How I Work

Start With the Problem, Not the SEO Checklist

SEO tools can identify thousands of conditions across a website.

That does not mean every condition is a problem.

A page with declining clicks may have a ranking problem. It may also have stable rankings but lower search demand. A poorly performing article may need better content, or it may be competing with another URL targeting the same intent. An orphan page may represent an internal linking problem, or it may be a page that should not exist at all.

The first objective is therefore not to find as many SEO issues as possible.

It is to determine what is actually limiting search performance and what evidence supports that conclusion.

That principle guides my entire process.

My SEO Process

Every project moves through the same six stages. The depth of each stage depends on the website and the problem being investigated.

1. Discover — understand the business, website, audience and search environment.
2. Diagnose — investigate what is actually limiting search performance.
3. Prioritise — decide what deserves action first.
4. Implement — turn priorities into documented website changes.
5. Measure — evaluate the work against the problem it was meant to solve.
6. Refine — use the new evidence to decide what happens next.

01. Discover

Every project begins with understanding the environment in which the website operates.

Before recommending changes, I want to understand the business, website, audience, search landscape and existing performance.

Discovery can include:

• Business objectives.
• Products and services.
• Target audiences.
• Priority markets.
• Website structure.
• Existing content.
• Organic search visibility.
• Competitors and search-result competitors.
• Available analytics and search data.
• Previous SEO work.
• Technical constraints.
• Content production processes.

The objective is to establish context.

Without that context, an SEO recommendation can be technically correct and still be strategically wrong.

02. Diagnose

Once the context is understood, I investigate how the website currently performs and where problems or opportunities exist.

Diagnosis can span several connected areas.

Technical SEO

I examine factors affecting how search engines discover, crawl, render, index and interpret the website.

This can include crawlability, indexation, canonicalisation, redirects, status codes, XML sitemaps, robots directives, structured data, rendering and site performance.

Information Architecture

I examine how pages are organised and how relationships between topics, services, locations and supporting content are communicated.

This can include URL hierarchy, navigation, crawl depth, internal linking, orphan pages and hub relationships.

Search Intent

I analyse what users appear to be trying to find, understand, compare or accomplish and whether the appropriate page satisfies that intent.

Content

I evaluate whether existing pages provide useful, relevant and sufficiently complete information for their intended search purpose.

Semantic and Entity SEO

I examine important entities, concepts and relationships and whether the website provides enough context for users and search systems to understand them.

Search Performance

I analyse available performance data to understand what is actually happening at page and query level.

This can include clicks, impressions, click-through rate, average position, landing-page performance and changes over time.

AEO and GEO

Where relevant, I extend the analysis into answer-driven and generative search environments.

This can involve question coverage, answer structure, entity representation, brand mentions, citations, cited sources and observable AI search behaviour.

The result of diagnosis should be an explanation of the problem, not simply an export from an SEO tool.

03. Prioritise

Not every identified issue deserves immediate action.

SEO work competes for development time, content resources and business attention. Prioritisation determines what should be addressed first.

I evaluate recommendations using factors such as:

The objective is to turn a large collection of observations into a manageable SEO roadmap.

A prioritised roadmap is more useful than a 200-item audit with no indication of what matters.

04. Implement

Once priorities are established, recommendations need to become actual website changes.

Depending on the project, implementation can involve:

Implementation is documented so recommendations can be distinguished from changes that were actually completed.

This distinction becomes important when evaluating results.

05. Measure

SEO work should be evaluated against the problem it was intended to solve.

The metrics therefore depend on the project.

Organic Search Performance

Clicks, impressions, rankings, click-through rate and landing-page visibility.

Technical SEO

Indexation, crawlability, canonicalisation, internal linking and resolution of validated technical problems.

Content

Query visibility, search performance, engagement and coverage of relevant search intent.

Information Architecture

Crawl depth, orphan pages, internal link distribution, page relationships and search visibility across important sections.

AEO

Observable answer visibility and coverage across relevant question-based search experiences.

GEO

Brand mentions, citations, cited URLs, entity accuracy, topic association and source selection across a defined set of generative search observations.

Measurement should also account for external variables.

Search demand changes. Competitors change. Search engines update their systems. Generative platforms change their retrieval and response behaviour.

A performance movement should therefore be investigated before being attributed to a specific optimisation.

06. Refine

SEO does not end when recommendations are implemented.

The resulting data creates new evidence.

I use that evidence to determine:

• What improved.
• What did not change.
• What performed differently from expectations.
• Whether the original diagnosis remains valid.
• Whether new search opportunities have emerged.
• Whether priorities need to change.
• What should be tested or investigated next.

This creates an iterative optimisation cycle.

Discover. Diagnose. Prioritise. Implement. Measure. Refine.

Then repeat where necessary.

How SEO, AEO and GEO Fit Together

I do not treat SEO, Answer Engine Optimisation and Generative Engine Optimisation as three independent marketing channels.

They operate across overlapping information systems.

SEO Builds the Foundation

Technical accessibility, search intent, content quality, website architecture and authority remain fundamental to search visibility.

AEO Extends Content Into Answer-Driven Search

AEO examines whether information clearly addresses relevant questions and can be interpreted within search experiences that provide direct answers.

GEO Extends Visibility Into Generative Search

GEO examines how entities, information and sources are represented within generative search environments and what can be improved within the surrounding information ecosystem.

The relationship can be represented as:

1. Technical Foundation
2. Information Architecture
3. Search Intent and Content
4. Semantic and Entity Understanding
5. Evidence and Authority
6. SEO, AEO and GEO Visibility
7. Measurement
8. Refinement

The exact optimisation mix depends on the website and its audience.

Data Before Assumptions

Search optimisation involves uncertainty.

A ranking decline can have several possible causes. An indexing problem can originate from multiple technical conditions. An AI-generated answer can vary between systems, prompts and even repeated tests.

For that reason, I separate three things:

Maintaining that distinction reduces the risk of turning assumptions into SEO recommendations.

Tools Support the Process. They Do Not Make the Decisions.

SEO platforms are useful for collecting data, identifying patterns and accelerating analysis.

But an SEO tool reporting an issue does not automatically mean the issue deserves action.

I use tools according to the problem being investigated.

These can include:

The tool provides information.

The SEO specialist still needs to determine what that information means.

Documentation Is Part of the Work

SEO recommendations become difficult to manage when decisions exist only in chat messages, temporary reports or someone’s memory.

I prefer structured documentation.

Depending on the project, this can include:

• URL inventories.
• SEO audit sheets.
• Keyword maps.
• Search intent maps.
• Content inventories.
• Information architecture maps.
• Internal linking recommendations.
• Technical issue logs.
• Content briefs.
• Implementation trackers.
• Search performance reports.
• AEO observations.
• GEO visibility records.

This creates a record of what was identified, what was recommended, what was implemented and what happened afterwards.

What My Deliverables Look Like

The final deliverable depends on the problem.

It may be:

The format changes.

The principle does not.

Every deliverable should help answer:

What is happening? Why does it matter? What should we do next?

See the Process Applied

The methodology is easier to evaluate when you can inspect the work.

My Projects section contains individual audits, analyses, maps and optimisation work.

My Case Studies show how those pieces connect when solving larger search problems.