When 99% of your audience tells you they want more investigative journalism, don't believe them

A beginners guide to audience surveys and interpreting research data.

If you work in digital journalism, especially in audience revenue or distribution-focused roles, you've likely encountered some version of these two statements:

"99.9% of the audience says they want more investigative journalism"

and

"15/20/25% of people in [your country] pay for news"

That’s when you start to wonder whether your newsroom is just bad at its job, because those hundreds of thousands aren’t showing up in your revenue dashboard, or if you’re living in a simulation where infinite copies of Hugo Weaving are answering surveys instead of actual humans.

I don't mean to say research isn't useful, far from it. I like audience research almost as much as newsletters (and as you might know: I LOVE newsletters). In fact, over the past few years, our team has helped publishers design, distribute and analyze audience surveys in almost every CEE country. Since I'm still in Bucharest between two survey-related engagements, and we promised to bring you something on this towards the end of last year, I want to spend today's edition going over 9 practical tips around audience research and data.

1. Do audience research

Without an audience, journalism has neither a purpose nor an income. So even if you're not crazy about your audience (I know a surprising number of people who work in journalism and have, to put it mildly, complicated feelings toward their own readers/viewers/listeners), it makes sense to want to understand them better. This is especially true if you are in, or planning to move into, audience revenues. People pay for a great many things they need, want, or like, but in order for you to get their money, you need to understand what these things are.

Audience surveys are surprisingly low-lift. You can go from idea to live survey in under 12 hours. And even if you want to read some guides (I'll recommend at the end) and develop a detailed distribution strategy, you're looking at maybe 2-3 person days of investment. The only stage where things can get complicated is data analysis, if you have lots of responses. When we worked with Recorder in Romania and 29,000 people answered their questionnaire, we had to get creative with the analysis. But that's a really nice problem to have, and something that's also solvable.

Surveys can help you make business and editorial decisions and they are also an excellent tool to convince stakeholders, external (show a funder through survey data why you are asking money for a specific project) and internal (explain to the newsroom why you want to launch a newsletter product).

2. The free tools available are sufficient 90% of the time

Google Forms, while not perfect and most certainly not good-looking, is a very capable tool for newsrooms to run audience research. Unless you're a magazine focusing on design or some other publication where looks are especially important, Google Forms will do a perfect job, and it's free to use. I have much less experience in the Microsoft ecosystem, but I imagine Microsoft Forms is equally free and competent. The benefit of these tools - in addition to being free - is that they're familiar to your audience, and comfort translates to better response rates.

On the other hand, if you're running a weekly online magazine about design, architecture, or cakes imitating everyday objects, stuff where visuals are very important, you may want to look at Typeform, which looks much better by default and allows for more customization. But it can cost money, and your audience will be less familiar with it, which introduces friction you generally want to avoid. (More on that later.)

There are sophisticated tools like SurveyMonkey or Qualtrics that allow for branching surveys, conditional logic, and other fancy features, but for most newsroom-driven audience surveys, you simply don't need them. (Qurio can do cool automated stuff around content, check it out.)

3. The two basic settings most publishers get wrong at first