Does search volume tell you anything about applicants?

Behind a single term there are usually three kinds of people, and all three want something different.

The first one is looking for work. Someone typing in talent, or a job title, may just as easily be after a job as after a colleague. For an employer advertising on that term this is confusing news and good news at the same time, because that group is exactly the audience for a vacancy campaign, just not for an advertisement about your services.

The second one is studying the field. Students, HR professionals reading up, people writing a thesis, colleagues from the industry checking what a term means. They want a definition, not a supplier and not a job. They click, they read, and they never respond to anything.

The third one is an employer with a problem. That person is looking for help, and it is the only group that responds to an advertisement about services. In the English terms around the word talent that is 24 per cent of the volume. And that is before the nine purely candidate-facing terms our measurement leaves out, worth another 14,460 searches a month. Count those in and roughly five sixths of the traffic you buy on such a term is not the party you are looking for, which the keyword tool does not show you, because it shows only the number.

That distinction runs through everything. In a check on 28 August 2026 we re-measured 2,005 groups of searches whose meaning we did not yet know. Of the search volume in them, 87 per cent turned out to be about something other than hiring: 1,684 groups worth 2,284,830 searches a month dropped out, against 125 groups worth 151,960 that stayed. Not because the terms were nonsense, but because they were about something other than what we assumed.

We measured 78,975 searches in four languages, of which 48,407 turned out to have measurable volume, and then went back over them to work out who is searching. That is a lot of work for a question you can usually answer in five minutes. Three tests get you most of the way.

Read the results, not the number. Type the term in and look at what comes back. Job boards, CV tips and interview questions mean someone is looking for work. Suppliers, prices and comparisons mean an employer is looking for help. The search engine has already answered the question for you, and all you have to do is read it.

Look at what sits at the very top. One job board in the results can be a coincidence. If the first three results are full of vacancies, the term belongs to candidates, however high the volume is. Advertising a service on a term like that costs money and buys visits that have nothing to do with your question.

Be suspicious of brand names and double meanings. A sizeable part of the volume we had to strike out turned out to be about something that happens to share a name: a software product, a company, a television programme. In the tool a term like that looks like a gold mine. In practice it is a leak, and it is the single most common reason a big number produces nothing.

The reasoning that follows is less complicated than it sounds: a small keyword that the right people search for beats a large keyword that they do not. Not because small is more elegant, but because on the large keyword you pay for the overwhelming majority of your impressions to reach people who were never going to do anything.

That has two consequences for how you judge a campaign. First, traffic is not a goal. A campaign that delivers plenty of visits and no applications is usually not a bad campaign but a campaign on the wrong terms. Second, the low volume of a good term is no reason to leave it alone. A modest term that only employers search for is worth more than a busy one with a mixed audience.

In practice it comes down to deciding who is searching before you decide how much to spend. That is the work behind job marketing and the wider recruitment marketing, and it is why in data and optimisation we look at the composition of the traffic before we look at its size.

Can I trust the volume in a keyword tool at all?

You can trust it as a count of searches. What it does not tell you is who those searchers are, and that is the part that decides whether a campaign works. In the English terms around the word talent, 24 per cent of the volume comes from an employer who is hiring, measured on 26 August 2026. For other groups of terms the share is higher or lower. You only know once you look at the term itself.

My campaign gets clicks but no applications. Why?

Usually because it runs on terms where most searchers want something else. Clicks cost money and tell you only that an advertisement stood out. So look first at who responds and only then at how many. If this sounds familiar, there is more about it on too few applicants.

How do I tell whether a term belongs to candidates or to employers?

By typing it in and reading the results. Job boards and application tips at the top mean candidates, suppliers and prices mean employers. On top of that, check whether the term is also a brand name or has a second meaning, because that is the most common reason a large volume delivers nothing in practice.

Is a small keyword worth building a campaign on?

Often yes, provided the people behind it are employers and you accept the ceiling that comes with it. A term with a few hundred searches a month from employers only is a better starting point than a term with tens of thousands from a mixed audience. What it will not do is fix the rest of the funnel, which is where wrong applicants and our approach come in.

We measure this distinction because we need it ourselves when dividing campaign budget. If you want to know what the terms around your own roles look like, SocialFind is happy to take a look with you. If you would like measurements like this in your inbox, sign up for our newsletter.