and help discovery.
• Rendering and JS: ensure critical content is server-rendered or reliably hydrated; verify what search engines see after rendering.
• Schema and structured data: provide explicit signals for rich presentation and feature eligibility.

On-page SEO mechanics

Practical on-page responsibilities include:

• Title tags and headings that reflect user intent and include target terms naturally.
• Content structure, depth, and E-E-A-T signals: use factual sourcing where appropriate and clear author/brand attribution.
• Internal linking that distributes authority and helps users find related content.
• Meta descriptions and on-page snippets that increase click-through rate from SERPs.

Verification checklist: tools and exact checks

Separate checks you can run from outside the publisher (third-party checks) versus checks for pages you own. Use the right tool for each.

Checks for pages you own

1. Use Google Search Console URL Inspection to check crawl, index, and last fetch status. This is authoritative for your property.
2. Run the Rich Results Test and the Schema Markup Validator for structured data validation.
3. Measure Core Web Vitals in field data (Page Experience) and in lab with Chrome DevTools Performance panel.

Checks you can run from outside any site (useful for third-party publishers or audits)

1. Fetch and inspect HTML with curl. Example to retrieve headers only while impersonating Googlebot: curl -I -A "Googlebot/2.1 (+http://www.google.com/bot.html)" https://example.com/page
2. To fetch the rendered HTML (not just headers), omit -I and inspect the body: curl -A "Googlebot/2.1 (+http://www.google.com/bot.html)" https://example.com/page
3. Open the page in Chrome, use DevTools → Elements to inspect the rendered DOM, and DevTools → Network to verify resource responses and load order.
4. Use server-log analysis to confirm which user-agents visited and what HTTP responses they received.
5. Use the site: operator sparingly as a public signal that a page may be known to Google; it is not definitive for indexation.

Common mistakes and how to diagnose them

Misconfigured canonical tags

Symptom: the page does not appear in results despite being crawlable. Diagnose by comparing canonical tags, HTTP headers, and sitemap entries. Use a direct curl fetch of the page and inspect the <link rel="canonical"> value in the HTML.

Content not visible to crawlers due to client-side rendering

Symptom: crawlers fetch the HTML but the content appears only after JavaScript runs in the browser. Diagnose with curl (server HTML) and Chrome DevTools (rendered DOM). If critical content requires JS, consider server-side rendering or hybrid rendering.

Over-optimization or mismatched intent

Symptom: a page ranks but has low engagement or poor conversions. Review on-page signals: does the title and H1 match the user's task? Is the content depth aligned to the query type? Use query intent mapping before broad rewrites.

Practical overlap: handoffs, priorities, and workflow

Teams often split responsibilities: developers and platform engineers handle technical SEO; content and SEO strategists handle on-page. For efficient work, define clear acceptance criteria that include both technical and content checkpoints.

A pragmatic priority order for a page launch or audit:

• Ensure the page is reachable and returns a 200-range response for the intended canonical URL.
• Confirm canonicalization and sitemap inclusion so indexation is possible.
• Validate on-page relevance: title/H1 alignment, content structure, and internal links.
• Measure page experience and fix high-impact performance issues before large-scale publishing.

If you want a deeper walkthrough of the technical checks referenced above, Read the Technical SEO Guide

Link attributes and editorial context (brief)

Links on a page are both on-page signals and technical elements when they affect crawling. Use the correct rel attributes depending on context. Example syntax:

A standard link without a special rel value: example

For paid or sponsored placements use example; for user-generated content use example. There is no rel="dofollow" attribute — a normal link is simply a link without rel="nofollow", rel="sponsored" or rel="ugc".

FAQ