Skip to content

02.03 / Grow

AI search and SEO.

Technical SEO and generative engine optimisation on one job: server-rendered content, correct crawler access, schema.org, hreflang and Core Web Vitals, plus measurement of citations in ChatGPT, Perplexity and AI Overviews. It starts with an AI visibility audit at a fixed price.

Book a callTell us about your project
characters of body text readable by AI crawlers across 65 URLs, up from zeroMalacca Tables
105,212
indexable URLs, Spanish and EnglishHotel Talavera
16
A tablet on someone's lap: a winery's AI search report, with what ChatGPT, Claude, Perplexity and Gemini answer to a real buyer question and the tasks for this week.

In detail

What you should know.

A hotel manager with reading glasses in her hair compares an AI assistant's list of recommended hotels with a crawler report, an old town street outside

For example: 105,212 characters of body text readable by AI crawlers across 65 URLs, up from zero. Malacca Tables

Book a callTell us about your project

What it is

Search has two front doors now. One is Google, where a page still has to be crawlable, fast and correctly marked up. The other is an assistant: ChatGPT, Perplexity, Claude, Copilot or Google's AI Overviews, which read a handful of pages and answer on your behalf. Polargate treats both as one job. Technical SEO is the floor, generative engine optimisation is the layer above it, and they share one requirement: the answer has to sit in the HTML the crawler receives, in plain language, before any JavaScript runs.

That is where most sites fail. GPTBot, OAI-SearchBot, ClaudeBot and PerplexityBot fetch raw HTML and do not execute JavaScript, so a client-rendered React site shows them an empty shell. ChatGPT in its fastest mode sees little more than the title, the H1 and the first couple of hundred characters. Bot protection at the edge can quietly serve those fetchers a challenge they cannot solve, and your analytics will never show it, because bots do not run the tracking script either.

How we do it

We start with an audit, not a proposal. We fetch every important URL the way a crawler does, with JavaScript off, and count the characters that are actually readable. We read robots.txt group by group, check the Vercel or Cloudflare rules in front of the site, verify hreflang return tags and canonicals, look at Core Web Vitals field data rather than lab scores, and check whether you exist in Bing at all, since ChatGPT and Copilot lean on that index.

What a visitor reads, and what an AI crawler receives

What a visitor reads, and what an AI crawler receivesA JavaScript site serves a full page to a browser that runs its code, and an almost empty shell to the AI crawlers that do not run it. A pre-rendered site serves the same complete HTML to both.A JavaScript siteVisitor in a browserruns the code, reads everything11,421The same siteGPTBot, ClaudeBot,Perplexitydoes not run the code14Pre-renderedVisitor in a browsersame complete HTML11,421Pre-renderedThe same crawlerssame complete HTML11,421
Characters of body text from one real news page, before and after. An assistant recommends what it could read, so the second number is the one that decides whether you are in the answer.

Then we fix things in the order that changes outcomes. Server-rendered pages first, so the content exists without JavaScript. Answer-first structure next: one H1, headings phrased as the questions buyers actually ask, a direct forty to seventy word answer under each. One schema.org entity graph, built from a single source of truth and serialised into the HTML, never injected at runtime. Both have a limit: a 2026 study of 1,885 marked-up pages against 4,000 controls found no measurable citation uplift from schema alone, and no engine has shown a citation lift from llms.txt. We ship both because they cost little and keep the site consistent, not because they are a lever.

What you get

What you take away.

Everything is written down before we start, and everything stays in accounts your company owns.

01Deliverables

  1. 01

    Crawler-truth audit: every key URL fetched with JavaScript disabled, readable characters counted page by page

  2. 02

    robots.txt rebuilt with explicit groups for search and answer bots, and a deliberate, separate decision on training crawlers

  3. 03

    Edge check on Vercel or Cloudflare so bot protection is not silently serving AI fetchers a challenge

  4. 04

    Server-rendered, answer-first pages: one H1, question headings, a direct answer under each

  5. 05

    One schema.org entity graph (Organization, WebSite, Service, Article, BreadcrumbList) written into the built HTML, never injected at runtime

  6. 06

    hreflang with return tags and x-default across every language, mirrored in the sitemap and its lastmod dates

  7. 07

    Core Web Vitals work driven by field data (LCP, INP, CLS), not lab scores

  8. 08

    Measurement stack: Bing Webmaster Tools and IndexNow, AI user agents in server logs, an AI referrer group in GA4, and a fixed monthly prompt panel

02How it runs

  1. 01 · Week 1 to 2

    Audit and baseline

    We fetch your site the way GPTBot, Claude, PerplexityBot and Googlebot fetch it, read Search Console and analytics, check the structured data against the visible text, and run a panel of buyer prompts to see where you appear today. The output is a written baseline and a ranked list of findings.

  2. 02 · Week 3 to 6

    Fix at the source

    We work in the repository: real HTML per route, the schema.org graph, hreflang, sitemap, robots and page speed. Every fix ships as a pull request with the test that proves it, so the same problem does not come back on the next deploy.

  3. 03 · Month 2 onwards

    Content and campaigns

    We publish focused pages that each answer one buyer question, in every language you sell in, and run paid advertising where it is the fastest way to test a message. Volume is set by the retainer, so you know in advance what a month contains.

  4. 04 · Every month

    Measure, report, repeat

    Each month we re-run the audit and the prompt panel, compare against the baseline, and report clicks, impressions, average position and assistant referrals in your client portal. Then we agree the next month's list with you before touching anything.

For example: 16 indexable URLs, Spanish and English. Hotel Talavera

Book a callTell us about your project

Who it is for

We don't take every project.

We would rather tell you now than halfway through. If you recognise yourself on the left, let's talk.

It fits if

  • Your site already exists and your buyers search on Google or ask ChatGPT
  • You want the fixes made in the code, not a keyword report
  • You are happy to measure every month against a recorded baseline
  • You can give access to the site, Search Console and analytics

It is not for you if

  • You want a guaranteed first position or a guaranteed mention in ChatGPT
  • You want to buy links or publish unreviewed text in bulk
  • You expect results within a week
  • Nobody can touch the code of your site, and there are no plans to change that

For example: 65 routes pregenerated at build time with their own metadata and JSON-LD. Malacca Tables

Book a callTell us about your project

Timing and budget

How we quote.

No price list and no hourly billing during a build. After a short call you get a written quote for your project.

Timeline
Audit delivered in about two weeks. The first implementation block usually runs three to six weeks, depending on whether the site already renders on the server. Citation measurement starts in month one and reports monthly.
Team
Our team runs the audit and the implementation, with content, translation or paid media when the project needs them. Andrés Ciordia, Chief AI Officer, leads how AI assistants read the site, and Pedro Ciordia, CTO, reviews the technical changes before they ship.
How the price is set
The AI visibility audit has its own fixed price, quoted privately after a short call. Implementation is quoted as a fixed-price block; ongoing work runs on a monthly retainer.
The stack
ViteReactTypeScriptVercelSupabaseSearch ConsoleGA4GitHub ActionsVitestPlaywrightCloudflareWordPress

For example: 27 published FAQs with FAQPage schema. Hotel Talavera

Book a callTell us about your projectAsk for your quote

Work

Where we did it.

Real projects. Open each one to see what we built and the figures we measured.

For example: 24 products migrated, across 19 series and 9 categories. Malacca Tables

Book a callTell us about your projectAll work

Same line

Other services in Grow.

Each service has its own page with what it includes, the usual timeline and the projects where we did it.

For example: 48 audit findings confirmed and fixed. Hotel Talavera

Book a callTell us about your project

FAQ

What people ask us.

How much does an AI visibility and SEO audit cost?
Polargate's AI visibility audit has a fixed price that depends on the size of the site and how many languages it publishes, and you get it in writing before anything starts. It covers the HTML crawlers actually receive, robots.txt, edge and firewall rules, hreflang, structured data, Core Web Vitals field data and your presence in Bing. Implementation is quoted separately as a fixed-price block once we know what the site needs. Show us the site and you get a private quote.
How long does it take to show up in ChatGPT or AI Overviews?
The technical fixes land in days. Once pages are server-rendered and the fetchers are allowed through, GPTBot, ClaudeBot and PerplexityBot can read your site on their next visit. Citations run on a slower clock, driven by authority, freshness and third-party mentions, so we measure monthly and look for movement across a quarter. Nobody can guarantee a citation or a ranking, and we do not.
Does llms.txt help my site get cited by AI?
Not on its own. Consumer crawlers do not request llms.txt, Google Search ignores it, and no study shows a citation lift from it. It is genuinely useful for coding and agent tools that read it deliberately. We generate it from the built HTML because it costs almost nothing and stays accurate, but the work that moves citations is server-rendered, answer-first content.
Will schema.org markup get me into AI Overviews?
Structured data helps machines read your page correctly, but it is not the lever people think it is. A 2026 study of 1,885 marked-up pages against 4,000 controls found no measurable citation uplift from schema alone. Visible text does the work. We write one accurate entity graph into the served HTML, mark up only facts a visitor can see, and skip the schema types Google retired.
Can you do this on WordPress, or only on custom builds?
Both. Malacca Tables ran on WordPress and WooCommerce before we rebuilt it, and Hotel Talavera is a Vite and React site pre-rendered with vite-react-ssg. The audit is platform-neutral because it only looks at what the crawler receives. On WordPress we fix rendering, robots, structured data and speed in place. If the platform itself is the blocker, we say so and price a rebuild separately.
How do you measure whether AI assistants are citing my site?
With server-side data, because AI bots do not run JavaScript and never appear in GA4 or Plausible. We track requests by user agent in your server logs, use Bing Webmaster Tools and its AI Performance report, add an AI referrer group in GA4 for the humans who arrive from chatgpt.com or perplexity.ai, and re-run a fixed panel of buyer prompts every month.

For example: 93 product variants verified one by one against the old catalogue. Malacca Tables

Book a callTell us about your project

Get started

Tell us what you want to build.

Websites, apps, custom systems and AI for companies that need them in production in weeks. Fill in the form or book a 30 minute call: a senior engineer answers within 48 business hours with a first take on scope, timeline and price.

Book a 30 minute call
What do you need?
Budget
When

At least ten characters.

Answer in writing within 48 business hours. No newsletter, no automatic sequence.

Rather answer the full brief?

First step

Discovery Sprint

Two weeks to close scope, architecture and price before anything is built.

No price list. You get a private quote, in writing, after a short call.

Then, to build, always at a fixed price

Ask for your quoteHow we quote

Not ready yet?

AI-native digital studio for mid-market companies

POLARGATE, S.L. · CIF B24895005
Calle Electrónica 6, 28027 Madrid, Spain