Azhan Shafique

The fix is not automatically more links; it is making existing authority and company information easy for search systems to discover, understand and connect. In this view, SEO and GEO converge around entity understanding: who the company is, what it does and which information represents it.

A surgical instrument manufacturer had ISO certifications, regulatory references, industry memberships, established business information and a long industry history, but ChatGPT and Google AI still misdescribed the company.

• Shafique worked on the information layer around company identity, products, location, certifications, expertise, business history and entity details, then had important pages re-indexed through Google Search Console.
• The post says AI systems then began identifying the company correctly and surfacing more accurate information.

Read original

A B2B site may attract visitors but lose them before contact because it does not answer the questions that decide a purchase: what makes the product different, whether the company meets technical requirements, whether it has the experience and capabilities, and why it should be trusted.

• The same questions are now put to ChatGPT and other AI platforms, so a brand absent from those answers can be missed before anyone reaches the website.
• The recommended work combines technical SEO, buyer-question content, off-site trust signals and measurement of AI visibility, including which sources are cited and where competitors appear.

The post describes B2B companies investing in SEO to rank while overlooking what happens after discovery: a buyer searches the site for products and cannot find clear answers about differentiation, technical requirements, experience or trust. The goal is not just to be found but to become a credible option when buyers are ready to decide.

Read original

A page can have plenty of words and still repeat one vague claim without explaining what it means to a buyer.

• The method described is to rebuild the page from search intent upward: primary keyword, meta title, meta description, H1, H2s, H3s, semantic entities, internal links, external references and call to action.
• Semantic SEO, in this account, means understanding what the user is searching for, what questions they have, what helps them decide and what evidence demonstrates expertise.

During an audit of a medical B2B page about instrument quality, Shafique found repeated claims of high quality but no explanation of materials and manufacturing processes, quality control across different instruments, OEM and private-label standards, how quality is maintained during production, what international buyers should verify, or related products and manufacturing information. The page was rebuilt in that fixed sequence to make it more complete and useful, rather than merely longer.

Read original

High volume alone does not show whether the searcher is ready to buy.

• The approach described is to target low-competition terms first, check intent one by one, get backlinks from sources actually related to the product, and build one page around one keyword cluster so pages do not compete with each other.
• AI remains useful for speeding up research, drafting and structuring; the strategy still requires judgement.

A prospect running a surgical instrument company told Shafique he did not need an SEO agency because he had Claude.

• The tool returned high-volume terms with no difficulty scores, no search intent and no indication of who currently ranks for them.
• Shafique's point is that a site built on that list would spend months chasing terms held by established medical suppliers, and that the list cannot distinguish a hospital procurement manager ready to buy from a student doing coursework.

Read original

Agents can find keywords, analyze competitors, audit technical issues, generate content, suggest internal links, monitor rankings and create reports.

• But someone must first decide who the company is trying to sell to, which products matter most, which searches have commercial intent, which competitors matter, what the site architecture should look like, what to fix first and what success looks like.
• Those are business decisions, not SEO tasks.
• The proposed division is human strategy and judgement, AI execution and scale, and data as feedback.

A manufacturer entering the US market had a developer building AI agents on the belief that the agents could do SEO completely.

• Shafique listed the audience choices that still had to be made, including distributors, OEM buyers, hospitals and procurement teams, and asked who decides what work actually matters.
• The post concludes that automating an unchosen strategy simply moves the wrong work faster.

This is a consistent tension across the posts: Shafique is positive about AI tools for speed and scale, but treats human judgement as the part that selects the right target. The posts have no dates, so it is not possible to establish whether this view changed over time.

Read original

Search data can show which countries search for a product, which topics are growing, what language people use, whether they are researching, comparing or ready to buy, who already dominates the results and how difficult competition will be. This makes SEO more than a traffic channel: it can help test a market before committing heavily to localization, content, paid acquisition or expansion.

Before entering a new market, Shafique suggests using inputs such as Google Trends geographic interest data to see where demand is building and how hard competing will be. Ask separately about pricing, regulations, competition, logistics, purchasing power and customer fit.

Read original

It is the clearest single statement of what Azhan Shafique does and why the work goes beyond link building.

A surgical instrument manufacturer had a strong backlink profile, ISO certifications, regulatory references, industry memberships and a long trading history, yet ChatGPT and Google AI still described the company inaccurately.

• Rather than build hundreds of new links, Shafique rewrote and connected the information layer: who the company is, what it manufactures, its location, product categories, certifications, expertise and business history.
• They then had key pages re-indexed through Search Console.
• After that, both systems began identifying the business correctly.
• The post’s point is that authority already existed; the problem was making the business discoverable and understandable to search systems.

For brands planning a campaign

For a brand with real-world credentials that AI assistants misstate, budget should go to entity clarity and re-indexing before outreach volume. This is a low-cost entry point and a natural first milestone: correct the entity description across priority pages, then verify it by testing prompts against the assistant’s answers.

This concise account of a real conversation shows the gap between tools and judgment.

A prospect running a surgical instrument business said he did not need an agency because he had Claude, and had already asked an AI tool for keywords to target.

• It returned high-volume terms that looked impressive but had no difficulty scores or search intent, and gave no indication of who currently ranked for them.
• Shafique’s point is that a site built around that list would spend months chasing terms held by established medical suppliers.
• The list also failed to distinguish a hospital procurement manager ready to buy from a student doing homework.
• The preferred approach is smaller and more deliberate: start with low-competition terms, check intent one at a time, get backlinks from sources related to surgical instruments, and build one page around each keyword cluster so pages do not compete with each other.

For brands planning a campaign

Useful when a client is considering replacing agency support with a general AI tool. A campaign evaluating that replacement can start with intent classification and difficulty review, before any content is commissioned, and set an acceptance criterion that goes beyond volume.

For newcomers, it quickly shows how Shafique separates executable tasks from business decisions.

A manufacturer entering the US market had a developer building AI agents to automate SEO, believing the agents could do SEO completely.

• Shafique lists what those agents can do — find keywords, analyze competitors, audit technical issues, generate content, suggest internal links, monitor rankings and produce reports — then sets out what must be decided before any of it runs: which US buyers to target (distributors, OEM buyers, hospitals or procurement teams), which products matter most, which searches carry commercial intent, which competitors matter, what the site architecture should look like, what to fix first and what success looks like.
• The conclusion: automating a strategy nobody has chosen just moves the wrong work faster.
• The preferred division of labor is human strategy, AI execution and data as feedback.

For brands planning a campaign

Sets realistic expectations for a client seeking automation. It supports scoping a small strategy phase before tooling and offers a practical checklist — target buyer, priority products, intent, competitors, architecture and order of work — for a kickoff agenda.

It is the set’s most detailed technical example, showing the sequence from intent to call to action.

While auditing a medical B2B site, Shafique found a page about instrument quality that had plenty of words but repeated one claim: that the instruments were high quality.

• It did not explain what that meant to a buyer: materials and manufacturing processes, quality control across different instruments, OEM and private-label standards, how quality is maintained during production, what international buyers should verify or which products were related.
• The page was rebuilt in a fixed sequence: primary keyword, meta title, meta description, H1, H2s, H3s, semantic entities, internal links, external references and call to action.
• The goal was completeness, not length.
• Semantic SEO, the post argues, means understanding what the user is deciding, not repeating a keyword.

For brands planning a campaign

A page-level work order a client can approve directly. It defines what a rewritten product or capability page must contain and sets a review standard: if the page does not help a buyer verify materials, process and standards, it is not finished, regardless of word count.

It is the strongest introduction to the SEO-and-AEO pairing that runs through the rest of the work.

The post starts with a familiar pattern: a B2B site attracts visitors but still loses customers before contact because buyers cannot find clear answers about differentiation, technical requirements, experience or trust.

• The same problem extends to AI search: prospective buyers put those questions to ChatGPT and other assistants, and brands missing from the answers lose the opportunity before anyone reaches the website.
• Four moves are recommended: fix technical SEO so important pages can be crawled, understood and indexed; write content that answers buyer questions beyond generic product descriptions; build trust outside the website through industry mentions and consistent business information; and measure AI visibility, including which sources are cited and where competitors appear.

For brands planning a campaign

Gives clients a measurable frame: track brand mentions in relevant AI answers and the sources cited, not just rankings. It pairs naturally with the entity work in the featured post and can be built into milestone reporting.

It reframes SEO as a way to test a market before committing money to it.

Most companies use SEO data to ask how to get more traffic; Shafique argues the more valuable question is where demand is already building.

• Before entering a new market, the data can show which countries search for the product, which topics are growing, what language people actually use, whether they are researching, comparing or ready to buy, who already dominates the results and how hard it will be to compete.
• Google Trends is named as one source of geographic interest data.
• Search demand is not market demand: pricing, regulations, competition, logistics, purchasing power and customer fit still need to be validated separately.

For brands planning a campaign

Useful for a client weighing a move into a new geography or category. It supports a low-cost research stage before localization spend, with a short list of what search data can and cannot answer, plus a clear hand-off to pricing and regulatory checks.

Yet ChatGPT and Google AI still described the company inaccurately.

• Rather than building new links, the work focused on the information layer: who the company is, what it manufactures, its location, product categories, certifications, expertise, business history, and entity details.
• Important pages were then re-indexed through Google Search Console.
• The results cited in the post: AI systems identified the company correctly and surfaced more accurate information.
• Shafique’s point: sometimes a business does not need more authority; it needs the authority it already has to be understandable.
• That is where SEO and GEO converge.

For brands planning a campaign

For a medical device manufacturer, start by auditing how AI systems describe the company, then correct and complete the pages covering its business, products, certifications, and expertise instead of buying link volume.

Examples: Source 1

Buyers ask ChatGPT and other AI systems the same questions, they note, and brands absent from those answers can be missed before buyers reach the site. Their priorities: technical SEO so important pages can be crawled, understood, and indexed; content that answers buyer questions instead of repeating generic product descriptions; trust signals beyond the website, including industry mentions, credible sources, and consistent business information; and AI visibility measurement, including which sources are cited and where competitors appear instead.

Four-point working agenda: technical SEO, buyer-question content, off-site trust signals, and AI visibility measurement.

For brands planning a campaign

Set milestones for crawl and index health on priority pages and an inventory of buyer-question content. Track AI answer mentions and cited sources separately from Google rankings.

Examples: Source 1

Missing details included materials and manufacturing processes, quality control across different instruments, OEM and private-label standards, how quality is maintained in production, what international buyers should verify, and related products and manufacturing information.

• They rebuilt the page in an explicit order: primary keyword, meta title, meta description, H1, H2s, H3s, semantic entities, internal links, external references, and call to action.
• The goal, Shafique says, was completeness and usefulness rather than length.
• Semantic SEO means understanding what the user is searching for, what questions they have, what helps them decide, and what evidence demonstrates expertise.

A named build sequence—from primary keyword through H1, heading hierarchy, semantic entities, internal links, external references, and CTA—guided by the buyer’s decision rather than keyword repetition.

For brands planning a campaign

Offers a repeatable content brief for technical product and quality pages, plus a way to diagnose pages with substantial word counts but little buyer-relevant detail.

Examples: Source 1

The tool returned high-volume terms without keyword difficulty, search intent, or information about who currently ranked for them.

• Shafique’s point: a site built on that list could spend months chasing terms dominated by established medical suppliers.
• And volume says nothing about whether the searcher is a hospital procurement manager ready to buy or a student doing coursework.
• Their approach: start with low-competition terms, check intent one by one, choose backlinks from sources related to the product, and build every page around one specific keyword cluster so pages do not compete with each other.
• Shafique says AI tools are genuinely useful for speeding up research, drafting, and structuring, while judgement guides the underlying strategy.

For brands planning a campaign

Use AI-generated keyword lists as a starting pool, then validate difficulty and intent. For a new or thin-content site, prioritize commercially intended, lower-competition clusters.

Examples: Source 1

Shafique replied that agents can find keywords, analyze competitors, audit technical issues, generate content, suggest internal links, monitor rankings, and produce reports.

• But someone still has to decide the audience—distributors, OEM buyers, hospitals, or procurement teams—which products matter most, which searches have commercial intent, which competitors matter, what the site architecture should look like, what to fix first, and what success means.
• Shafique classifies those as business decisions, not SEO tasks, and proposes a division of human strategy and judgement, AI execution and scale, and data as feedback.

For brands planning a campaign

Before commissioning automated SEO execution for a US market entry, define audience segments, priority products, the competitive set, and success metrics. Then let agents scale execution under that strategy.

Examples: Source 1

Before entering a new market, they suggest checking which countries search for the product, which topics are growing, what language people actually use, whether they are researching, comparing, or ready to buy, who dominates the results, and how difficult the competition will be. Shafique cites Google Trends geographic interest data as one input and describes SEO as a market intelligence layer, not just a traffic channel.

Search demand is not market demand. Pricing, regulations, competition, logistics, purchasing power, and customer fit still need separate validation.

For brands planning a campaign

A manufacturer entering a new region can use search data to screen countries and topics for priority, subject to commercial validation. Build milestones around demand signals, not rankings alone.

Examples: Source 1

They sign off with this role, and their content positions the service around search strategy, AI search visibility, and buyer intent.

View example

The company already had backlinks, ISO certifications, and regulatory references, but ChatGPT and Google AI were misreading it.

• They say they rewrote pages covering identity, products, location, certifications, and business history, then reindexed key pages; AI systems subsequently identified the company correctly.
• The post presents this as completed work rather than an ongoing contract.

View example

One post lists ChatGPT prompts for featured snippets, topical maps, content briefs, internal linking and ranking inside ChatGPT; another checks off free Google tools, from Search Console to Trends, and explains what to look at in each. This reusable, teach-the-workflow approach serves the same practitioner audience and covers several subjects Shafique also writes about, including topical authority, keyword research and AI search.

Jain’s format is a distributable list of prompts and tools readers can apply immediately, illustrated with hypothetical prompts rather than client outcomes.

• Shafique writes first-person accounts of named client situations in medical and surgical instrument manufacturing, using them to make strategic arguments.
• Their posts address a narrower audience and read more like consulting arguments than a toolkit.

Read original · Original 2

His posts argue that AI answers reward clear category positioning, consistent terminology across pages and citations, specific outcomes and credible proof; generic product pages stay invisible; and product pages need entity work, intent mapping, schema and AI-search readiness. When Shafique rebuilds a manufacturer’s information layer, they make the same case for consistent brand terminology and clear entities.

Nisar generalises across ecommerce, SaaS, publishers and local businesses, offering a broad product-page framework with a mind map. Shafique focuses on B2B manufacturing and export, where the reader is a procurement or OEM buyer, and reasons from individual client engagements rather than applying a framework across sectors.

Read original · Original 2

Shafique also recommends choosing backlinks from relevant sources. Their entity post shows that acquiring links is not automatically the answer, giving readers weighing link work both perspectives.

Haider focuses on link-profile quality and penalty recovery.

• His posts centre on a panic call, a cheap competitor’s 250 links, disavowals and reconsideration requests, with monthly retainer pricing woven into the argument.
• Shafique leaves commercial pricing out of these posts and treats links as one part of making existing authority and information readable.

Read original · Original 2

The two posts offer twenty ChatGPT prompts covering featured-snippet targeting, competitor analysis, topical maps, content briefs, intent mapping, internal linking, FAQ schema and content-gap analysis, plus a walkthrough of eight free Google tools — Search Console, Analytics 4, PageSpeed Insights, Keyword Planner, Business Profile, Tag Manager, Merchant Center and Trends — and the decision each supports. For a project focused partly on upskilling an in-house SEO and content team on a modest budget, this is a more directly usable starting kit than client-work commentary.

The posts are generic and platform-level.

• They do not address medical devices, manufacturing, regulatory credentials, OEM or private-label buyers, or the challenge of correcting what AI assistants already say about an established company.
• They describe no completed buyer task: no diagnosed client, rebuilt page or before-and-after measurement of AI answers.
• The strategic decisions — which buyers, products and competitors to target, and what success looks like — would need to come from elsewhere.
• Prompt lists and tool inventories also age quickly as assistants and search surfaces change.

Examples: Source 1 · Source 2

One post sets out a product-page optimisation framework covering an entity-attribute-value structure; intent mapping across transactional, commercial, informational and navigational queries; title tags, meta descriptions and URL structure; schema markup; image optimisation; reviews and user-generated content; internal linking; variant and discontinued-product handling; and AI search readiness.

• The other argues that AI answers reward solution specificity and consistent category terminology, citing Amazon, Reddit, Docusign, G Adventures and Campfire Labs.
• That suits a manufacturer with a large, uneven device or instrument catalogue that needs a repeatable page pattern across hundreds of SKUs.

The framework is ecommerce-shaped and delivered as a mind map and lead magnet — comment to receive a PDF — rather than as an engagement.

• Neither post documents medical, surgical or regulated manufacturing work.
• The AI-search post comments on third-party analysis rather than describing a firsthand diagnosis of one company's AI answers.
• The entity-attribute-value structure transfers to a device catalogue, but the framework omits what this work requires: certifications, regulatory references and what international buyers should verify.
• A manufacturer would need to source that regulated-manufacturing layer separately.

Examples: Source 1 · Source 2

The two posts tell link-building case stories: one client’s rankings collapsed after 250 exact-match links were built from a single blog network alongside another provider's work; a second client’s fifty links produced no traffic because of repeated anchor text, irrelevant hosts, dead target pages, footer placement and a single-month spike.

• Both offer concrete repair steps — disavowal, reconsideration requests, an anchor-mix ratio, niche relevance over domain authority, traffic-qualified target pages, in-body placement and a paced monthly linking rate.
• If the manufacturer has previously paid for links and suspects the link profile caused lost visibility, this is the closer match.

The method covers off-page link work only.

• Neither post addresses entity definition, page-level information completeness, whether AI assistants describe the company accurately, or search-led market screening, so it does not meet the AI-visibility need.
• The delivery is a first-person case narrative with a stated monthly rate and penalty-recovery framing, unlike a technical working session with a procurement or engineering audience.
• Medical device and manufacturing clients are not indicated, and the outcomes described are traffic and ranking movements rather than qualified pipeline.

Examples: Source 1 · Source 2