Lately, I’ve heard more and more marketers and founders say that content has become a commodity.

I understand where that comes from.

AI has made it much easier and cheaper to create content. What used to take a content team days or weeks can now be done in just a few hours.

But producing more content is not the same as having a content strategy.

Writing 50 articles instead of 5 does not automatically build more authority, trust, or visibility.

In fact, I believe the human role is now even more important.

Someone still needs to decide what is worth publishing, what is actually different, what is supported by real experience, and what is just another version of something that already exists.

Having a person involved as an editor, strategist, and subject-matter expert still makes a big difference.

The tools have changed.

The need for judgment has not.

And this matters even more in AI search.

If every company can quickly produce technically correct content, the real advantage comes from sharing real experience, original information, good judgment, and a clear point of view.

I wrote about this broader issue in Content Velocity Is Not a Content Strategy.

The same principle applies to GEO.

There is no special formatting trick that will make ChatGPT, Google AI, Perplexity, or any other AI system pick your content.

Good AI search optimization starts with a simple idea:

Create information that is clear, useful, original, and worth finding.

What is AI search optimization?

AI search optimization means making your content easier for AI-powered search systems to understand, find, and possibly use in their answers.

It overlaps heavily with SEO.

Technical accessibility still matters.

Content quality still matters.

Authority still matters.

Internal linking still matters.

But the interface is changing.

Traditional SEO often focused on getting a page to rank for a query.

AI search adds another objective:

Can this information become part of the answer?

That’s why I don’t see AI search optimization as just another formatting step.

It’s really a part of your overall strategy.

Start with user intent, not the engine

A lot of SEO planning traditionally started with keywords.

What phrase are we targeting?

What is the search volume?

What is the intent?

Which page should rank?

Those questions are still useful.

But with AI search, you need to dig deeper.

The better question is:

What is the user actually trying to solve?

And then:

What context, constraints and decision criteria are included in that question?

A user might not simply want information.

They may want:

  • a recommendation
  • a comparison
  • a shortlist
  • a decision
  • an explanation tailored to their situation
  • a solution that fits a specific budget or use case

This changes how you should plan your content.

Search behavior itself has changed

I find this to be one of the most interesting changes.

Traditional search queries were often short.

Two words.

Three words.

Four words.

SEO strategies naturally followed that behavior.

We built keyword lists around short combinations and then expanded them into variations.

I also think there was a behavioral reason for this.

When we searched Google, we expected a list of websites.

We knew we would probably need to click into several results, scan pages, compare information and continue the research ourselves.

So there was often limited value in spending a lot of effort writing an extremely detailed query.

We already knew the search engine would send us into a research process.

AI search sets a different expectation.

Users increasingly expect the answer itself.

As a result, the prompt can become much more specific.

Instead of:

best CRM small business

a user might ask:

What CRM would you recommend for a 20-person B2B sales team that needs HubSpot integration, simple reporting and a low implementation cost?

That is not simply a longer keyword.

It includes:

  • context
  • constraints
  • preferences
  • use case
  • decision criteria

This is a meaningful shift for content strategy.

With traditional SEO, we got used to thinking mainly about keywords.

In AI search, we need to think much more in terms of:

questions, context, intent and decision criteria.

Keywords still matter.

But they are no longer enough to describe what the user actually wants.

Stop optimizing only for keywords

This does not mean keyword research is obsolete.

It means keywords should be just one part of your strategy, not the whole thing.

For example, suppose the target keyword is:

CRM for startups

That phrase tells us something.

But it does not tell us:

What stage is the startup in?

How large is the team?

Does it need sales automation?

Does it already use HubSpot?

Does it need a low-cost option?

Does implementation speed matter?

Does it need strong reporting?

These questions are more like how someone would actually interact with an AI system.

So when I plan content, I don’t just ask:

What keyword should this page rank for?

I would also ask:

What set of real questions should this page help answer?

Answer first, explain second

One of the easiest ways to improve your content is to give the answer right away.

If the heading is:

What is GEO?

do not start with four paragraphs of background.

Start with the answer.

For example:

GEO, or Generative Engine Optimization, is the process of improving the likelihood that a brand, source or piece of content will be included in AI-generated answers.

Then explain the details.

I like the principle:

Answer first. Explain second.

This is useful for people.

It is also useful for systems trying to identify the relationship between a question and the answer underneath it.

Direct answers do not make content shallow.

They make it clear.

Build clear information architecture

Good content structure has always mattered.

It’s even more important now, since content might be pulled in sections instead of read from start to finish.

A useful page should have:

  • descriptive headings
  • clear subheadings
  • short, focused sections
  • definitions
  • structured comparisons
  • tables when appropriate
  • concise paragraphs
  • clear terminology
  • self-contained answers

A heading like:

How do you measure AI visibility?

is more useful than:

The next step

The first heading has meaning on its own.

That is important for both users and retrieval systems.

Make entities explicit

AI systems need context.

So do readers.

If the content discusses a company, product, method, person or concept, use clear and consistent naming.

For example:

GEO, or Generative Engine Optimization…

Google AI Overviews…

ChatGPT…

Perplexity…

Avoid switching between vague references when the exact entity matters.

The goal is not to write robotically.

The goal is to remove unnecessary ambiguity.

Add first-hand experience

Sharing first-hand experience is one of the most valuable things your company can offer.

AI can summarize existing information.

It can structure it.

It can rewrite it.

But it cannot recreate the internal experience of your business unless you provide it.

That can include:

  • experiment results
  • campaign performance
  • product data
  • customer interviews
  • sales objections
  • support questions
  • failed tests
  • operational lessons
  • pricing experiments
  • implementation challenges
  • internal benchmarks

Compare these two statements.

Generic:

Internal linking is important for SEO.

More useful:

After reorganizing a fragmented set of articles into one clear topic cluster and linking them back to a central pillar page, we saw related pages begin surfacing together across more adjacent search queries.

The second version has context.

It has experience.

It gives the reader something they cannot get from a generic summary.

AI can structure experience. It cannot manufacture authentic experience.

Create information gain

Before publishing, I like to ask one question:

What will the reader know after this page that they probably did not know before?

That is a useful practical definition of information gain.

The answer could be:

  • a new data point
  • a framework
  • a benchmark
  • an expert opinion
  • a case study
  • a detailed example
  • a comparison
  • a process
  • a template
  • a tool
  • a different interpretation

You don’t need groundbreaking research in every article.

But there should be a reason the page exists.

If the page says exactly what ten other pages already say, using slightly different wording, it is hard to argue that it adds much value.

Use evidence, examples and sources

Broad claims are easy to make.

Specific evidence makes them more useful.

Instead of:

Users are asking more complex questions in AI search.

Show examples of the difference between a traditional keyword query and a detailed AI prompt.

Instead of:

Our content improved.

Show what changed:

  • conversion rate
  • engagement
  • citation frequency
  • branded search
  • rankings
  • qualified traffic
  • customer feedback

Not every article needs to become a research paper.

But the more important the claim, the stronger the evidence should be.

Internal linking should create context

Internal linking should do more than distribute SEO value.

It should explain how the ideas on the site connect.

For example, this article belongs inside a broader topic around Organic Growth & AI Visibility.

It also connects directly to SEO vs GEO: What Actually Changes in the AI Search Era.

And the discussion about quality versus scale connects to Content Velocity Is Not a Content Strategy.

That creates a clear topic cluster.

One page explains the strategic shift.

Another explains execution.

Another explains content quality.

Measurement is another layer. I cover that process in How to Measure AI Visibility.

This approach is more useful than publishing lots of separate articles on unrelated keywords.

Brand authority matters beyond your website

Your website is only one source of information about your company.

A brand can also appear in:

  • LinkedIn
  • Reddit
  • YouTube
  • podcasts
  • industry publications
  • customer reviews
  • partner websites
  • comparison pages
  • communities
  • research
  • conference websites

This matters because AI visibility is not necessarily created by your own domain alone.

The broader context around the brand can matter too.

That means GEO increasingly overlaps with:

SEO, PR, distribution, thought leadership and brand building.

I would not treat these as completely separate systems.

What not to do

There are a few patterns I would avoid.

Do not create AI keyword stuffing

Repeating phrases such as “AI search”, “GEO”, “ChatGPT optimization” and “generative search” dozens of times does not create useful content.

Do not create FAQ spam

FAQs are helpful when they answer real questions.

They are not useful when they simply repeat the article in slightly different wording.

Do not publish dozens of near-duplicate pages

If five pages answer essentially the same question, consider whether one stronger page would be better.

Do not write only for machines

If content sounds unnatural, repetitive or overly engineered, the problem is not that the reader does not understand GEO.

The content simply is not good enough.

Do not assume volume equals authority

Publishing hundreds of pages may increase the size of a website.

It does not automatically increase credibility.

Do not remove the human editor

AI can accelerate drafting.

But good judgment still matters.

Someone needs to ask:

Is this accurate?

Is this useful?

Is this different?

Does it reflect real experience?

Should we publish this at all?

That role has not disappeared.

A simple framework for AI search content

I like using a simple sequence:

Question → Direct Answer → Evidence → Example → Context → Internal Links

Question

What is the real problem the user is trying to solve?

Direct Answer

What is the clearest possible response?

Evidence

What supports the answer?

This could be data, experience, a credible source or a clear argument.

Example

What does it look like in practice?

Context

What else does the reader need to know before acting on the answer?

What additional information on the site helps them go deeper?

This framework is straightforward.

That’s on purpose.

You can use it for almost any educational article, comparison page, or expert guide.

A practical pre-publish checklist

Before publishing content for SEO and AI search, I would ask:

  • Is the main question clear?
  • Is the answer visible early?
  • Are the headings descriptive?
  • Is the page easy to scan?
  • Are important entities clearly named?
  • Is the terminology consistent?
  • Is there first-hand experience?
  • Is there something original?
  • Are important claims supported?
  • Are the examples specific?
  • Are internal links useful?
  • Does the page belong to a larger topic cluster?
  • Is there a clear reason this page deserves to exist?

If most of those answers are yes, the content is probably moving in the right direction.

AI search optimization is mostly quality optimization

There is a risk that GEO becomes another industry built around hacks.

I do not think that is where the sustainable opportunity is.

The strongest content is still content that is:

clear.

useful.

credible.

specific.

original.

well structured.

connected to real expertise.

supported by evidence.

AI search changes how information is discovered and presented.

But it does not eliminate the value of judgment, expertise or originality.

If anything, as content becomes easier to produce, those things become more important.

The goal should not be:

How do we make an AI engine choose this page?

A better question is:

How do we make this one of the best sources available for the question?

If you have something valuable to share, publish it so it’s easy to understand, easy to find, and worth citing.

Frequently Asked Questions

What is AI search optimization?

AI search optimization is the process of improving content so it is easier for AI-powered search systems to discover, understand, retrieve and potentially use inside generated answers.

How is AI search optimization different from SEO?

SEO primarily focuses on visibility in traditional search results. AI search optimization also considers visibility inside generated answers, citations, mentions and recommendations. The two disciplines overlap significantly.

Do I need separate content for ChatGPT?

Not necessarily. The goal should not be to create separate pages for individual AI engines. Strong content should answer real user questions clearly and provide useful, credible information that can work across multiple search environments.

Does GEO replace SEO?

No. GEO builds on many of the same foundations as SEO, including technical accessibility, content quality, authority, internal linking and relevance.

Do FAQs help with GEO?

They can, when they answer genuine questions clearly. FAQ sections created only to add more text are unlikely to add much value.

How important is original information?

Very important. Original data, first-hand experience, expert insight and unique examples give a page a stronger reason to be used, referenced or cited.

How do I know if content is optimized for AI search?

A good page should answer a clear question, provide the answer early, use descriptive structure, contain useful evidence or experience, define important entities and connect naturally to related content.