How to conduct employer brand audience research

Summary

• The more specific the audience, the more likely you'll need bespoke research.
• Combine different types of research for a more rounded understanding.
• Try to find the right balance between the quantity and quality of your research.
• Be realistic about who your audience are and what you can afford.

First things first, audience research for an employer brand is a lot harder than you might think. And it all depends on the who your audience is. The most straightforward type of audience research comes when you're trying to understand general attitudes. For this, you can usually get a nationally representative sample without too much difficulty. If you want to research a niche or specific audience, then it becomes more challenging, and this is more often the case.

Where do you start with audience research?

If the audience for your employer brand is large (e.g., retail workers, nurses, teachers), there is actually a wealth of high‑quality public data available, particularly from national statistics bodies like the ONS in the UK. This data can be broken down by region or demographic, although as a heads up, the categorisation on these sites can sometimes be counter‑intuitive or grouped in unhelpful ways. The more specific the audience is that you're looking for, the less likely it is that they already exist on a ready‑made panel. This means you'll need to source these people yourselves and build a bespoke panel to conduct your research.

How do I approach audience research?

Desk research

There is usually huge value in doing desk research before starting primary research. This includes reviewing public reports, existing market analysis, social listening, industry publications, and any internal historical insights.

Qualitative research

Qualitative research uses deep, exploratory methods like interviews or focus groups. This can help you to uncover the motivations, behaviours, and reasons behind people's attitudes.

Quantitative research

Quantitative research uses structured surveys to give us measurable patterns, trends, and a more representative picture of an audience.

Behavioural research

As well as listening to what people say, we can often look at downstream behavioural data too. This includes elements like funnel analytics, hiring trends, engagement signals, or performance data. This helps us compare stated attitudes with actual behaviour, which can give a more rounded view.

Compare your findings

The most reliable audience understanding usually comes from combining multiple sources of insight, for example, using desk research to frame hypotheses, qualitative research to explore the "why", quantitative research to validate scale, and behavioural data to test whether real‑world actions match what people report.

Even imperfect data can help to guide decision‑making, test messaging, and illuminate attitudes, and that makes the research absolutely worth doing. It simply means you also need to budget more than you might expect if you want to do it well.

How do you decide which research you need?

Let's talk about a specific example. If we want to understand submarine engineers within 100km of Amsterdam, we need to decide whether we want 200 respondents for a broad, representative picture, or 10 for a more detailed but unrepresentative view.

The thing is, submarine engineers in Amsterdam aren't lining up on a public database, ready to be contacted. And sourcing niche audiences usually requires proactive outreach that is both difficult and expensive.

Quality also matters. You want insights from submarine engineers who are genuinely experienced and relevant, not loosely associated or inaccurately self‑reported participants.

I always ask clients, "How many of your best people regularly complete long surveys or take part in competitor research?" The reality is, not many, and this affects the quality of respondents.

So, ultimately, you need to decide what kind of a view you're looking to create and what you can afford.

Are there any other ways to conduct research?

You can, in some cases, use internal colleagues as a strong and cost‑effective proxy audience, especially when the audience you're exploring closely resembles your own people. This approach won't work if you're entering a new market or exploring an audience you don't employ, but in many scenarios it's a highly effective alternative.

As for off‑the‑shelf audience research tools, we would always recommend caution. They claim cheap access to niche audiences, but unless they have an illegal database, building a panel of Amsterdam‑based submarine engineers, for example, is neither simple nor low‑cost. If a tool is cheap, it's likely using an overly broad or poorly‑matched audience, or increasingly, relying heavily on synthetic data. This is risky because it often reflects the assumptions and biases of the model rather than the real attitudes of the audience, which can lead to false patterns and misleading strategic decisions.

Ultimately, conducting audience research in employer branding means being realistic about who your audience are, how reachable they are, what quality looks like, and how to combine the right methods to get credible insight.

If you'd like to discuss audience research further or you'd like to work with us to research your audience, you can get in touch at hello@thirtythree.co.uk.