01EDUCATE
Category and problem pages
Define the market without trapping the strategy inside terminology nobody searches yet.
SEO for AI startups
Your product may be novel. Your market still searches in familiar language.
I help AI startups translate new capabilities into the problems, workflows, use cases and buying decisions people already understand—then build the organic system that captures and shapes that demand.
4,705%organic traffic increase
3,500%organic growth
190%AI-referred sign-up growth
88×organic traffic growth
The demand translation
Stop leading with the model. Start with the job.
Buyers rarely begin by searching for your architecture. They search for the process they want to automate, the risk they need to reduce or the outcome they need to reach.
PRODUCT LANGUAGE
Agentic workflow orchestration
Multimodal retrieval
Autonomous context layer
MARKET LANGUAGE
Automate compliance reviews
Search video archives
Answer customer questions faster
AI-startup organic blueprint
Build the surface area from pain to product decision.
02DEMONSTRATE
Use cases and workflows
Show the product completing real jobs for identifiable teams, industries and operating contexts.
03CONNECT
Integrations and ecosystems
Capture demand around the tools, data and workflows your product improves or replaces.
04CONVERT
Alternatives and comparisons
Own the buying moment with credible differentiation, evidence and implementation detail.
Where scale becomes possible
Programmatic SEO without programmatic thinness.
AI products often sit across repeatable combinations—use case, role, industry, integration, template or data source. That can become a scalable acquisition system when every page has genuine product value.
Pages for who performs the job.
Examples: AI research assistant for analysts, AI intake for personal-injury teams, AI support for product operations.
Pages for where the product fits.
Examples: connect the product with Slack, Salesforce, Notion or an industry-specific system.
Pages for what the product changes.
Examples: turn calls into tasks, contracts into risk flags, or datasets into decision-ready answers.
The proprietary-data moat
Do not begin with 10,000 keywords. Begin with 10,000 useful things.
AI has made page production cheap. The durable advantage is the product, operational, user or transaction data a competitor cannot reproduce—and whether it helps a buyer make a better decision.
Exclusivity
Does the information come from your product, workflow or market activity?
Decision value
Does it help someone compare, choose, plan or act with more confidence?
Entity coverage
Can it support distinct pages across real suppliers, workflows, assets or use cases?
Freshness
Does the data improve as the product is used and the market changes?
Commercial proximity
Is the useful next step naturally connected to what the product enables?
The pipeline reversal
Turn the product database into an acquisition engine.
Traditional programmatic SEO starts with public keywords and hopes the resulting pages are useful. This system starts with proprietary value, then finds the search demand it can serve.
Model the data
Create a clean ontology for the entities, attributes and relationships the product genuinely knows.
Generate candidates
Combine entities with decision modifiers: role, location, workflow, quality, price, risk or outcome.
Cluster by meaning
Group candidates by semantic intent so similar language does not create duplicate experiences.
Validate demand
Use search data plus first-party signals such as product searches, views, saves, enquiries and transactions.
Set eligibility
Publish only where inventory depth, unique data and decision value clear a defined threshold.
Match the page type
Use inventory pages, commercial research or buying guides according to the job behind the query.
Connect to action
Let discovery flow into a live product experience: filter, compare, generate, evaluate, enquire or sign up.
PRODUCT USE → BETTER DATA → MORE USEFUL PAGES → QUALIFIED DISCOVERY → MORE PRODUCT USE
What this looks like in practice
Find the graph hidden inside the product.
The opportunity is rarely another generic integrations directory. It is the relationship between the data your product creates and the decisions your customers repeatedly make.
Inventory × supplier × demand.
Pages can reveal availability, quality, geography, pricing patterns or buyer demand—then connect the visitor to the relevant inventory or supplier.
Product action: filter, compare or enquireRole × workflow × outcome.
Anonymised workflow patterns can show how teams complete a job, what inputs are involved and what successful output looks like.
Product action: run or adapt the workflowEntity × signal × decision.
Original market observations, benchmarks or risk signals can support research pages that become more valuable as the underlying dataset grows.
Product action: inspect or analyse the entityThe publication gate
A URL is earned. Not generated.
Every page type needs an explicit launch, refresh and retirement rule. This protects the domain from thin inventory, duplicated intent and stale promises.
Demand is real, the page has enough unique data, and the experience improves a buyer decision.
Inventory, product activity, market evidence or the supported workflow materially changes.
The page is useful inside the product but has insufficient external demand or standalone value.
Supply disappears, the entity becomes obsolete or the page can no longer satisfy the intent credibly.
THE CONTRARIAN TAKE
Most AI startups do not need 100 “AI tools” articles.
They need a smaller number of commercially connected pages that explain the category, demonstrate product truth, answer evaluation questions and create a path from discovery to activation. Scale comes after the information model works—and privacy, clean taxonomy and product truth are non-negotiable.
See how we could work together →Turn novelty into demand
Your product may be new. The buyer’s problem is not.
Let’s translate the product into a search market, identify the acquisition surfaces worth building and avoid wasting months on generic AI content.
Book a strategy conversation →30 minutes · your actual growth opportunity · no generic pitch deck