Search intent is the least glamorous idea in SEO and the one that decides most outcomes. It is simply this: what did the person want when they typed that?
Get it wrong and everything downstream fails quietly. You write a genuinely good page, it does not rank, and nobody can tell you why — because the page is not bad, it is answering a question nobody asked.
The same words, different jobs
Take trekking Nepal. Three people type it in the same hour.
One is deciding whether to come to Nepal at all. One has booked flights and is choosing a route. One has chosen a route and is comparing three agencies on price.
One query, three completely different pages required. A pricing page shown to the first person is useless. An inspirational overview shown to the third is a waste of her time.
Google resolves this by looking at what people actually clicked and stayed on. Which is why the fastest way to understand intent is not to theorise about it, but to look at what is already ranking.
Read the results page, not the keyword tool
This is the whole technique, and it takes two minutes per query.
Search your target term. Ignore the ads. Look at the first five organic results and ask what kind of page they are:
- All listicles? Google has decided this query wants options to compare. Your single-service page will not rank, no matter how good it is.
- All product or service pages? This is a buying query. A blog post will not win it.
- All guides and explainers? Informational. Sending buying traffic here is unlikely to work.
- A map pack at the top? Local intent. Your Google Business Profile matters more than your website for this one.
- Mixed? Google is unsure, which means there is room — this is often the best kind of query to target.
Then check the format. Are they all long? All short? Do they all lead with a price table? If the top five all answer the question in the first paragraph, that is what this query rewards.
You are not copying them. You are reading the verdict Google has already reached about what this query means, so you can compete on the right terms.
The four intents, and what each is worth
Informational — "how long is the annapurna trek". Highest volume, lowest commercial value, and increasingly absorbed by AI answers before anyone clicks. Worth writing, but not as your main strategy.
Navigational — "nepal hiking team contact". They want a specific place. You cannot win someone else's navigational queries and should not try.
Commercial investigation — "best trekking company in nepal", "annapurna vs everest base camp". They intend to buy and are choosing. This is the most valuable and most neglected group. Comparison pages, honest pros and cons, real differences.
Transactional — "book annapurna trek 7 days", "root canal price kathmandu". Ready now. Lowest volume, highest value, and usually the queries the site never addressed.
Most Nepali business content plans are almost entirely informational, because that is where the volume is. The customers are in the last two groups.
Where to find real intent without paying for tools
Keyword tools cover Nepal poorly, and they cover Romanised Nepali worse. Four better sources:
Your own inbox. Every question a customer sends on Facebook, Viber or WhatsApp before buying is a query in your customers' exact words. Keep a list for a month. It will be more accurate than any tool.
Search Console. Filter for queries containing "price", "cost", "near", "best", "vs", "kati". Whatever appears is a commercial query you already get impressions for. Those are your fastest wins, because Google has already decided you are somewhat relevant.
Autocomplete on a phone. Type your service and stop. Read the suggestions. Then type it and add "ko" or "kati" and read them again. These are real queries with real volume.
People Also Ask. Expand a few and note the questions. They are Google telling you what else this searcher wanted to know.
The test before you write anything
Before committing to a page, answer three questions:
- What does this person want to do next? If the honest answer is "nothing, they just wondered", it is a low-value page.
- What kind of page is already winning? If it is a comparison and you are planning a service page, you will lose.
- What can I say that the current results cannot? If nothing, do not write it. A tenth version of an existing article has no reason to outrank the nine.
That third question is where most content plans quietly die, and it should. If you cannot answer it, the page will not work — and now that AI systems assemble answers from existing sources, a page that only restates what already exists has less value than ever.
Fewer pages, better chosen
The advice most Nepali businesses receive is to publish more. Two posts a week, consistency, momentum.
The businesses I have seen do well went the other way. Eight pages, each aimed at a query with a customer behind it, each genuinely better than what ranked. Then they improved those eight for a year instead of adding forty more.
Eight good pages beat eighty thin ones, and they are less work. I have written about why the opposite approach actively backfires in The SEO Content Trap.
The uncomfortable part of intent-led planning is that it produces a much shorter list than a keyword tool does, and short lists look lazy in a proposal. They are not. They are the result of throwing away the queries that were never going to produce a customer — which is most of them.
If you want help sorting a keyword list into these four groups, send it over. It is usually an hour of work and it changes what people write for the next year.