Skip to content
traze

Step 4 · Publish

Technical optimization

A site that Google and AI crawlers can read, index and cite without obstacles — and that stays that way.

Traze crawls your site, lays the findings next to your Search Console data and sorts by what actually costs traffic. What it can fix in your CMS itself, it does after your approval. What lives in your theme, server or hosting comes back as a ready-to-run ticket for your developer. Then Traze checks that it worked.

What you get · concretely

What technical optimization covers

  • 01

    Crawl and index control

    Crawl depth, orphan pages, robots.txt, sitemaps, canonicals, noindex conflicts, redirect chains and 404s. Tied to index coverage from Search Console, so you see which errors actually keep pages out of the index.

  • 02

    Core Web Vitals per template

    LCP, INP and CLS from field data, grouped by page type: category, product, article. One fix in a template resolves hundreds of pages at once.

  • 03

    Structured data

    Product, Article, FAQ, Organization and BreadcrumbList schema, validated. Pages Traze publishes ship with schema included.

  • 04

    Internal link architecture

    Hub-and-spoke structure, click depth and anchor text. New pages get links from relevant existing pages, and orphaned pages get reconnected.

  • 05

    AI crawlability

    Whether GPTBot, PerplexityBot, ClaudeBot and Google-Extended are allowed and able to read your content: robots rules, rendering of JavaScript content and a maintained llms.txt.

  • 06

    International and multilingual

    Hreflang pairs, canonical consistency and language-specific sitemaps, so every market gets the right version and your pages don't crowd each other out.

What it is

What technical optimization means at Traze

Technical SEO is the floor, not the ceiling. A technically perfect site with no good content ranks nowhere, but a site with crawl traps, duplicate URLs or blocked AI bots drags all of its content down. So Traze treats the technical side as a standing condition: it measures, prioritises, fixes what it can and watches that it stays fixed. The first measurement is part of the holistic SEO analysis.

The difference from an audit comes down to two things. First, prioritisation: a list of three hundred findings helps nobody. Traze weighs every finding against the pages it touches and what those pages are worth in traffic and revenue. Second, follow-through: findings become either a change in your CMS that you approve, or a ticket a developer can pick up without an explanation. Run the free Tech Scan to see what's going on right now.

New compared with classic technical SEO is the AI layer. ChatGPT, Perplexity and Claude use their own crawlers and read a page differently from Google. Content that only appears after JavaScript, a robots.txt that shuts out AI bots or missing entity information means you simply don't exist in AI answers. Traze checks that layer separately. Read how AI visibility works.

For web stores there's one more layer: the product feed. Products Google Merchant Center rejects don't show up in Shopping, and increasingly not in AI shopping answers either. Traze checks the feed alongside the site. Test your feed for free with the Shopping Feed Tester.

How it works · step by step

How it works

  1. Step 1

    Crawl like a bot

    Traze crawls your site respecting robots.txt, renders JavaScript where needed and compares what it sees with the index coverage and Core Web Vitals data from Search Console.

  2. Step 2

    Prioritise by impact

    Every finding gets a score: how many pages it touches, how much traffic and revenue is attached, and how big the fix is. What returns the most per hour of work goes to the top.

  3. Step 3

    Change it in your CMS

    Metadata, canonicals, schema, internal links, sitemaps and redirects on pages Traze manages, it changes itself — as a proposal you approve, or automatically if you set it that way.

  4. Step 4

    Ticket the rest

    Template, server and hosting issues — a slow LCP caused by a hero image, a redirect chain in the web server — become a ticket with exact location, expected gain and acceptance criteria.

  5. Step 5

    Verify

    After every change Traze re-crawls the affected pages and follows index status in Search Console. A fix is only done when the effect is visible.

  6. Step 6

    Guard

    Deploys break things. Traze crawls periodically and flags regressions: an accidentally added noindex, a sitemap returning a 500, an AI bot suddenly blocked.

Why it matters

Why it matters

Every page Traze writes and publishes depends on the same foundation. If your category pages sit in a crawl trap or your sitemap is stale, that costs return on everything built on top. Keeping the technical side in order is the cheapest way to make the rest work. Especially with programmatic SEO, where hundreds of pages share one template.

AI engines are stricter than Google in what they can read. They render less JavaScript, have less patience with slow pages and rely on clear structure to work out what a page is about. A site that's technically fine for Google isn't automatically readable for an AI crawler.

And it keeps changing. New releases, new plugins, a colleague who accidentally sets a page to noindex. Without monitoring you only notice once the traffic is already gone. Performance monitoring keeps an eye on the technical signals daily.

Core Web Vitals grouped, not per single page
Per template
AI crawlers checked alongside Googlebot
4
re-crawled and verified
After every fix

Comparison · old way versus Traze

One-off audit versus Traze

One-off audit versus Traze
AreaOne-off technical auditTraze
RhythmOnce, then the report agesContinuous, with alerts on regressions
OutputSpreadsheet with hundreds of findingsList scored on traffic, revenue and fix size
ExecutionYou find a developer and explain itCMS changes after approval, dev tickets for the rest
VerificationNone; you hope it workedRe-crawl and index status after every change
SpeedLighthouse score of one pageField data per template, tied to real users
CrawlersGooglebot onlyGooglebot plus GPTBot, PerplexityBot, ClaudeBot and Google-Extended
ContextSeparate from contentTied to the pages Traze publishes and optimises

Use cases · where it works

Where it has the most impact

  • E-commerce

    Faceted navigation generating millions of URLs, product variants duplicating each other, missing Product schema and a shopping feed Google rejects. Traze handles the crawl and index side and checks your feed.

  • SaaS and B2B

    React, Vue or Next sites where content only appears after JavaScript, documentation with thousands of pages and pricing pages that need to be fast. Traze checks what bots actually see.

  • International sites

    Hreflang pointing to the wrong language, canonicals merging markets, sitemaps per country. A mistake here makes entire markets invisible.

  • Content-heavy sites

    Thousands of articles, pagination, tags and archives eating crawl budget. Traze decides what should and shouldn't be indexed, and reconnects old content to the structure.

  • Multiple domains or brands

    The same checks across every site in one account, with a separate priority list per domain and shared templates where possible.

Honestly · risks and limits

Honest about the limits

Traze changes what's reachable through your CMS or API. Your theme, your server, your CDN and your hosting contract are outside that; for those you get tickets, not miracles. Expect developer hours on Core Web Vitals improvements, even when the diagnosis is done.

A perfect score isn't the goal. A Lighthouse score of 100 brings no traffic if the content doesn't answer the question. Traze prioritises by effect and won't put a slow but well-performing page at the top of the list because a tool turns red.

Automatic changes are powerful, which is why they're off by default. Canonicals, redirects and noindex rules can fix a lot and break a lot in one go. That's why Traze works with per-change approval until you decide to automate a category of changes, and every change stays reversible from the log.

What you risk without monitoring: silent damage. A noindex in a template, a blocked AI bot or a sitemap that stops updating costs weeks of traffic before anyone notices. This is exactly the kind of error Traze flags within a day.

Try it yourself · free

See it at work on your own site

This free tool shows part of this capability. No login, instant results.

Tech Scan

Tech Scan: free technical SEO audit

Enter a URL. Traze crawls the start page plus the internal pages linked from it and lines up the issues quietly capping your rankings — each with its fix, and with a separate score for how well AI engines can read your site.

No login. We do not store your pages; the crawl stays within 26 pages and respects robots.txt.

Try:

Questions · about Technical optimization

Frequently asked questions

Not in your theme or server. Traze changes what's reachable through your CMS or API — metadata, canonicals, schema, internal links, sitemaps and redirects on the pages it manages — and delivers everything else as a ticket with exact location and expected effect. Every change is visible in the log and can be reverted.

Get started · self-serve or with a demo

Let Traze run this for you

Start on the Start plan yourself, or book a demo where we show this capability on your own site. Pricing is public, with a rolling monthly option.