A business owner shows me a graph going up and to the right. Organic traffic has tripled in eight months. Then he tells me, more quietly, that enquiries have not changed at all.
He thinks he has a conversion problem. He wants to talk about button colours and form fields.
He almost never has a conversion problem. He has a traffic problem that looks like success on the only chart anyone showed him.
The two kinds of visitor
Search traffic is not one thing. Consider two people arriving at a trekking company's website.
The first searched height of annapurna base camp. She is a student writing an assignment. She reads one number and leaves in nine seconds.
The second searched annapurna base camp trek 7 days price. He has three weeks of leave and a budget. He reads two pages and opens the contact form.
In your analytics these are both "one organic session". They are not remotely the same thing. And here is the trap: the first query is much easier to rank for and has far more volume, so a content strategy aimed at growing traffic will produce hundreds of the first and almost none of the second.
The graph goes up. The business does not.
How the wrong traffic gets built on purpose
This usually happens because somebody was paid to increase traffic and did exactly that.
Informational keywords are cheap to rank for. "What is X", "history of Y", "10 facts about Z". Write twenty of those and traffic rises measurably within a few months. The report looks excellent.
Commercial keywords — the ones with a buyer behind them — are contested, because everyone competing for that customer wants them. Ranking takes longer and the volume numbers look unimpressive next to the informational wins.
So the incentive, if you are being judged on traffic, is to chase the easy queries. It is not usually dishonest. It is just optimising the number that was put in front of everyone.
Find the gap in twenty minutes
Open Google Search Console, go to Performance, set the date range to the last three months, and sort queries by clicks.
Now read the top twenty and sort them into three groups by hand:
- Buying — price, cost, booking, near me, best, your service plus a place name, comparisons.
- Researching — how, what, why, guides, definitions.
- Irrelevant — queries where the person could never become a customer.
Count the clicks in each group. In most sites I have looked at where the owner has this complaint, buying queries are a small fraction of total clicks, sometimes under a tenth.
That ratio is the answer. You do not have a conversion problem. You have ranked for the researching group and not the buying group.
The second check: which pages do the work
Still in Search Console, switch to Pages. Then compare against your enquiries.
What you usually find is that the pages with the most traffic and the pages that produce customers are two different sets, with very little overlap. The blog post about trekking seasons gets 4,000 clicks a month and produces nothing. The unglamorous page listing what is included in the 7-day package gets 180 clicks and produces most of the enquiries.
Once you can see that, the strategy writes itself. You have been improving the wrong page for a year.
What to do about it
Write the pages buyers search for. This is the whole fix and it is unglamorous. The pages that convert are the ones people are slightly embarrassed to ask about directly: what it costs, what is included, what is not included, what happens if it goes wrong, how you compare to the alternative.
Most Nepali business sites will not publish prices. I understand the reasoning — it varies, competitors will see it, you want them to call. But price is one of the highest-intent words a customer can type, and refusing to address it does not stop them searching. It just means somebody else answers.
You do not have to publish a single number. A range, plus an honest explanation of what moves it up or down, ranks for the query and builds more trust than silence.
Make the researching traffic earn its place. You do not have to delete it. But a guide to trekking seasons should end by pointing at the treks you actually run in those seasons. Right now most of these pages end and stop, having given the reader everything and asked for nothing.
Fix the obvious friction, but do it last. If your contact form asks for eleven fields, cut it to three. If your phone number is an image, make it a link. If your enquiry form breaks on mobile — worth checking, because a surprising number do — fix that first of all.
But be clear that this is a small effect compared to the traffic mix. Doubling conversion rate on the wrong audience still leaves you with the wrong audience.
Change what you report
If you take one thing from this: stop reporting sessions as the headline number.
A more honest monthly summary is four lines:
- Clicks from buying-intent queries
- Enquiries
- Enquiries divided by clicks from buying-intent queries
- Which pages produced them
This is a less impressive report. It will sometimes show a month where traffic fell and the business improved. That is the point — that month actually happened, and the old report would have hidden it.
This matters more now than it did two years ago. With AI answers absorbing a growing share of informational queries, informational traffic is falling for reasons that have nothing to do with your performance. If sessions are your headline metric you will read that as failure and respond by chasing more of exactly the traffic that is disappearing. I have written about that in more detail in Google Search vs AI Search.
The uncomfortable version
Sometimes the traffic is right, the pages are right, and the enquiries still do not come. Then it is worth considering that the problem is not the website.
It might be that your prices are genuinely uncompetitive and the page makes that clear. It might be that your reviews say something you have not read recently. It might be that the thing you are selling is not what people want to buy.
Before you conclude the offer is the problem, check the form still works — contact forms fail silently more often than anyone expects. If the numbers themselves look wrong rather than the offer, that is a different investigation — start here. A website makes you findable. It cannot make an offer attractive. Every so often the most useful thing an SEO can tell a client is that the search data is working correctly and it is delivering an honest verdict.
If you want a second pair of eyes on which group your traffic falls into, get in touch. The Search Console export usually settles it in one sitting. There is more about choosing queries by intent if you would rather work through it yourself.