Meeting user needs pays: How reader-centric content drives revenue

We spoke with Ariel Zirulnick to explore why newsrooms are struggling to meet audience, revenue, and impact goals—and what steps they need to take to turn things around.

As journalists, we like to believe we understand our audiences. We’re often confident that we know what people should care about—sometimes more than they do themselves. Yet, many newsrooms are struggling to adapt to shifting audience habits.

In response, some news organizations are experimenting with “user needs models” to help figure out what audiences actually want and build strategies to meet their demands. We spoke to Ariel Zirulnick, who ran the Membership Puzzle Project and developed a user needs framework at LAist, where she served as Director of News Experimentation until recently.

Ioana Epure: What are we really talking about when we say “user needs”?

Ariel Zirulnick: In most newsrooms the journalism is organized and labeled in a couple different ways: maybe you’d have “beats”, “story types”, and “stages of the funnel” [for audience revenues and conversions]. They’re all useful in some way – but all of these are rooted in what the newsroom needs. 

The user needs model creates another taxonomy for a newsroom to work within that is rooted in what the audience members need. 

“User needs” answers the “why” behind our information-seeking behavior. When people choose to come to a news organization, they are trying to accomplish something. And the “user needs” give labels to what they might be trying to accomplish. Or if they stumble across it, what keeps them there? What thing are you doing for them that is encouraging them to continue to engage? 

You can also think of it as one of many segmentation models a newsroom can use to ground its work in what audience members may be looking for. Personas, archetypes, and jobs to be done are also strategies for segmentation and serving our audiences. 

User needs are a distillation of the things people are trying to accomplish and their motivations when engaging with your product. They’re not inherently a journalism thing, but they’ve been successfully applied to journalism as an editorial strategy over the last couple of years. The user needs model developed by Dmitry Shishkin is really what popularized it in journalism. 

Why are these models more relevant now than a decade ago?

Well, we are consistently falling short of audience, revenue, and impact goals, so we need to change something about how we work.

The only way to grow audience revenue - whether it is donations, membership, or subscription - is to increase the number of people who are coming to us, to increase how often they are coming, and to increase our value to them so that they are willing to pay to support our work. 

The user needs model gives us a framework for understanding what types of stories make that more likely to happen. 

How can these frameworks help boost audience revenue?

It can help you devote reporting resources to the stories most likely to reach and resonate. It can help you move people down the funnel by understanding what types of stories drive repeat visitors. It can help you write more resonant membership or subscription appeals – giving you language that isn’t focused on the importance of journalism.

I also think it can help you understand whether your stories actually have an impact on people, through surveys. The results of those surveys can be leveraged to raise philanthropic support. Philanthropic funders want to know that your journalism is having an impact. Philanthropic funding is usually very focused on driving impact in the community that you serve. 

But the user needs model only becomes meaningful for your organization and on the revenue side if it trickles down all the way to the beginning of the editorial process. The way you approach stories needs to change to see these benefits – to use the BBC user needs model, you need to do fewer “update me” stories and more “educate me” stories, for example. 

wAiT, iF wE wERe ONly tO wRItE wHaT AUdiEncEs wAnT, wE ARe jUsT bEArS iN A cIRcUS dOInG trICkS

Those arguments always bother me – they assume that what audience members want from us is cat videos or whatever. 

If you do research in your market, I think you’ll find that people want more than distractions, but they don’t know that they can expect journalism that is useful or meaningful to them because they’ve experienced so little of that in the past. Additionally, the need for that light, distracting content is often driven by a sense of disillusionment and disengagement, by alienation from media and the political system. 

Maybe producing more journalism that meets people’s variable needs can begin to chip away at that disillusionment with media by showing people that there is value in reengaging with the information ecosystem.

I want that for my newsroom. Where do I start?

You have two pathways to choose from: do I want to develop my own framework based on qualitative research that we do in our market, or do I want to start with a model from elsewhere and try and layer it over our work? 

Doing your own research means that it will speak much more directly to what your organization is trying to do. And when you actually get to the implementation stage, it will probably feel much more natural. That’s what we did at LAist. But it takes a lot longer, right? You have to do all the research first. 

But if you choose to bring in a model from elsewhere, you would probably start by holding that framework against a certain time block of stories and retroactively labeling those stories. (Most people who import a model from elsewhere use the user needs model developed by Dmitry Shishkin for the BBC. I’ve had some people tell me they’ve used the LAist model I developed as well.)

If I use the LAist model, this is a “discover” story, this is a “navigate” story, this is a “change” story. And then, you build a dashboard and start to look at the data. You'd look at things like: what percentage of the stories you do in a month correlate to each of these user needs? 

And then you might start to look at the performance data, like what percentage of traffic in a month correlates to each user need, or how percentage of traffic by user need shifts as you move further down the funnel. At LAist, one thing we looked at was referral sources so we could be smarter about how we deployed our audience team. What Modes did best on social? Let's prioritize promoting those types of stories on social. 

But if you want more of your journalism to truly meet people’s needs, eventually, you’ll have to  train the journalists on this new way of approaching stories. 

It can help build buy-in if you can point to examples of stories you’ve already done that map to the user needs.  When I was applying the Modes tags retroactively to stories to study the data early on, I found really strong examples of each of the Modes in our past coverage. It’s really valuable if you can  show people that this isn’t an entirely new thing. We’re doing some of this already, we  just need to do it more consistently. It needs to become the primary way that we commission stories, not a thing that we sort of luck into every once in a while. 

How can smaller outlets with limited resources use this approach?

I think you can go a really long way by just looking at trends in your content and labelling pieces. I helped a startup do an editorial audit a few years back, and we bucketed their stories in a variety of different ways. 

We bucketed them by story types; we bucketed them by geography: is it a neighborhood-level story, is this a city-level story, is this a regional story? 

They were a very “accountability journalism” focused organization, so we also identified the types of accountability that they wanted to achieve with their coverage. 

That act of looking at your content and thinking about “what are the different taxonomies that make sense for us” is a really good way to start bucketing your content and then looking at how that content performs. 

It depends on how frequently you're publishing but - say you're publishing a few stories a week - maybe you would look at two months of data and try to spot patterns in what has done really well and trends in what has done really poorly.  Think about what those high-performing stories have in common. What needs do they fulfill for readers? Maybe try holding one of the user needs models against those high and low-performing stories and see what user needs they correlate with. 

Now that you’ve identified a couple of types of stories that seem to correlate with strong performance, you can do some audience research to understand why those types of stories do well. 

So the end product of a “user needs models” is a mindset?

Well, that's when you can start to see the impact – when people start to really think about commissioning and writing stories differently. And also you start to see a real shared language across the organization. 

At LAist something that we hadn't even planned for, that ended up being really powerful, is that the framework could be used across the entire organization. The marketing team used it to design a marketing campaign. The membership team used it to overhaul membership appeals. The events team used it as a way to structure their events work.

And so it started out as an editorial strategy, but it ended up being a source of organizational alignment, to the point where people started using the Modes to make jokes about their personal lives. And that's when you knew it really stuck.