Understanding web visibility: ranking in search — and being cited by AI
An educational resource for B2B teams on SEO, AEO and GEO: how search engines and AI crawlers read your site. Lakora offers optional audit and implementation support.
About this resource
This website is an independent educational resource for B2B teams, covering web architecture, SEO, AEO and GEO. It explains how search engines and generative AI systems discover, read and cite web pages.
Lakora is a small consultancy that can optionally help implement these concepts: technical audits, website development and structured-data work for B2B organizations across the US, Europe, the UK and the Middle East.
The material here is written from practice: audits, content models and structured data work on live sites. Where a method only works with specialist help, the text says so rather than hiding it.
The explanatory material on this page is independent of our commercial service and may be cited on its own. Service information appears separately, in the pricing section, and is optional.
Global Engines & AI Indexing Rules
A 2026-edition guide to how pages get found, read and cited, written for the engines that matter now: Baidu, Google, Bing, Yandex, Perplexity, ChatGPT-Search and Gemini. Six modules, working through 200 practical knowledge points covering SEO, AEO, AIEO, GEO, GEO-Local and white-hat practice. Published one topic a day, and every claim carries a link to the official documentation it rests on.
10 of 200 published · a new topic every day at 09:30 US Eastern · 2026 edition with complete authoritative source links
What Is Mobile-First Indexing? How Google Crawls
Topic #10 - plain-English, sourced
Read the topic →Core Concepts
Crawling and indexing
Before a page can rank or be cited, a crawler has to find it and a search engine has to store it. Robots directives, sitemaps, canonical tags and internal links decide what gets discovered, what gets kept, and what is treated as a duplicate.
Entities and topics
Search engines and AI models reason about entities — a company, a service, a person — rather than keywords alone. Consistent naming, clear descriptions and shared identifiers make an organisation easier to verify and harder to misattribute.
Structured data
Schema.org markup describes a page in machine-readable terms: organisation, services, FAQ, breadcrumbs. It does not create rankings by itself, but it removes ambiguity about what a page is, who published it and what it answers.
SEO, AEO and GEO compared
SEO targets ranked result lists on Google and Bing. AEO targets retrieval and citation by assistants such as ChatGPT, Copilot and Claude. GEO targets generative answers such as AI Overviews and Perplexity. All three rest on the same foundation: crawlable, unambiguous, answer-first pages.
How AI engines choose sources
Answer engines favour pages that state the answer plainly, attribute it to a verifiable entity, and agree with other public sources. Publication dates, stated methodology and openly available data all affect whether a page is quoted rather than paraphrased.
Measurement
Visibility claims can be checked rather than believed. Server logs record which crawlers actually arrived, robots.txt and llms.txt show what they were allowed to read, and public data files show what each engine used. Every one of those checks can be run against any site, including this one: ours are published in full at /live/.
Live Signals From Our Own Network
A live demonstration of crawler behaviour: this dashboard shows bots visiting lakora.net, read straight from our server logs. You can watch how search engines and AI crawlers behave on a real website. These figures are an educational reference, not a benchmark for other sites. No visitor IP or client data is published.
Crawler visits — automated requests the site received today (UTC), hour by hourSearch & AI referrers — real visits arriving from Google, Bing, ChatGPT, Perplexity and similarToday’s overview — which engines are crawling today, and their shareCrawler share by engine — which engines account for the window of logs we haveTop engines in window — per-engine totals for that windowWindow summary — engine, hits, distinct IPs, last seen (UTC)
| Source class | Hits | Share |
|---|---|---|
| Search engines | 303 | 16% |
| AI retrieval / citation | 906 | 48% |
| AI training | 671 | 36% |
| Other / unclassified | 0 | 0% |
| Engine | Hits | IPs | Last |
|---|---|---|---|
| ClaudeBot | 1,422 | 13 | 10-08 04:20 |
| PerplexityBot | 918 | 13 | 10-08 04:20 |
| Bytespider | 507 | 14 | 10-08 04:20 |
| 502 | 20 | 10-08 04:20 | |
| OAI-SearchBot | 494 | 24 | 10-08 04:20 |
| CCBot | 491 | 13 | 10-08 04:20 |
| GPTBot | 490 | 14 | 10-08 04:20 |
| meta-externalagent | 482 | 14 | 10-08 04:20 |
| Amazonbot | 472 | 22 | 10-08 04:20 |
| ChatGPT-User | 469 | 13 | 10-08 04:20 |
Live monitoring of the sites we operate, filtered to Europe and North America — no client data is shown, and visitor IP addresses are never published. Country data is derived from visitor IP addresses and is approximate.
The Numbers Behind This Resource
Two measurements, both ours: what crawlers do on this site (read from the server logs) and what people do (counted by our own cookie-less script). We publish both in full - crawler signals on /live/, visitor statistics on /stats/ - with the same definition on the page and in the JSON.
Implementation Framework
Below is a framework for improving web visibility. If you choose to work with Lakora, these are the outputs we provide — and you can equally apply these concepts yourself without hiring anyone.
Technical audit, first
Crawlability, indexation, Core Web Vitals, schema coverage and AI crawler access, with a prioritised fix list scored by effort and impact.
Entity and content model
How your brand, services and people are described across the site, plus the schema that matches it — Organization, Service, Person, FAQ and breadcrumbs.
Answer-first page briefs
Page-by-page briefs written so a human gets the answer in the first screen, and so assistants can quote the page cleanly.
Implementation
We can work alongside your developers, or build the changes ourselves. Your call — the deliverable is the same either way.
AI visibility tracking
A monthly baseline across ChatGPT, Copilot, Claude, Perplexity and Google AI Overviews, using the prompts your buyers actually use, alongside rankings and organic traffic.
A written monthly summary
What changed, what moved, what we recommend next. Plain language, no dashboard you have to decode.
Typical timeline: audit delivered in 10–15 working days. First measurable movement in crawl and indexation within 2–4 weeks of implementation.
Recommended Workflow
1. Start with a technical audit
Assess your current site, content and entity profile before changing anything. This is where crawl, indexation and structured-data issues show up.
2. Entity-first content modeling & structured data
Define your organization, services and expertise as clear entities so search engines and AI models can understand and reference your content.
3. Track measurable signals
Track organic rankings, AI citations and qualified traffic to judge whether the changes worked.
Who We Work With
We partner with small and mid-sized B2B teams that don’t have in-house SEO or web specialists.
- B2B SaaS and technology companies
- Professional services firms
- Mid-market B2B teams without dedicated web & search marketing resources
We understand your constraints: limited team bandwidth, tight budgets, and the need to compete against larger players in an AI-changing search landscape.
“Lakora helped us rebuild our web foundation and optimize for AI discovery. We now see more qualified leads coming from both organic search and AI citations.”
— Composite scenario, built from our own patterns; no client data
Case studies
Case study: our own network under the same instrumentation
The same crawler instrumentation we install elsewhere runs on this site. Here is what it showed in its first two weeks, including the parts that were unflattering.
Read the case →Case study: making crawler traffic visible
A real instrumentation case: server-log crawler monitoring, the figures it produced, and why the crawl mix decides what can be cited.
Read the case →How Engagements Typically Look
Every card below is labelled, and the labelling is the point: card one is a real site we instrumented ourselves, the rest are composite scenarios drawn from patterns across our own work, with the ranges we aim at. No client data, no invented quotes, no borrowed screenshots.ther than client claims: your market, content and competition decide the outcome, and we will tell you honestly when we think a target is unrealistic.
Net manufacturer · instrumented · live dashboard
This card is a real, instrumented site. A manufacturer’s site with server-log crawler monitoring and the figures published: no analytics product, no cookies.
What it showed: 824 recognised crawler visits in 24 hours, and seven-day engine totals led by Amazonbot (7,647) and Bytespider (980) — figures anything else on the site can be checked against.
B2B SaaS · ~40 pages · 6 months
Starting point: strong product pages, no structured data, almost never cited by assistants.
Work: audit, entity model, rewritten service and comparison pages, schema, quarterly prompt tracking.
Range we work toward: 20–40% growth in qualified organic sessions; first consistent assistant citations within 3–4 months.
Professional services · 12 pages · 3 months
Starting point: trusted offline, invisible in search and assistants.
Work: technical fixes, service page rebuild, FAQ and schema, entity cleanup.
Range we work toward: top-10 placement for 5–10 commercial terms; citations for the firm’s name plus its two core services.
Mid-market manufacturer · 300+ pages · ongoing
Starting point: large legacy site, duplicate content, slow templates.
Work: crawl budget and template fixes, content consolidation, product data cleanup, ongoing technical SEO.
Range we work toward: priority sections indexed above 90%; sustained gains in crawl efficiency and organic revenue.
Pricing
If you decide to engage Lakora for hands-on implementation, these are our service options. Every package starts with a technical audit — which you can also run yourself using the guidance on this site.
Technical audit
from $1,450 USD one-off
- Full technical and content audit
- AI crawler and schema review
- Prioritised fix list (effort and impact)
- 45-minute walkthrough call
2 weeks. Best if you have an in-house team to implement.
Visibility retainer
from $2,400 USD / month
- Audit plus implementation
- Entity, schema and content model work
- Answer-first page briefs
- Monthly AI, ranking and traffic tracking
- Direct access to the consultant doing the work
3-month minimum. Most B2B teams start here.
Project build
from $9,000 USD per project
- Everything in the audit
- Rebuild or re-platform of key sections
- Content model, templates and migration QA
- Handover documentation for your team
4–10 weeks depending on scope.
Indicative ranges for typical engagements. We confirm a fixed scope and price in writing before any work starts. Fees exclude third-party costs such as hosting, tooling and licences. If a smaller job is the right answer, we will say so.
Who does the work
You work directly with the consultants doing the work — there is no account-management layer between you and the people making decisions. Engagements are led by a senior practitioner with a technical SEO and web engineering background, supported by a content strategist on briefs and editorial. We keep the roster small on purpose: it is the only way we can stay accountable for outcomes rather than activity.
Latest articles
Short, practical explanations of how search engines and AI answer engines find, read and cite websites. See all articles →
Five Layers of Scanner Defence, Tested on Our Own Site
Edge blocking, server hardening, behaviour analysis, honeypots and public reporting - the series in one model, with the one place where our practice diverges from the usual advice for each layer.
Read the article →Layer 5: Closing the Loop in Public
The fifth layer is not a control at all - it is the reporting. Publishing refusal counts next to their definitions is what turns four layers of defence into something a reader can check.
Read the article →Layer 4: Honeypots as Early Warning (and Their Limits)
A decoy path nobody legitimate would ever request is a high-signal tripwire. It is not a fingerprinting laboratory - here is what we actually learn, and the two rules that keep a honeypot from becoming a liability.
Read the article →Layer 3: Behaviour Analysis Without Hurting Your Crawlers
Rate limiting and automatic bans sound like precision tools. On a site that depends on being crawled, the precision is mostly in the whitelist - and that is exactly where they get dangerous.
Read the article →Layer 2: Server Hardening, and the 403-vs-444 Question
Hardening is the layer you control completely - and the one where the textbook advice (drop the connection with 444) collides with the value of a published, verifiable refusal.
Read the article →Free tools
Thirteen free tools we built and use ourselves. All of them run without sign-up, and none of them uploads your data. The three below are the ones we reach for first.
AI Visibility Prompt Tester
Turn your product and market into five buyer-style questions, then run them yourself in the AI assistants you care about and record whether your brand appears. It writes the prompts and nothing else.
AI Crawler Radar
Enter a domain and see which AI crawlers are allowed to read it, whether the structured data is machine-readable, and what is blocking AI citation. It fetches only the site's robots.txt, llms.txt and homepage.
Schema.org JSON-LD Generator
Fill in your brand details and copy ready-to-paste Organization, WebSite, WebPage, BreadcrumbList, FAQPage and Service or Product JSON-LD. Everything is generated in your browser.
Frequently Asked Questions
The questions below are educational explanations of SEO, AEO and GEO concepts. They can be referenced independently of Lakora’s services.
What is AEO (Answer Engine Optimization)?
AEO, also called AIEO, is the practice of structuring content so AI assistants and answer engines such as ChatGPT, Copilot and Claude can retrieve it easily and cite it as a source. It focuses on answer-first structure, clear entities and schema markup.
What is GEO (Generative Engine Optimization)?
GEO focuses on getting content featured and cited inside generative search results such as Google AI Overviews and Perplexity. It combines content modelling, entity strategy and structured data so a brand appears inside AI-generated answers.
How is AEO different from traditional SEO?
Traditional SEO optimizes for ranking in search result lists on Google and Bing. AEO optimizes for being retrieved and cited by AI assistants and answer engines. The two reinforce each other: a clean SEO foundation is what makes AEO possible.
Do I need an in-house SEO team to work with Lakora?
No. The framework on this site is written for small and mid-sized teams without in-house web or search specialists, and it can be applied independently. Technical SEO, content modelling and AI visibility work can also be handled by an external specialist.
How long before I see results from SEO, AEO and GEO work?
It depends on the starting state of the site and the competition. Technical fixes often show up in crawl and indexation within two to four weeks; ranking and citation changes usually build over three to six months.
Which platforms do you optimize for?
Search engines include Google and Bing. AI assistants include ChatGPT, Copilot and Claude. Generative answer engines include Google AI Overviews and Perplexity. The same principles apply to other major platforms as they emerge.
How much does an engagement cost?
Technical audits start at $1,450, retainers at $2,400 per month and project builds at $9,000. Every engagement is scoped and priced in writing before work starts, and we will tell you if we think you do not need us yet.
Who will work on my project?
You work directly with the consultants doing the work. There is no account-management layer, no handover to a junior team after the pitch. We keep a deliberately small roster so accountability stays with the people you speak to.
Contact
General enquiries: [email protected]
Business hours: Monday to Friday, 09:00 – 18:00
Ready to audit your website for search and AI discovery?
Get in touch today.