Before I write an SEO article, I check the SERP first.

Having an idea for an article is not the same as knowing what people search for.

And knowing a keyword exists is not the same as knowing what Google thinks that search means.

So before I write an SEO article, I search first.

Not because I want to copy whatever ranks.

I want to know whether the article I am thinking about writing belongs in that search result at all.

Sometimes the answer is no.

That is useful information to discover before spending several hours writing.

I start with the idea, not the keyword

Usually I already know roughly what I want to say.

For example, I recently wanted to write about something I have been changing in my own client process: research the market before deciding what website to build.

The idea came from a real proposal I had just prepared for a new beach club in North Phuket.

Instead of immediately quoting a website, the first engagement was designed to research:

• customer search demand
• competitors
• rankings
• Maps visibility
• reviews
• backlinks and authority
• positioning
• acquisition channels
• market gaps
• website and technology requirements

The research would determine what should actually be built.

I knew that was the article.

What I did not know was what search it belonged to.

That is where SERP research started.

The first keyword that sounds right can be wrong

My first instinct was:

market research before building a website

It sounded almost perfect.

So we searched it from the United States and looked at the actual results.

Google interpreted the query much more broadly than I did.

The results included advice about:

• validating business ideas
• interviewing potential customers
• demographics
• surveys
• startup planning
• testing demand before building a product

The U.S. Small Business Administration appeared prominently.

Google’s AI result also framed the subject largely around traditional market research and customer validation.

That was not the article I wanted to write.

I could have forced my article onto the keyword anyway.

Instead, I rejected the keyword.

That is one of the most useful things a SERP analysis can tell you.

The SERP is Google’s interpretation of the question

A keyword tool can tell you that people search a phrase.

The SERP tells you what Google believes those people are trying to accomplish.

Those are different pieces of information.

Before writing, I look at things like:

• what type of pages rank
• what industries or publishers dominate
• whether the results are guides, service pages, tools, videos or forums
• what Google includes in the AI Overview
• People Also Ask questions
• related searches
• whether the results are local, commercial or informational
• whether Reddit, YouTube, directories or publishers dominate
• how closely the ranking content matches the article I actually want to create

The point is not to recreate the first result.

It is to understand the environment the article would have to enter.

Then I search several versions of the idea

After rejecting market research before building a website, we tried related searches:

• competitor research before building a website
• website competitor analysis
• SEO competitor analysis before building a website
• SEO research before building a website
• how to plan SEO before building a website

The intent changed.

Now the results were talking about:

• identifying actual search competitors
• keyword and content gaps
• backlinks
• site architecture
• search intent
• technical SEO
• page hierarchy

That was much closer.

It also exposed the gap.

Most of the ranking content approached the subject primarily as an SEO planning exercise.

My argument was broader:

You should research the market itself before deciding what digital system to build.

SEO was evidence inside that research, not the entire research process.

That became the angle.

Only then did I write the article about researching the market before building a website.

The order the work runs in

1. 01Idea
2. 02Query options
3. 03SERP
4. 04Intent
5. 05Competitors
6. 06Gap
7. 07Brief
8. 08Write
9. 09Measure

Writing is the eighth step of nine. Everything before it is deciding whether there is an article here at all, and which search it belongs to.

I am not looking for a keyword to insert into my article

There is a common version of SEO writing that looks like this:

1. choose a keyword
2. put it in the title
3. put it in a few headings
4. mention it several times
5. write a meta description
6. publish

That is optimization.

It is not strategy.

The harder question comes earlier:

Does this article deserve to exist for this search?

If ten excellent pages already answer the question completely and I have no original experience, evidence, data or useful disagreement to add, publishing an eleventh version may not be the best use of my time.

A SERP analysis should help answer that too.

I want to know who owns the search

The ranking domains matter.

Not simply because a high-authority domain is difficult to outrank.

They tell me what type of information Google currently trusts for the subject.

For one query, the result might be dominated by:

• software companies
• universities
• government sources
• agencies
• specialist publishers

For another:

• Reddit
• YouTube
• independent experts
• small businesses

Those environments are very different.

If the page is full of giant software companies publishing broad beginner guides, I probably do not want to compete by writing another broad beginner guide.

I need a different reason for the page to exist.

That might be:

• firsthand experience
• a real case
• original research
• a narrower problem
• a useful tool
• a strong disagreement
• a better framework
• evidence the other results do not have

Otherwise I am producing content because SEO software told me there was a keyword.

I read what the ranking pages actually say

The SERP itself is the first pass.

Then I open the pages that are genuinely closest to what I want to publish.

For this article, for example, the current results already include good guidance saying that SERP research should happen before drafting. One recent guide from Stellar makes a particularly useful distinction: analyze the SERP once when deciding whether a topic deserves to be created, then again when building the actual content brief.

That is good advice.

So there would be little value in publishing an Aroc article whose entire argument was:

Analyze the SERP first.

The more useful addition is showing what happens when the research tells you the original keyword or article concept is wrong.

That is something we had just experienced ourselves.

I look for the missing part, not missing word count

A content gap is not necessarily:

Competitor A has 1,500 words and Competitor B has 2,000, so we should write 2,500.

It can be a missing idea.

A missing example.

A missing audience.

A missing piece of evidence.

A missing objection.

A missing distinction.

For the previous article, the missing distinction was between SEO research for a website, and researching the market the website has to compete in.

Those overlap, but they are not identical.

That distinction became the article.

Sometimes the correct decision is not to write

This is a useful possibility to preserve.

Content calendars create pressure to publish.

SEO tools create lists of opportunities.

AI makes generating the actual article almost free.

That combination makes it very easy to create hundreds of pages nobody needed.

I would rather abandon an article during research than publish something I cannot justify.

The questions I want answered before writing are:

1. Is anyone actually searching around this problem?
2. What does Google think they mean?
3. What type of answer are they expecting?
4. Who currently owns the result?
5. What do those pages already explain well?
6. What are they missing?
7. Do I have real experience or evidence to contribute?
8. Does this topic belong on Aroc?
9. If the article gets zero search traffic, would it still be useful to somebody?

If the answers are weak, the article can wait.

Search volume comes after intent for me

Search volume matters.

But I do not want it to be the first filter.

A phrase can have impressive volume and terrible relevance to the business.

Another may have modest demand but describe exactly the problem a potential client is trying to solve.

For Aroc, I care about whether the search sits close to a problem we actually understand and can help solve.

That is why I looked at the SERPs before worrying about exact volume for the first article.

First I wanted to know:

Are we even talking about the right question?

Once the query and intent are right, volume and difficulty become much more useful.

Location matters too

Search results are not universal.

The same query can look different depending on location, language, device and personalization.

For the previous article, we specifically checked the U.S. results because the article is intended primarily for the U.S. business market.

That does not create a perfectly objective SERP.

But it is much better than assuming the results I happen to see from wherever I am physically located represent the target customer.

For local-business content, this becomes even more important.

Someone researching best med spa or roofing contractor needs a geographically relevant SERP, not a generic national one.

My pre-writing workflow now looks like this

Before writing an SEO article, I work through this sequence:

1. Start with a real idea
   Preferably something that came from client work, a question, an assessment, Search Console, a sales objection, original research, or something I actually noticed.
2. Generate several possible search formulations
   Do not assume the first phrase is correct.
3. Search them in the target market
   Look at the actual results rather than only keyword-tool metrics.
4. Determine the real search intent
   What is the person trying to accomplish?
5. Identify who owns the SERP
   Which publishers, businesses, tools, forums and content formats repeatedly appear?
6. Read the closest competitors
   Not every result. The pages genuinely competing for the article I might create.
7. Find the information gap
   What useful thing can I add that is not simply another version of the same article?
8. Decide whether to write
   The answer is allowed to be no.
9. Build the brief
   Only now do I lock the primary query, secondary intent, title direction, article thesis, structure, evidence, internal links, CTA, and the claims I should not make.
10. Write
    Writing comes surprisingly late. That is intentional.

The article becomes easier once the research is done

There is another advantage to doing this work first.

The writing gets easier.

By the time I sat down to write the article about researching before building a website, I already knew:

• which query it belonged to
• what Google thought that query meant
• which adjacent query we had rejected
• what competing articles covered
• where my position differed
• which real proposal I could use as evidence
• which claims I could not make
• where the article should lead a potential client

The blank page was no longer very blank.

The research had already created most of the article.

This is the same principle I use before building a website

The interesting part is that this is really the same process as the previous article.

Before building a website:

Research the market it has to compete in.

Before writing an article:

Research the search result it has to compete in.

In both cases, the mistake is the same.

Production starts before anybody has understood the environment.

Modern tools make producing websites and articles incredibly fast.

That makes research more important, not less.

Because when producing the wrong thing becomes cheap, it becomes very easy to produce a lot more of the wrong thing.

I would rather spend the first part of the process deciding what deserves to exist. Then build it — which is what Assess does before Build, and what the Growth System keeps doing once a site is live.

Read next