Search intent is the bit nobody actually does properly

Search intent is the most over-explained, under-applied concept in SEO. Everyone writes about it. Almost nobody actually checks it before they publish. Here’s how we do, and why it’s the single highest-leverage habit in SEO content writing.

Pick any SEO blog post written in the last five years (Ahrefs has a particularly clear explainer if you want the canonical version) and there’ll be a paragraph about search intent. Usually a list of four categories, sometimes a Venn diagram, often a worked example involving “best running shoes.”

The problem isn’t the explanation. The problem is that almost no one applies it. They tick the “we did search intent analysis” box, they label the keyword “informational,” and then they write the article they were going to write anyway. The intent label was decorative.

This post is how we actually use search intent in practice. It’s less elegant than the textbook version but it’s the version that makes the difference between content that ranks and content that gets buried on page 3.

The four buckets, briefly

For completeness, the categories everyone teaches:

• Informational: the searcher wants to know something. (“How do espresso machines work?”)
• Navigational: the searcher wants to get to a specific website or page. (“Jura coffee machines”)
• Commercial: the searcher is comparing options before buying. (“Best commercial coffee machines under $5,000”)
• Transactional: the searcher is ready to buy. (“Jura WE8 commercial coffee machine buy”)

Real queries blend categories. “Commercial coffee machines for cafes” sits between commercial and transactional. That’s fine. The labels are a starting point, not an answer.

The thing nobody does

Here’s the move that 90% of SEO content writers skip: before you write anything, type the target query into Google and study the top 10 results.

Not skim. Study. Specifically:

1. How long are the top-ranking pages? (Average, not the longest or shortest.)
2. What format are they? (Listicle, how-to, comparison, single deep-dive, video page, product page?)
3. What’s the dominant H2 structure? (For competitive head terms, you’ll often see silo-style structures dominate.) (Numbered list of products? Decision-tree style? Question-and-answer? Process steps?)
4. Is there a featured snippet? What does it grab from?
5. Are the top pages from publishers (Forbes, Wirecutter), retailers (Amazon, your competitors), or specialists (industry blogs)?

The answers to those five questions tell you the SHAPE the article needs to take. Search intent isn’t really about whether the query is informational or commercial. It’s about what KIND of page Google has already decided wins for that query.

A worked example

Let’s take a real query we worked on recently: “commercial coffee machine for office of 50”.

By the textbook, that’s a commercial-intent query. Someone is comparing options before buying. So you’d write a comparison article: “The 7 best commercial coffee machines for offices of 50+.”

But when we actually searched it, the top 10 was a mess:

• 3 product pages from coffee suppliers (transactional pages, not comparison content).
• 2 “how to choose” guides (informational/decision-tree format, around 2,000 words).
• 1 listicle (“Top 5 machines for medium offices”).
• 3 niche industry blog posts on capacity calculations.
• 1 featured snippet pulled from a 400-word definition of “office coffee” on a glossary page.

This told us several things the textbook label didn’t:

• The format isn’t settled. Listicles, decision guides, and product pages all rank. There’s room.
• Capacity is the dominant subtopic across multiple pages. It’s clearly something the searcher actually cares about.
• Specific machine recommendations are valued (3 of 10 results are actual products).
• The featured snippet sits on a 400-word page. Beatable with a tightly-written definition paragraph.

So we wrote a decision-tree guide structured around capacity calculations, with specific machine recommendations embedded at each capacity tier, and a 60-word definition near the top targeting the featured snippet. That article ranked on page 1 within six weeks and pulled the featured snippet within ten.

The textbook label would have produced a generic “Top 7 machines” listicle. Studying the SERP told us what to write instead.

The four shapes you’ll actually see

After about 1,000 SERPs, you start noticing the same shapes:

1. “The settled topic”

9 of the top 10 are the same format. (Usually listicles or how-to guides for the obvious commercial queries.) You either match the format and win on quality, or you skip the query because you can’t outrank entrenched publishers.

2. “The transactional ambush”

Half or more of the top 10 are product pages, not articles. Google has decided the user wants to buy. If you don’t sell the thing, this query is not for you, no matter how high the search volume is. We see this a lot in legal and finance. Sometimes the SERP is just lawyer/firm pages, and a blog post can’t get in.

3. “The unsettled format”

The top 10 is a mix of listicles, guides, definitions, product pages. There’s room. This is where SEO content writing genuinely earns its money. Choose a format that fills a gap (a more decisive listicle, a more thorough guide, etc.).

4. “The featured-snippet leak”

The featured snippet is pulled from a small, weakly-optimised page. You can grab it with a clean 40-60 word definition or a structured list near the top of your post. Easy wins.

How to actually do this in 15 minutes

You don’t need a tool. You need 15 minutes per query and the willingness to actually read the top 10 pages.

1. Open the query in an incognito tab.
2. Note the SERP features: ads, featured snippet, “People also ask” panel, knowledge panel, video carousel.
3. Open each of the top 10 results. Note: domain type (publisher / retailer / specialist), format (listicle / guide / product), word count (rough), and dominant H2 structure.
4. Write one sentence summarising the dominant shape. “Most top-rankers are 1,800-2,500 word decision guides organised by use case, written by specialists, with product recommendations embedded throughout.”
5. Decide: match the shape and win on quality, or find a gap in the shape and exploit it.

That sentence is the actual answer to “what’s the search intent?” Not the label. The shape.

What we do at SEO Sandwich

Every piece in a Classic Sandwich or Full Stack Sandwich gets this treatment. We run the SERP analysis before we outline the article. The outline reflects the shape we found, not the shape we assumed. The result is the difference between a blog post that exists, and a blog post that ranks.

It’s also why our sprint timelines look longer than some agencies’. 30-40 days for 20,000 words isn’t slow. It’s because 4 of those days are SERP analysis and outline. The writing is the fast bit. The thinking is the bit nobody else does.

See the packages, or book a call to talk through which queries you’d actually go after.