The Website Visibility Field Guide

Introduction: Why This Guide Exists

For twenty years, "getting found online" meant one thing: ranking in a list of ten blue links. You picked keywords, optimized pages, built links, and watched your position climb or fall. The game had a scoreboard, and everyone agreed on what winning looked like.

That world is gone. Not the fundamentals — those are more important than ever — but the scoreboard.

Today, a prospective customer might discover your business through a traditional search result, a map listing, an AI-generated answer that cites you (or doesn't), a recommendation surfaced inside ChatGPT, a comparison thread on Reddit, a YouTube video, a LinkedIn post, or a review platform they trust more than any of the above. Visibility is no longer a ranking. It is a presence across every surface where your customers research decisions.

Here is the idea at the center of this entire guide:

Your website does not need more random traffic. It needs to become a trusted, discoverable resource for the people most likely to become customers.

That distinction matters more now than at any point in the history of search. Traffic that never converts was always a vanity metric; in an era where AI answers absorb a growing share of informational queries, chasing raw traffic is a strategy built on sand. The businesses that win are the ones that become the source — the site that search engines index confidently, AI systems cite, and customers trust when it's time to decide.

Rankings alone are no longer enough because rankings were never the goal. They were a proxy for the goal: being present at the moment someone is trying to solve a problem you solve, and being credible enough that they choose you. When the surfaces multiply — search results, AI answers, maps, social, video — the proxy breaks down, but the goal doesn't change. Visibility earns attention. Attention builds trust. Trust drives evaluation. Evaluation produces action. Every chapter in this guide serves some part of that chain, and anything that doesn't serve it has been deliberately left out.

Who this guide is for. This is written for business owners, marketing leads, and in-house teams responsible for a website that is supposed to produce customers — service businesses, professional practices, local companies, institutions, and organizations of any size where the website is a business asset rather than a hobby. You don't need a technical background. Where technical work is required, I'll tell you what to ask for and how to verify it was done.

How this guide is maintained. Unlike a blog post that captures one moment in time, this is a living document. Search and AI discovery are changing quickly, and advice that was sound eighteen months ago can be actively harmful today. I review the full guide quarterly, update individual chapters immediately when platforms make significant changes, and log every meaningful revision at the bottom of the page. When a recommendation is new or still being validated, it's labeled that way. You should never have to wonder whether you're reading current guidance.

One more note before we begin. Google's own position — stated plainly in its documentation — is that visibility in its AI-powered search features is an extension of sound SEO, not a separate discipline with separate tricks. Its generative results still depend on crawlability, index eligibility, useful content, and the same core quality systems that govern classic search. That framing runs through this entire guide: there is no secret "AI SEO." There is doing the fundamentals well enough that every discovery system — human or machine — can find you, understand you, and trust you.

What Matters Most Right Now

The current state of website visibility, in two minutes. This section is updated each quarter so the rest of the guide stays stable.

Still foundational

These have mattered for years and continue to carry most of the weight:

Clear website positioning — a site that says plainly who it serves and what it does
Crawlable, indexable pages — nothing else works if discovery systems can't reach your content
Strong information architecture — organized around customer journeys, not your org chart
Search intent alignment — pages that match what the searcher actually needs
Original, useful content — written for people, with something to add
Internal links — the connective tissue that distributes authority and guides visitors
Relevant external recognition — links and mentions from sources that matter in your space
Fast, usable mobile experiences — most of your visitors are on a phone
Accurate measurement — you cannot improve what you're measuring wrong

Increasing in importance

These separate the sites gaining visibility from the sites losing it:

Demonstrable first-hand experience — proof you've actually done the work you describe
Brand and entity clarity — discovery systems knowing exactly who you are, everywhere
Content that answers complex questions — not just definitions, but decisions
Original examples, data, processes, and opinions — things no AI summary can replicate
Visibility across multiple platforms — search, AI answers, video, communities, reviews
AI crawler accessibility — deliberately allowing the crawlers that feed AI search
Content built for comparison and evaluation — because that's how buyers actually research
Conversion paths connected to informational content — resources that lead somewhere

Commonly overstated

Spend less time and money here:

Publishing volume for its own sake — fifty thin posts lose to five substantial ones
Arbitrary content length — word count is not a quality signal
Keyword density — a relic; write like a person
Mass-produced AI content — Google's scaled content policies explicitly target this
Special "AI SEO" markup — Google states no special markup or llms.txt file is required for its generative search features
Link-building measured only by quantity — a hundred junk links lose to one real mention
Obsessing over individual ranking factors — systems evaluate pages holistically; you should too

If you read nothing else in this guide, read that third list twice. The most expensive mistakes I see aren't things businesses fail to do — they're things businesses pay for that were never going to work.

Part One: Build the Right Foundation

Chapter 3Start With the Business, Not the Keywords

Most visibility projects begin in the wrong place. Someone opens a keyword tool, exports a list of high-volume phrases, and starts assigning them to pages. Six months later, traffic may even be up — and leads haven't moved.

The problem isn't execution. It's sequence. Keyword research is a middle step, not a first step. Before you can decide what to be visible for, you have to be honest about what the business actually needs.

Effective optimization begins by understanding:

Who the business serves. Not "everyone who needs plumbing" — the actual customers who are profitable, pleasant to work with, and likely to return or refer. Most businesses have a clear picture of their best customer and a website that speaks to nobody in particular.

What problems it solves. Customers don't search for your services; they search for their problems. A firm that sells "digital transformation consulting" gets found by people searching "why is our checkout abandonment so high." The gap between how you describe your work and how customers describe their pain is where visibility is won or lost.

Which services or products matter most. Every business has offerings it would happily do all day and offerings it tolerates. Your visibility strategy should be weighted accordingly. It is entirely common — and entirely fixable — for a company's best-margin service to have the weakest page on the site.

Geographic reach. A business serving a 30-mile radius and a business serving a national market need fundamentally different architectures. Pretending to be both dilutes each.

Profitability and capacity. Visibility that generates demand you can't serve, or work you lose money on, is worse than no visibility. This sounds obvious. It is routinely ignored.

The sales cycle. A $200 purchase decided in an afternoon and a $200,000 engagement decided over nine months require completely different content. The longer the cycle, the more your website must support evaluation, comparison, and trust-building — not just discovery.

What constitutes a qualified lead. If sales and marketing can't agree on this, no analytics setup will save you. Define it in writing before you optimize anything.

Why someone should choose this business. If the honest answer is "no particular reason," that is the first problem to solve — and it's a positioning problem, not an SEO problem.

The Visibility Foundation Worksheet

Before touching a keyword tool, answer these six questions in writing. Short answers. Real answers.

What do we most want to be known for...
Who is the most valuable audience...
What triggers that audience to begin searching...
What questions do they ask before making a decision...
What evidence would make them trust us...
What action should they take next...

Every subsequent decision in this guide — site structure, content priorities, technical investments, measurement — flows from these answers. When a visibility program stalls, it's usually because one of these questions was skipped or answered dishonestly.

This is also where SimQua's approach diverges from tactical SEO. Visibility is a business strategy problem before it becomes an optimization problem. Optimizing a website nobody should choose just helps more people not choose it, faster.

Chapter 4Understand Demand and Search Intent

Once you know what the business needs, you can study what the market wants. The goal of this research is not a spreadsheet of keywords. It's a map of demand: what your future customers are trying to figure out, in their words, at each stage of their decision.

Where demand evidence lives

No single source tells the whole story. Triangulate across:

Keyword research tools. Useful for scale and comparison — which topics have meaningful demand, which are seasonal, which are climbing. Treat volume figures as directional, not precise.

Search Console data. The queries your site already appears for, including the ones you never targeted. Some of the best opportunities are queries where you rank on page two without trying.

Paid search data. If you or competitors run ads, this is demand data with money attached — which queries actually convert, not just which ones get searched.

Sales questions. The questions prospects ask on calls are the questions other prospects are typing into search bars and AI chats. Have your sales or intake team log them for a month. This list is usually more valuable than any tool export.

Customer interviews. Ask recent customers: what were you trying to solve, where did you look, what did you search, who else did you consider, what almost stopped you... Five interviews will reshape your content plan.

Search result analysis. Search your important queries and study what actually ranks. The results tell you what Google believes the intent is. If you want to rank a service page for a query where every result is a comparison guide, you've misread the intent.

Competitor visibility. What are competitors visible for that you aren't... Where they're winning with weak content, you have an opening.

Online communities. Reddit threads, industry forums, Facebook groups. This is demand in its rawest form — unfiltered questions, frustrations, and the exact vocabulary people use before they've learned the industry's terms.

AI prompts and conversational questions. Ask ChatGPT, Gemini, and Perplexity the questions your customers would ask. Note what gets recommended, what gets cited, and how the question itself differs from a typed search. Conversational queries are longer, more specific, and more decision-oriented — and they're a growing share of how people research.

The intent framework

Every query carries intent — what the searcher needs at that moment. Match content to intent, and everything gets easier. Mismatch it, and no amount of optimization helps.

Intent	What the visitor needs	Typical content
Discovery	Understand a problem	Guides, explanations, definitions
Evaluation	Compare approaches	Comparisons, alternatives, cost guides
Validation	Confirm credibility	Case studies, reviews, proof, credentials
Action	Contact or purchase	Service pages, product pages, booking pages

Most websites over-invest at the edges — a pile of discovery blog posts and a contact page — and leave the middle empty. But evaluation and validation content is where decisions actually get made, where AI systems find the comparison-ready material they synthesize from, and where your most qualified visitors spend their time. If your site has no honest cost guide, no "how to choose" content, and no substantial proof, you've built a funnel with the middle removed.

One reframe to carry forward: keywords are evidence of demand, not instructions for repeating phrases. Google's guidance still recommends using the language people actually search in prominent places — titles, headings, opening paragraphs — while keeping the content useful and written for people. Use the customer's vocabulary because it's the customer's vocabulary, not because you're feeding a machine. The era of mechanical keyword placement is over; the era of speaking your customer's language never ends.

Chapter 5Create a Website Information Architecture

Here is an uncomfortable exercise: open your website's navigation and ask whether it's organized around your customers' decisions or your company's org chart. Most sites fail this test. They mirror how the business thinks about itself — departments, service lines, internal jargon — rather than how a customer moves from "I have a problem" to "I'm ready to talk."

Information architecture is the discipline of organizing a site around topics and journeys. It matters for three audiences at once: visitors, who need to find their path without thinking; search engines, which infer what you're about from how pages relate; and AI systems, which extract and synthesize more reliably from clearly structured sites.

The pages a complete site needs

Core service or product pages. One page per meaningful offering — not one page listing everything you do. Each is a landing page in its own right, built to satisfy action intent.

Supporting topic pages. The guides, explanations, and answers that address the questions surrounding each service. These capture discovery and evaluation intent and funnel it toward the core pages.

Location pages — where legitimate. If you genuinely serve distinct locations with distinct teams, offices, or service realities, location pages earn their place. If you're generating fifty near-identical pages for cities you've never worked in, you're creating the kind of scaled, low-value content Google's policies target. The test is simple: does this page contain anything true and specific about serving this place...

About, team, and trust pages. Not an afterthought. Who is behind this site is now a first-order question for both customers and quality systems. (Part Four covers this in depth.)

Case studies and proof. Validation intent needs a home. Real projects, real outcomes, real specifics.

Conversion pages. Contact, booking, quote request — pages with one job, designed for that job.

Resource or field-guide content. Substantial reference material (like this page) that earns links, feeds AI citations, and demonstrates expertise at depth.

A simple model

The structure most business sites need is not complicated:

Homepage

  → Primary service categories

    → Individual services

      → Supporting questions and resources

  → Case studies and proof

  → Contact or conversion

Visitors and crawlers alike should be able to move from the homepage to any important page in a few obvious clicks. Breadcrumbs should make the hierarchy visible. And internal links should do deliberate work: every supporting article links to the service it supports; every service page links to its proof; every page suggests a sensible next step.

The five-part test for every important page

Before a page earns a place in your architecture, it should have:

A defined audience — who specifically is this for...
A distinct purpose — what does it do that no other page does...
A clear search intent — discovery, evaluation, validation, or action...
A place in the hierarchy — where does it live, and what links to it...
A logical next action — where should the reader go from here...

Pages that can't answer all five tend to become the site's dead weight: orphaned posts, duplicate near-misses, and pages that compete with each other for the same query. Fewer pages, each with a job, beat more pages with none.

Part Two: Make the Website Discoverable

Chapter 6Technical Search Accessibility

Technical SEO has a reputation problem. It's either dismissed as plumbing or fetishized as the secret to rankings. It's neither. Think of it as accessibility for machines: the work that ensures every discovery system — Google, Bing, AI crawlers — can find, read, and correctly interpret your pages. When it's right, it's invisible. When it's wrong, nothing else you do matters.

One framing to internalize first: Google describes crawling, indexing, and serving as separate stages. A page can be crawlable and still not get indexed. It can be indexed and still not rank. Accessibility earns you eligibility, not results. This is why technical work is necessary but never sufficient — and why "we did technical SEO, where are the rankings..." misunderstands what the work does.

The checklist

Rather than a developer textbook, here's what to verify, in plain terms:

HTTPS — the entire site served securely, with HTTP versions redirecting.
Indexable URLs — important pages return status 200 and don't carry noindex directives left over from staging. (This happens constantly after redesigns.)
Robots.txt — not blocking anything you want discovered. Review it; it's a two-minute check that catches catastrophic mistakes.
XML sitemaps — current, submitted to Search Console and Bing Webmaster Tools, containing only pages you actually want indexed.
Canonical tags — each page pointing to its preferred version, so near-duplicates don't split credit.
Redirects — old URLs pointing to their closest current equivalent, one hop, via 301. Redesigns without a redirect map are how sites lose years of accumulated visibility overnight.
Status codes — dead pages returning 404 or 410, not soft-200 "not found" pages.
Duplicate content — parameter variants, print versions, and www/non-www duplicates consolidated.
JavaScript rendering — critical content and links visible without requiring JS execution. Google renders JavaScript; many AI crawlers effectively don't. If your content only exists after a script runs, part of your audience of machines can't see it.
Mobile parity — the mobile version containing the same content and links as desktop, since mobile is what gets indexed.
Image accessibility — descriptive filenames and alt text; meaningful images not embedded as unreadable CSS backgrounds.
Structured data — valid markup where appropriate (Chapter 8).
Site speed and Core Web Vitals — fast enough that users don't leave. Google uses these signals, but is explicit that good scores alone don't produce strong rankings. Treat Core Web Vitals as a user-experience floor, not a ranking lever.
Broken links — internal links to dead pages fixed on a recurring schedule.
Orphan pages — every important page reachable through internal links, not just the sitemap.

The priority system

Not all technical issues deserve equal urgency. Triage everything into three tiers:

Critical — prevents discovery, crawling, indexing, or use. A blocked robots.txt, a stray noindex, a site-wide redirect failure. Fix these this week.
Important — weakens interpretation or performance. Duplicate content, weak internal linking, slow templates, rendering issues. Fix these this quarter.
Enhancement — improves presentation or efficiency. Marginal speed gains, sitemap tidiness, markup expansion. Batch these into normal maintenance.

Most technical audits fail businesses by presenting two hundred undifferentiated findings. The value isn't in the finding — it's in knowing which three findings are actually costing you money.

Chapter 7Search and AI Crawler Access

This deserves its own chapter because it's where a genuinely new operational task has appeared — and where three different questions get dangerously conflated.

When someone asks "should we let AI access our site...", they're actually asking three separate questions:

Search visibility — can traditional search crawlers (Googlebot, Bingbot) access and index the site...
AI search visibility — can the crawlers that feed AI answers (like OpenAI's OAI-SearchBot for ChatGPT search) access the site, so it can be cited and recommended...
Model training permissions — is the site's content available for training future AI models (controlled by different user agents, like GPTBot)...

These are related but not the same setting or the same crawler. A business can block model training while remaining fully visible in AI search results — or can accidentally block AI search visibility while thinking it was only opting out of training. OpenAI currently recommends allowing OAI-SearchBot when you want your pages considered for ChatGPT's search summaries and citations. If being recommended when prospects ask ChatGPT for options matters to your business — and for most businesses it increasingly does — verify that this crawler isn't being blocked.

What to verify

Robots.txt controls — review the file for rules affecting Googlebot, Bingbot, OAI-SearchBot, and other AI-related agents. Know what each rule does and why it's there.
Firewall and bot-management rules — this is the silent killer. CDNs and security layers (Cloudflare and similar) often block or challenge AI crawlers by default, regardless of what robots.txt says. Your robots.txt can welcome a crawler your firewall is turning away at the door. Check both layers.
XML sitemaps — submitted and current in both Google Search Console and Bing Webmaster Tools. Bing matters more than its market share suggests, because Bing's index feeds multiple AI systems.
IndexNow — Bing recommends IndexNow as an efficient way to notify participating search engines the moment pages are added, changed, or removed. It's a low-cost addition, particularly for sites that update frequently.
Search Console URL inspection — the ground truth for how Google sees any specific page: indexed or not, and why.
Server-log verification — the definitive answer to "which crawlers are actually reaching us..." Logs show every bot visit. If OAI-SearchBot never appears in your logs, no amount of robots.txt hospitality is working.

The practical recommendation for most businesses: allow the search and AI-search crawlers, decide the training question deliberately based on your own content strategy and comfort, and — whatever you decide — make sure your robots.txt, firewall, and business intentions all agree with each other. Most AI visibility "mysteries" I investigate turn out to be a security rule nobody knew existed.

Chapter 8Structured Data and Entity Clarity

Structured data is one of the most misunderstood tools in visibility work. It is not a ranking hack, and adding schema markup will not, by itself, move you up a single position. What it does is more subtle and more durable: it removes ambiguity. It tells discovery systems, in a machine-readable vocabulary, exactly what a page is about and exactly who is behind it.

Google's position is clear: structured data helps its systems understand page content and entities, and correct markup creates eligibility for supported search features — rich results, knowledge panels, enhanced listings. Eligibility, not entitlement. Valid markup can sit for months without producing a visible feature. Invalid or misleading markup, meanwhile, can get features suspended.

The markup that earns its keep

For most business websites, a short list covers nearly everything worthwhile:

Organization — sitewide; who the business is, its logo, its official profiles.
LocalBusiness — for businesses serving physical areas; hours, location, contact details.
Person / ProfilePage — for the founders, practitioners, and authors whose expertise the site claims.
Product — for ecommerce, with price, availability, and review data.
Service — where semantically appropriate for service offerings.
BreadcrumbList — reinforcing site hierarchy.
Article — for substantial editorial content, with author and dates.
Video — where video content is a meaningful asset.
FAQ — only where genuinely useful and currently supported; Google has sharply narrowed FAQ rich-result eligibility, and stuffing FAQ markup on every page is wasted effort.

Entity consistency: the other half

Markup is only half of entity clarity. The other half is boring, unglamorous consistency — your business appearing identically everywhere it appears:

Business name (exactly, everywhere)
Address and phone
Logo
Founder and team information
Social profiles
Service areas
Business descriptions
Author names and credentials

Discovery systems — search engines and AI models alike — build a picture of your business by cross-referencing every mention of it they can find. When your website says one name, your Google Business Profile says a variation, your LinkedIn says a third, and an old directory says a fourth, you've made your own entity blurry. Blurry entities get cited less, recommended less, and trusted less. This is tedious work with no dashboard to celebrate it, and it is some of the highest-leverage effort in AI-era visibility.

Part Three: Create Content Worth Finding

Chapter 9The Standard for Useful Content

Everything in Parts One and Two exists to deliver an audience to your content. This chapter is about deserving that audience.

The bar has moved. When any AI system can generate a competent summary of publicly known information in four seconds, publishing competent summaries of publicly known information is no longer a strategy. Google's guidance has converged on the same point from a different direction: it rewards content made primarily for people, with original value and a clear intended audience, and it increasingly demotes content that exists primarily to occupy a search result.

What a strong page provides

A genuinely useful page delivers some combination of:

A direct answer — stated early, not buried under eight paragraphs of preamble.
First-hand experience — evidence you've actually done this, not just read about it.
Expert interpretation — what the facts mean, which is where expertise actually lives.
Original research or data — numbers nobody else has.
Useful examples — real ones, with specifics.
A clear process — how to actually do the thing, step by step.
Decision criteria — how to choose between options, honestly.
Risks and limitations — what can go wrong, when this doesn't apply. Nothing signals credibility faster than acknowledging the limits of your own advice.
Comparisons — because comparing is how buyers think.
Next steps — what to do with what they just learned.

No single page needs all ten. A page with none of them shouldn't be published.

The Five-Question Content Test

Before anything goes live on a site I'm responsible for, it has to pass five questions:

Does this page answer a real customer question... Not a question a keyword tool invented — a question an actual prospect has asked or would ask.
Does it add something beyond what already ranks... Search the query. Read the top results. If your draft could be shuffled in among them unnoticed, it isn't ready.
Is the information specific enough to be useful... "It depends on several factors" is not an answer. Which factors... Depends how...
Can the claims be supported... By your experience, your data, or a citable source. Unsupported claims are liabilities everywhere — with readers, with quality systems, and with AI systems deciding what to cite.
Does the page help the reader make a decision... The ultimate test. Information that doesn't move someone toward a decision is trivia.

Five honest yeses, and the page will hold its own on any surface it appears — a search result, an AI citation, a sales email, a prospect's evaluation shortlist. That's the point: content built to this standard doesn't need a different version for each channel.

Chapter 10Build Non-Commodity Content

Here is the position I'll state as plainly as I can:

The internet does not need another summary of information that already exists. It needs clearer evidence, better interpretation, and useful guidance.

Commodity content is any page whose information could be assembled by anyone with a search engine — or generated by any AI model — because it contains nothing that originated with you. "What is X," "10 benefits of Y," the definitional post, the listicle of generic tips. This content was already losing before AI answers; now the AI answer is the commodity summary, delivered instantly, above your listing. You cannot out-publish that. You can only out-know it.

Non-commodity content is built from materials only you possess:

Proprietary data — patterns from your own projects, customers, or operations. Even modest datasets ("across 40 client sites, here's what we found") are more citable than any opinion.
Internal benchmarks — what "good" looks like in your field, from your own measurements.
Before-and-after examples — real transformations with real numbers.
Client questions — the actual questions you're asked, answered the way you actually answer them. Your inbox is a content calendar.
Screenshots — of real work, real dashboards, real results. Proof that this happened.
Original diagrams — how you visualize a process is itself intellectual property, and it's what gets referenced and shared.
Pricing observations — what things actually cost in your market. Businesses avoid this topic; buyers search it constantly; the few who address it honestly own it.
Process documentation — exactly how you do the work. The fear that "they'll just do it themselves" is almost always backwards: showing your process is what convinces serious buyers you're worth hiring.
Expert commentary — your read on industry changes, published while they're happening.
Contrarian but defensible opinions — where the standard advice is wrong and you can show why. Agreeing with everyone is invisible.
Regional or industry-specific details — the local knowledge and niche specifics that generic content physically cannot include.
Tools, calculators, templates, and worksheets — assets that do work for the reader. These earn links and return visits for years.

Notice what all twelve have in common: they cannot be produced by summarizing the existing internet. That's the definition of defensible content in 2026 — and conveniently, it's also exactly the kind of first-hand, evidence-rich material that both Google's quality guidance and AI citation behavior favor.

Chapter 11Optimize Individual Pages

Once a page has substance, page-level optimization is the craft of making that substance easy to find, easy to parse, and easy to act on — for readers first, and for the systems that serve readers second.

Every important page should include

A clear title and main heading — saying what the page is about in the language your audience uses. Clever titles lose to clear ones.
A direct opening answer — address the core question in the first screen. Readers skim; AI systems extract; both reward pages that get to the point.
Logical subheadings — a reader should understand the page's full argument from the headings alone.
Relevant search language — the customer's vocabulary in prominent places, naturally.
Original examples or evidence — the Chapter 10 material, present on this page, not just somewhere on the site.
Contextual internal links — descriptive links to related pages, placed where they genuinely help. Anchor text that says where the link goes ("our website audit process"), not "click here."
Supporting external citations where appropriate — showing your claims trace to sources.
Useful images or media — that add information, with descriptive alt text.
Author or business credibility — who is saying this, and why they're qualified to.
A clear next step — matched to the page's intent. A discovery page might offer a deeper resource; a service page should offer contact.
Accurate metadata — a title tag and meta description that honestly represent the page and earn the click.
Appropriate structured data — per Chapter 8.

The details, briefly

Title tags: roughly 50–60 characters, primary topic first, brand last, unique per page.
Meta descriptions: not a ranking factor, but your search-result ad copy. Under ~160 characters, specific, with a reason to click.
Headings: one H1; H2s and H3s reflecting actual document structure, not styling choices.
URLs: short, lowercase, hyphenated, descriptive, stable. Once published, a URL is an address people and systems rely on — don't change it without a redirect.
Images: compressed, dimensioned, descriptively named, alt-texted.
Anchor text: descriptive and varied; it tells both readers and crawlers what the destination is.
Internal links: every new page should link out to related pages and receive links from them. Publishing a page nobody links to is publishing an orphan.
Calls to action: one primary action per page, matched to visitor readiness (Chapter 16).
Content updates: refresh important pages on a schedule; update the revision date only when the substance actually changed. Fake freshness is transparent to everyone.

None of this is exotic. That's the point — page optimization in 2026 is mostly the disciplined application of things that have been true for a decade, applied to content that finally deserves it.

Part Four: Earn Trust and Recognition

Chapter 12Demonstrate Experience and Credibility

There was a time when a website could be effectively anonymous and still rank. That time is over — for good reasons. Search quality systems increasingly evaluate who is behind content, and AI systems deciding what to cite and recommend lean heavily on signals of identity and legitimacy. More importantly, so do your customers. The question "who are these people, and can I trust them..." is asked on every visit, whether or not it's asked out loud.

The work of this chapter is not manufacturing trust signals. It's making legitimate expertise visible and verifiable — taking credibility that exists in the real world and putting it where visitors and evaluation systems can see it.

What that looks like in practice:

A detailed About page. Not three sentences of mission-statement fog. The real story: who founded the business, why, what it actually does, who it serves, and what it believes about its work. About pages are consistently among the most-visited pages during evaluation. Treat yours accordingly.

Author profiles. Every substantial piece of content attributed to a real person, with a bio, photo, credentials, and a profile page linking to their work elsewhere. Anonymous content is increasingly treated as low-trust content — by readers and systems alike.

Team credentials. Licenses, certifications, degrees, years of practice — stated specifically, where relevant. "Our experienced team" says nothing. "Licensed in Iowa since 2011" says something.

Editorial or review processes. If your content is reviewed for accuracy — by a practitioner, an expert, an editor — say so, and say by whom. This matters enormously in fields where accuracy has consequences.

Case studies. The single most persuasive content type most businesses under-produce. Real client, real problem, real work, real outcome, real numbers where possible.

Testimonials and reviews. With names, businesses, and specifics. "Great to work with!" is filler; "They found a tracking error that had been overstating our leads by 40%" is evidence.

Client and project examples. Even a simple portfolio of "here's who we've worked with and what we did" beats abstraction.

Awards and affiliations. Industry memberships, certifications, recognitions — displayed where they're relevant, not wallpapered everywhere.

Original photography. Your actual team, office, work, and projects. Stock photography is a small dishonesty that visitors detect instantly, and a site full of it signals that nothing else on the page should be taken at face value either.

Transparent contact information. A real address, a real phone number, real people. Businesses that are easy to find are easier to trust.

Policies and guarantees. What you stand behind and how — stated plainly.

Publication and revision dates. On everything substantial. Undated content is unaccountable content.

A caution: don't turn this into a mechanical "E-E-A-T checklist" where you bolt on trust-badges and boilerplate bios because a blog post told you to. Quality systems aren't checking boxes, and neither are customers. The order of operations is: be credible, then make that credibility impossible to miss.

Chapter 13Authority Beyond the Website

Your website is your case. Everything said about you elsewhere is your corroboration — and no discovery system, human or machine, takes an uncorroborated case at face value.

External recognition has always mattered for search rankings, primarily through links. In the AI era its role has widened: language models and answer engines form their picture of "who's credible in this space" from the whole visible web — mentions, citations, reviews, interviews, community presence — not just hyperlinks. That changes what "link building" should mean.

The reframe: external visibility should create real recognition and referral opportunities, not merely manufacture links. A link is a byproduct of being genuinely worth mentioning. Chase the recognition and the links follow; chase the links and you accumulate exactly the manipulative patterns Google's spam policies prohibit — and its current AI-search guidance specifically warns against seeking inauthentic mentions. The same behavior that used to be low-risk gray-hat is now flagged from two directions at once.

Where legitimate external authority comes from:

Relevant editorial links — earned because you published something worth referencing: original data, useful tools, a definitive resource. One link from a respected industry publication outweighs a hundred directory listings.
Industry associations — membership, participation, and the profile pages that come with them.
Digital PR — giving journalists and publishers something genuinely newsworthy: research, data, expert availability on developing stories.
Podcasts and interviews — high-leverage for founders and practitioners. One good interview produces a citation, an audience, an entity signal, and reusable content.
Expert contributions — quotes, guest pieces, and panel participation in the places your customers already read.
Local citations — for local businesses, accurate and consistent listings in the directories and maps platforms that actually matter (see entity consistency, Chapter 8).
Partnerships — real business relationships that naturally produce cross-references.
Community involvement — sponsorships, events, and civic participation that generate authentic local mentions.
Reviews — systematically requested, on the platforms your customers check, responded to professionally.
Social profiles — complete, consistent, active enough to prove the business is alive.
YouTube and video — a search engine in its own right, and a major input to AI systems' understanding of who explains things well.
Forums and community participation — genuinely helpful presence in Reddit, industry forums, and professional communities. These surfaces are heavily weighted in both search results and AI training data; helpful participation there compounds. Self-promotional spam there backfires immediately.

What to avoid should now be obvious, but for the record: purchased link schemes, private blog networks, mass directory submissions, reciprocal-link arrangements, syndicating the same article across dozens of low-grade platforms, and paying for fake mentions or reviews. Every one of these is a documented spam-policy violation, and every one produces the kind of pattern that's trivially detectable at scale. The uncomfortable truth about authority is that there's no legitimate shortcut — which is precisely why it's a moat once you've built it.

Part Five: Visibility in AI-Generated Answers

Chapter 14How AI Visibility Actually Works

There is a lot of confident nonsense being sold about AI search optimization right now. So let's start with an honest account of what's actually happening — stated broadly, because these systems differ and change, and anyone who tells you they know exactly how each one weighs each signal is guessing.

At a high level, every AI answer system does some version of five things:

It discovers or obtains access to information — by crawling the web, licensing content, drawing on a search index (its own or a partner's), or some combination.
It interprets pages, entities, and relationships — building an internal picture of what content is about, who published it, and how sources relate to one another.
It retrieves sources relevant to a user's question — for search-connected systems, this often looks a lot like a search query behind the scenes, which is why classic search visibility remains upstream of AI visibility.
It synthesizes an answer — composing a response from what it retrieved and what it already "knows."
It may cite, link, mention, or recommend sources — depending on the platform, the question, and how clearly attributable the underlying material is.

Understand step 3 and you understand why this guide spends most of its pages on fundamentals: for the systems driving most AI-answer traffic, retrieval runs through search infrastructure. If you're not crawlable, indexable, and findable for a topic, you're not in the pool of sources an AI answer draws from. Google says as much about its own generative features — they depend on the same crawlability, index eligibility, and quality systems as classic search. There is no side door.

What AI visibility looks like when you have it

It's worth being precise, because "showing up in AI" isn't one thing. Your business might appear as:

A citation — your page listed as a source behind an answer
A linked source — a clickable reference driving actual referral traffic
A brand mention — named in the answer text, with or without a link
A recommendation — "consider SimQua for..." in response to a "who should I hire" prompt
A product or service listing — surfaced in shopping- or provider-style results
A summarized answer — your content shaping the answer without explicit attribution
A navigational result — the system pointing users directly to your site

These have different business value and different drivers. A citation rewards being the clearest source on a specific question. A recommendation rewards entity strength and widespread third-party recognition. Measuring and pursuing "AI visibility" starts with deciding which of these forms actually matters for your business.

Chapter 15Practical AI Visibility Recommendations

Everything in this chapter follows from one premise, which happens to be Google's stated position and matches what practitioners observe across other platforms: AI visibility is built on search fundamentals, executed to a higher standard. Here is the honest to-do list.

Do

Make pages crawlable and indexable — including verified access for AI-search crawlers (Chapter 7). This is the gate everything else stands behind.
Provide direct, well-structured answers — a clear question addressed near the top of a clearly headed section is the easiest material for any system to retrieve and attribute.
Establish clear business and author entities — consistent identity everywhere, supported by markup (Chapter 8). Systems recommend entities they can resolve confidently.
Publish first-hand evidence — the non-commodity material of Chapter 10. Original data and experience are what make your page the source rather than a paraphrase of one.
Cover topics comprehensively — depth across a topic, not one page per keyword. Systems asked complex questions favor sources that handle complexity.
Use descriptive headings — your heading structure is the machine-readable outline of your expertise.
Keep factual information current — prices, dates, specs, availability. Stale facts get you excluded from answers that need current ones.
Provide citations and source transparency — pages that show their sources are easier to trust, for readers and retrieval systems alike.
Earn legitimate mentions elsewhere — Chapter 13. AI recommendations are heavily shaped by the broader web's opinion of you.
Maintain consistent business information — everywhere, always (Chapter 8 again — it keeps coming up because it keeps mattering).
Track AI referral traffic — set up referral reporting for ChatGPT, Perplexity, Copilot, Gemini, and AI-overview traffic so you can see what this channel actually produces (Chapter 17).
Test representative prompts regularly — monthly, ask the major AI platforms the questions your customers would ask. Record who gets recommended, who gets cited, and how you're described. This is the closest thing this channel currently has to rank tracking.

What not to do

Do not create hundreds of shallow FAQ pages. Fragmenting topics into thin Q&A pages is the scaled low-value pattern quality systems are built to demote.
Do not rewrite competitor content. If it originated elsewhere, you've added a worse copy to the pool — nothing retrievable, nothing citable.
Do not publish unsupported claims. AI systems increasingly cross-check; being confidently wrong is a durable way to be excluded.
Do not assume schema guarantees citations. Markup clarifies; it does not entitle (Chapter 8).
Do not buy fake brand mentions. Explicitly warned against in Google's AI-search guidance, and a spam-policy violation besides.
Do not treat llms.txt as a universal requirement. Google states plainly it isn't needed for its generative features. Adding a file is harmless; paying someone who claims it's the key to AI rankings is not.
Do not create separate "AI content" that's worse for humans. There is one audience: people, plus the systems trying to serve them. Content degraded for machine consumption fails with both.

If a vendor's AI-visibility pitch consists mostly of items from the second list, keep your money.

Part Six: Turn Visibility Into Business

Chapter 16Conversion-Focused Website Strategy

This is the chapter that separates a business asset from a publishing hobby. Every preceding page of this guide is worthless if the visitors it produces arrive, read, and leave.

Conversion is not a pop-up strategy. It's the discipline of matching what you ask of visitors to where they are in their decision — and removing every unnecessary obstacle between readiness and action.

The progression every customer moves through:

Discover → Understand → Trust → Evaluate → Act

Most websites only build for the two ends: content for discovery, a contact form for action. The three stages in between — where decisions are actually made — are left to chance. Building for the full progression means:

Match calls to action to visitor readiness. A first-time reader of an informational guide is not ready for "Start an audit" — but they might take a checklist or a deeper resource. A visitor on their third session reading your pricing page is ready for exactly that call. Every page should offer a next step proportionate to its intent (the framework from Chapter 4 maps directly here).

Reduce unnecessary form fields. Every field is friction, and most forms collect information nobody uses. Ask for what the next conversation actually requires — usually a name, a contact method, and a sentence about the need. You can qualify on the call.

Establish trust near conversion points. The moment before someone submits a form or dials a number is the moment doubt peaks. Testimonials, credentials, guarantees, and recognizable client names belong beside the form, not just on the About page.

Explain what happens after submission. "We'll respond within one business day, and here's what the first conversation covers" measurably outperforms a bare Submit button. Uncertainty kills conversions; process descriptions kill uncertainty.

Use low-commitment and high-commitment CTAs together. "Start an audit" for the ready; "Get the checklist" or "Read the field guide" for the not-yet. A site with only high-commitment CTAs silently discards the majority of its qualified future customers.

Connect resource content to service pages. Every guide, article, and tool should link contextually to the service it relates to — not as a banner ad, but as a genuine "if you'd rather have this done for you" pathway. Informational content that dead-ends is the most common conversion leak I find.

Address objections. Whatever stops people — price fears, timeline worries, "will this work for a business like mine" — answer it on the page, before they have to ask. Your sales conversations already told you what the objections are.

Show price expectations where appropriate. Full pricing isn't always feasible, but ranges, starting points, or "how pricing works" content qualifies visitors, builds trust, and captures the enormous volume of cost-related search demand your competitors are avoiding.

Improve mobile conversion specifically. Test your forms, click-to-call, and booking flows on an actual phone. A form that's mildly annoying on desktop is abandoned on mobile.

Track every conversion action — form, phone, booking, and email — which brings us to measurement.

Chapter 17Measurement That Matters

Most website reporting fails in one of two ways: it measures nothing (a traffic chart and a shrug) or it measures everything (forty metrics, no decisions). The fix is a hierarchy — four levels, each answering a different question, read together.

Level 1: Visibility — Can we be found...

Impressions — how often you appear in search, regardless of clicks
Ranking coverage — the breadth of queries and topics where you appear
Search features — presence in maps, rich results, and other enhanced placements
AI mentions and citations — from your monthly prompt testing (Chapter 15) and any monitoring tools
Brand search demand — people searching your name is the truest measure that broader marketing is working

Level 2: Traffic — Are the right people arriving...

Organic users — trend, not day-to-day noise
Landing-page sessions — which pages actually earn entry traffic
Qualified visits — traffic to pages that indicate commercial intent, from geographies you serve
Referral traffic — from the recognition work of Chapter 13
AI platform referrals — sessions from ChatGPT, Perplexity, Gemini, Copilot, and AI overviews, tracked as their own channel so you can see this shift as it happens

Level 3: Engagement — Is the site doing its job...

Content consumption — are people actually reading, or bouncing in seconds
Next-page movement — do resource visitors continue toward service content
Service-page visits — the clearest mid-funnel signal a business site has
Returning users — evaluation for anything considered happens across multiple sessions
Meaningful interactions — downloads, tool usage, video plays, pricing views

Level 4: Business outcomes — Is any of this making money...

Form submissions
Calls — tracked, not guessed; for many service businesses, calls are the majority of leads
Bookings
Qualified leads — as defined in writing back in Chapter 3
Sales
Revenue
Cost per acquisition

In GA4 terms, Level 4 is where key events live — GA4's mechanism for designating the actions that particularly matter to the business, so every channel and page can be evaluated by its contribution to those actions rather than by raw traffic. If your GA4 property has no key events configured, or its key events are things like "scrolled 90%," your reporting is decorative. Configuring honest key events is the single highest-leverage hour in most analytics setups.

Read the hierarchy top-down when diagnosing ("visibility is up, outcomes are flat — where does the chain break...") and bottom-up when reporting ("here's what the website produced, and here's the visibility pipeline behind it").

And the principle this entire guide keeps returning to, because every reporting meeting needs it said out loud:

Traffic is useful only when it contributes to awareness, evaluation, leads, sales, or another defined business outcome. A traffic increase is not a result. It's a leading indicator that a result may be coming — or evidence that you've gotten better at attracting the wrong people. The metrics exist to tell you which.

Part Seven: Diagnose and Improve

Chapter 18Website Visibility Troubleshooting

Visibility problems rarely announce their causes. What you observe is a symptom — traffic fell, leads dried up, pages vanished from the index — and the discipline is working from symptom to cause methodically instead of grabbing the first plausible explanation. This chapter is organized the way problems actually arrive.

A framing note first: Google's own recommendation for meaningful traffic losses is to assess the situation broadly rather than attempting isolated fixes based on assumed individual ranking factors — and it cautions that recovery after substantial improvements can take time. In practice: diagnose the whole system, fix the real causes, and judge results in quarters, not days.

Symptom: Rankings dropped

Work through these in order — the cheap checks first:

Search demand changes — did the queries decline, not your position... Seasonality and shifting demand masquerade as ranking losses constantly. Check impressions before panicking about positions.
Algorithm updates — did the drop coincide with a confirmed update... If so, the question becomes "what does the update reward that we lack," not "what page do we tweak."
Technical changes — what shipped recently... Redesigns, migrations, plugin updates, and CMS changes are behind a remarkable share of "mysterious" drops. Check for new noindex tags, broken redirects, and robots.txt edits first.
Competitor improvements — sometimes you didn't fall; someone else climbed. Look at who took your positions and what they're doing.
Intent shifts — Google periodically reinterprets what a query means. If a query that used to return service pages now returns guides, the intent moved and your content type no longer matches.
Indexing — are the affected pages still indexed at all... URL inspection answers this in seconds.
Content quality — honestly re-run the Five-Question Test (Chapter 9) against what now outranks you.
Brand demand — declining branded search often precedes declining non-branded rankings; entity strength and rankings travel together.
SERP layout changes — sometimes your rank held but an AI overview, map pack, or ad block moved above you and absorbed the clicks. Same position, less traffic, different problem.

Symptom: Traffic increased but leads did not

The most common — and most misdiagnosed — situation in this entire guide. Review:

Search intent — is the new traffic informational visitors with no buying context... Growth from a viral-adjacent blog post produces readers, not leads. Check which pages grew.
Geographic relevance — is the traffic from places you serve... National traffic to a regional business inflates every chart and converts at zero.
Conversion paths — can a motivated visitor get from the growing pages to a service page in one obvious click... (Chapter 16's most common leak.)
Offer clarity — does the site make plain what you sell, to whom, and what to do next...
CTA placement — are next steps present and proportionate on the pages people actually land on...
Mobile usability — test the conversion flow on a phone; this is where "traffic without leads" often lives.
Tracking accuracy — before concluding leads didn't happen, confirm you'd know if they had. Broken form tracking and untracked phone calls make successful sites look like failures every day.

Symptom: Pages are not being indexed

Robots directives — noindex tags and robots.txt blocks, including ones a plugin or template added silently.
Canonicals — is the page declaring some other URL the preferred version...
Internal links — does anything on the site actually link to this page... Orphans get deprioritized.
Sitemap inclusion — present, and in a sitemap that's submitted and error-free...
Duplicate content — if it's substantially similar to an existing page, Google may simply choose the other one.
Page quality — indexing is not guaranteed; thin pages on sites with many thin pages get left out. Sometimes "not indexed" is feedback.
Rendering — does the content exist without JavaScript... Test what the crawler sees, not what your browser sees.
Server responses — intermittent errors, timeouts, and firewall challenges can quietly block crawling while everything looks fine to humans.

Symptom: The brand is not appearing in AI answers

Crawler access — are AI-search crawlers actually reaching the site... Robots.txt and firewall, verified in server logs (Chapter 7).
Search index visibility — are you findable in classic search for these topics... If not, start there; retrieval runs through search.
Entity clarity — can systems confidently resolve who you are... (Chapter 8.)
Topic coverage — do you have substantial content on the specific question being asked, or only adjacent material...
Original evidence — is there anything on your pages that makes them the source worth citing rather than one of forty paraphrases...
External recognition — does the wider web corroborate that you're a credible option... Recommendations especially depend on this.
Prompt relevance — finally, be honest about the prompt. If you're testing "best national providers" and you're a regional firm, absence isn't a visibility failure; it's an accurate answer.

Part Eight: Practical Playbooks

Chapter 19Action Plans by Website Situation

Everything so far is the methodology. This chapter compresses it into starting sequences for the situations businesses actually arrive in. Each playbook gives the first five checks, the priority actions, what to measure, the mistakes to avoid, and the realistic order of work.

Playbook: New website

First five checks: positioning clarity (Ch. 3), site architecture plan (Ch. 5), indexability of launched pages, GA4 + Search Console + key events configured from day one, entity basics (Organization markup, consistent profiles).
Highest-priority actions: nail the core service pages before any blog content; build the About/trust layer; establish Google Business Profile and essential citations if local.
Measure: indexation coverage, impression growth, first key events. Expect months, not weeks.
Common mistakes: publishing thin content for volume; skipping measurement setup "until later"; expecting rankings before authority exists.
Sequence: foundation → core pages → trust layer → supporting content → external recognition.

Playbook: Website redesign

First five checks: full URL inventory of the current site, redirect map drafted, current rankings/traffic benchmarked, staging site blocked from indexing, analytics migration plan.
Highest-priority actions: preserve or improve every page with existing search value; 301 every changed URL; verify no noindex/robots blocks ship to production; re-verify Search Console and sitemaps at launch.
Measure: daily indexation and traffic monitoring for the first month post-launch against the benchmark.
Common mistakes: treating the redesign as purely visual; deleting "old" pages that held rankings; launching on a Friday.
Sequence: benchmark → map → migrate → verify → monitor.

Playbook: Established site with declining traffic

First five checks: demand vs. position (did queries decline or did you...), algorithm-update timeline overlay, recent technical changes, index coverage trends, SERP layout changes on your money queries.
Highest-priority actions: run the full Chapter 18 rankings diagnosis; consolidate or improve weak content rather than adding more; strengthen the pages that still earn traffic.
Measure: the four-level hierarchy (Ch. 17), quarterly.
Common mistakes: publishing more as the reflexive response to decline; making isolated tweaks instead of assessing broadly.
Sequence: diagnose → fix causes → improve survivors → rebuild authority → be patient.

Playbook: Local service business

First five checks: Google Business Profile completeness and category accuracy, citation consistency (Ch. 8), service pages for each core offering, location legitimacy of any location pages, review volume and recency.
Highest-priority actions: one strong page per service; systematic review generation; local entity consistency; original photos of real work.
Measure: map-pack visibility, calls (tracked), direction requests, form leads by service.
Common mistakes: mass-generating city pages; neglecting the review pipeline; hiding the service area.
Sequence: GBP + citations → service pages → reviews → local content and community recognition.

Playbook: Professional service business

First five checks: practitioner profiles and credentials visible, case studies or matter examples present, evaluation-stage content (how to choose, pricing approach, process), About page depth, individual expertise visible off-site.
Highest-priority actions: make the experts visible (Ch. 12); build decision-support content (Ch. 4's evaluation layer); publish first-hand perspective content under real bylines.
Measure: consultation requests, evaluation-content engagement, branded search growth.
Common mistakes: anonymous firm-voice content; hiding pricing philosophy; treating the site as a brochure.
Sequence: trust layer → core service pages → evaluation content → practitioner authority off-site.

Playbook: Ecommerce business

First five checks: product page indexability at scale, Product structured data validity, faceted-navigation crawl control, category page content quality, key event tracking through checkout.
Highest-priority actions: category pages as ranking assets; unique product information (not manufacturer boilerplate); buying-guide content connected to categories.
Measure: revenue by landing page, product impressions/features, category ranking coverage.
Common mistakes: duplicate manufacturer descriptions; letting filters spawn infinite crawlable URLs; ignoring the informational layer entirely.
Sequence: technical scale issues → category strength → product content → guides → recognition.

Playbook: Multi-location organization

First five checks: one location page per real location with genuinely local substance, per-location GBP management, consistent NAP data at scale, local content duplication risk, per-location tracking.
Highest-priority actions: location page template with real local content (staff, photos, services, service-area specifics); centralized citation management; local review programs per location.
Measure: per-location visibility, calls, and leads — aggregate numbers hide failing locations.
Common mistakes: identical boilerplate across locations; managing profiles nobody owns; measuring only the aggregate.
Sequence: data consistency → location pages → per-location reviews → local recognition.

Playbook: Business entering AI search optimization

First five checks: AI-search crawler access verified in logs (Ch. 7), classic search visibility for target topics, entity clarity audit, prompt testing baseline across major platforms, AI referral tracking configured.
Highest-priority actions: fix crawler access; strengthen entity signals; upgrade the top 10 pages with direct answers and first-hand evidence; earn corroborating mentions.
Measure: monthly prompt-test results, AI citations observed, AI referral sessions and their conversion rate.
Common mistakes: buying "AI SEO" services built on llms.txt files and fake mentions; creating separate AI-targeted content; expecting attribution where platforms simply don't send much traffic yet.
Sequence: access → fundamentals → evidence-rich content → recognition → measurement.

Playbook: Site with traffic but few leads

First five checks: which pages hold the traffic and what intent they serve, geographic fit, conversion-path audit from top pages, tracking verification (would you know about a lead...), mobile conversion flow test.
Highest-priority actions: the full Chapter 16 program — CTAs matched to readiness, resource-to-service pathways, trust at conversion points, form friction reduction.
Measure: Level 3 and 4 metrics; conversion rate by landing page, not sitewide.
Common mistakes: buying more traffic to fix a conversion problem; concluding "SEO doesn't work" when the leak is downstream.
Sequence: verify tracking → fix paths → align CTAs → add trust → then grow traffic.

Playbook: Site recovering from low-quality content production

First five checks: full content inventory with traffic/engagement per URL, scale of thin or generated content, index bloat assessment, which pages retain genuine value, whether the pattern matches Google's scaled-content policies.
Highest-priority actions: prune ruthlessly — improve pages worth improving, consolidate overlapping ones, remove (with appropriate status codes) pages with no value or prospects; then rebuild publishing standards around Chapters 9–10.
Measure: index coverage quality, engagement on surviving content, gradual impression recovery.
Common mistakes: keeping everything "because it's indexed"; expecting fast recovery — Google is explicit that regaining trust after substantial improvement takes time; resuming the old publishing habits that caused the problem.
Sequence: inventory → prune → improve survivors → new standards → patience.

Chapter 20The 90-Day Visibility Plan

For a business starting — or restarting — visibility work in earnest, here is the sequence I'd run. Ninety days won't finish the job; it will establish the foundation, fix what's broken, and produce the first honest measurement of what's working.

Days 1–30: Establish the baseline

You cannot improve what you haven't measured, and you shouldn't build on ground you haven't inspected.

Define audiences and outcomes — the Chapter 3 worksheet, completed in writing, with a written definition of a qualified lead.
Audit tracking — GA4 configured, key events defined for real business actions, call tracking in place, Search Console and Bing Webmaster Tools verified. Test every conversion path yourself.
Review technical accessibility — the Chapter 6 checklist, triaged Critical / Important / Enhancement; crawler access verified per Chapter 7, including AI-search crawlers, in the server logs.
Map services and topics — current architecture vs. the Chapter 5 model; where are the gaps, overlaps, and orphans...
Identify the important pages — the 10–20 pages that matter most to revenue. These get the next thirty days.
Establish benchmark reporting — the four-level hierarchy (Ch. 17), recorded now, so every future claim of progress has a "compared to what."

Days 31–60: Improve priority pages

Resist the urge to create new things. The fastest gains almost always come from making the pages you already have deserve their positions.

Rewrite core service pages — against the Chapter 11 framework, with the Chapter 10 evidence: real examples, real proof, direct answers, honest specifics.
Strengthen internal linking — every supporting page linking to the service it supports; every service page linked from everything relevant; no important orphans.
Improve evidence and credibility — About page rebuilt, author profiles live, case studies drafted, real photography replacing stock.
Resolve technical barriers — every Critical item closed; Important items scheduled.
Build missing decision content — the evaluation-stage gaps from your intent mapping: the cost guide, the "how to choose," the comparison your prospects are searching for.
Improve conversion paths — Chapter 16 applied to the priority pages: matched CTAs, trust at the point of action, mobile flows tested.

Days 61–90: Expand and validate

Publish supporting resources — the first substantial pieces built to the Five-Question standard, each connected to a service page.
Pursue legitimate external exposure — the first recognition efforts from Chapter 13: an association profile, a podcast pitch, a data-driven piece worth referencing, systematic review requests.
Test AI prompts — the baseline prompt panel across ChatGPT, Gemini, Perplexity, and Google's AI features: who's recommended, who's cited, where you appear. Record it; this is your AI-visibility benchmark.
Review early performance — thirty-plus days of clean data against the Day-30 benchmark. Read the hierarchy top-down: where is the chain moving, and where is it stuck...
Improve underperforming pages — the priority pages that haven't responded get a second, evidence-based pass.
Establish the next quarterly plan — because Day 90 is not a finish line. It's the first checkpoint of a permanent operating rhythm: measure, diagnose, improve, expand, repeat.

Set expectations accordingly: by Day 90 you should see improved indexation, growing impressions, stronger engagement on rebuilt pages, and — for sites that had conversion leaks — often the first lead improvements. Ranking and AI-visibility gains compound over the following quarters. Anyone promising the full result in ninety days is selling the calendar, not the work.

The Core Editorial Principle

This guide will keep growing. Chapters will be revised, playbooks added, recommendations promoted or retired as the evidence changes. But every addition has to answer at least one of five questions:

How can a website become easier to discover...
How can it become a more trusted source...
How can it attract a more qualified audience...
How can it turn that attention into action...
How can the business measure whether the work is succeeding...

Anything that doesn't serve one of those outcomes doesn't belong on this page — no matter how novel the tactic or loud the trend. That's the standard I hold client work to, and it's the standard this guide is held to.

Recent Field Guide Updates

July 2026

Updated AI crawler guidance (→ Chapter 7)
Clarified Google's position on generative search optimization (→ Chapters 14–15)
Added AI referral measurement recommendations (→ Chapter 17)
Expanded guidance on non-commodity content (→ Chapter 10)

Review cadence: full review quarterly · immediate updates for significant search or AI platform changes · structural review annually · broken links and outdated references checked monthly.

Confidence labels used in this guide: Established practice · Supported by platform documentation · Observed across multiple sites · Promising but still being tested · Unproven or overstated.

Need help applying this to your website... Request a visibility audit — or Use the contact form to talk through where your site stands today.

Visibility now depends on trust, usefulness, and measurability across the full discovery journey.

Need help turning the field guide into a focused next step?