Receive Weekly Recap

Pia Frey is a co-founder of Opinary. In this episode, she speaks about how her engagement tool was built from the newsroom, how user-centric thinking can uncover what audience thinks and opinion/polling trends and her plans ahead.

Podcast Transcription

Vahe Arabian: Welcome to the State of Digital Publishing Podcast. State of Digital Publishing is an online publication and community providing resources, perspectives, collaboration, and news for digital media publishing professionals in the media and technology. Our aim is to help industry professionals get back more time to work on what really matters – monetizing and growing real relationships. In Episode 13, I speak with Pia Frey, co-founder of Opinary, an audience insights tool based in Germany and recently expanded to the US. Let’s begin.

Hi, Pia, how are you?

Pia Frey: I’m good, hello.

Vahe Arabian: It’s good to hear you. You’re in Germany at the moment, is that correct?

Pia Frey: I just got back to Berlin. It just became spring here, which totally changes the game of being in Berlin.

Vahe Arabian: Nice, nice. Just for people who don’t know much about you and about Opinary, if you can start off by providing a background?

Pia Frey: Yes. So, I’m Pia and I’m a co-founder of Opinary. I used to be a journalist, which is how Opinary got started. And I live in Berlin and in New York, I go back and forth a lot, and I started Opinary about four years back together with my brother when we discussed and figured out that something needs to be done in the way how content creators, publishers, and also brands have conversations with their users.

And there was a rotation that the vast internet is providing a lot, but not a way for users to share their opinion and content effectively, and then insights or rewarding way. So, we did it.

Vahe Arabian: How did you come up with that insight though? What was the detail behind that?

Pia Frey: So, at the publisher where I was working, Axel Springer, it’s a big European German publisher. I was working on premium products for subscribers. But the actual problem lay not in the products for the subscribers, but in the funnel where you get people to subscribe, which is basically just a loyalty or a loyalty metric from users.

And we were thinking that when you’re building a relationship with a user, with a person, it’s quite hard to do this without being in a conversation, without speaking to each other. I think real, live relationships are not too different in this way from digital relationships. And when you’re not interacting, it’s hard to build a connection.

So, the number that intrigued me was less than 0.5% engagement rate in the comment section on the site. Because the comment section is traditionally the place where conversations with users are happening. But the users who were joining these conversations on the site, at the publisher where I was working, we’re just randomly called the grumpy old men, and we were rather, or the moderators were rather happy that there are not too many grumpy old men because it’s just hard to moderate and not scalable.

So, we were thinking how can you make it rewarding and insightful for users to join the conversation? And to share their opinion in order to get them engaged, in order to build a relationship with them, a closer relationship. And to convert them, and to keep them along the side, and to get them to share stuff, and to read more. And these were the just very rough hypothesis that we had, in engaging users in smart ways pays onto these loyalty metrics.

And we tried it on a very experimental level. And so wow, that it’s something, and because the tools that we built back then. It was just a side project that we did with a friend, but this thing that we built there looks awful from today. And really shitty technology in the first place. But it’s true in the case of an increased use of engagement rate from 0.5% to 18%. And this is what kept us going since then.

Vahe Arabian: That’s a massive jump, and that must’ve made a big difference in terms of new subscribers as well.

Pia Frey: Yeah.

Vahe Arabian: With that in mind, when you speak about comments in that section now, and you speak about Opinary, people might potentially think of it first off as a widget or an additional add-on. You mention, on your website, the fact that it’s not that. It’s a platform. Can you explain in more detail behind why it’s a platform for publishers?

Pia Frey: It’s true. The reason why I don’t identify it as a widget is because I think conversation and interaction is the very base of a relationship with a user. And by working with more than 18 publishers across the UK and the US and with many, many brands who use our technology to drive these conversations, we have a database and a back end that basically helps enable us to get into any conversation on any given topic with any demographic user group across this entire network.

But this is nothing that we use, but what we make available for our partners. So, as a brand, you are not embedding a widget, but you are able to interact with a user group that you like to target across our entire network. And this is what makes Opinary, we call it a decentralized platform, but I learned that decentralized is one of these German English words that are not actually used.

Germans tend to complicate their language, is what I learned. But yeah, it’s a decentralized platform, meaning we are not a destination platform like Facebook. But we are driving conversations across a huge network of publishers and brands.

Vahe Arabian: Does that mean publishers can tap into audiences from other publishing sides as well, through this platform? How does that work?

Pia Frey: They can, yes. And especially publishers brands teams are interested in this. When, as for example, The Economist has a brand client Toyota, and I want to enable Toyota to engage with users and not only to generate some clicks. And we enable them to distribute this interactive brand that micro-content for Toyota across our entire network, and place their polls, basically, into any contextually relevant article that is related to this question that is asked by Toyota.

For example, Toyota might like to know, or is engaging users, with a question “do you think there will be self-driving cars on London streets in five years?”. We would enable The Economist to distribute this poll into any article from our network that is related to future driving or diesel or self-driving cars.

Vahe Arabian: That’s really interesting. And then people will engage and The Economist will see the results and then will they share that with Toyota?

Pia Frey: Exactly, exactly. And these insights that you generate are sometimes extremely revealing about the target groups and audiences a brand aims to reach. So, for example, the marketing team at Toyota might think we want to reach 35 to 45-year-old men from south of England. But what we can actually reveal is that women between 25 and 30 have a much stronger opinion about this topic than the actually targeted user group.

So, it also enables our brand partners and also publishers to find new target groups and new audiences.

Vahe Arabian: That’s really interesting to me. I think there’s a lot of technology behind this, and I like to go into detail about these. But I just wanna take a step back first, just to give a broader picture of what kind of other similar solutions are out there. To my end, at the moment, there’re things like Disqus or even like New York Times recently put an AI chatbot in their comment moderation. What other technology do you think is out there which fits this similar landscape that you guys are trying to engage in?

Pia Frey: Of course, we do have many competitors from many sides. And basically, when it comes to our business model, biggest competitor is Facebook because the brand is deciding to engage an audience on a walled garden on Facebook. Or are we using the distribution model like Opinary, where we engage users? But on the engagement side, and different ways and technologies to get users into conversations, I’d say there are two categories.

One is these actual conversation drivers that are enhancements of the traditional comment section. And yeah, there’s Disqus, for example. There’s a really great project from the Crow called The Coral Project. Talk, for example, is also a smarter comment section driven by questions, asking users questions and letting them answer. There are proprietary solutions from publishers like the New York Times chatbots. And Die Zeit, a German publisher, is doing great work on the comment section.

We decided, even before we got started, that we don’t want users to write comments, just because I never wrote a comment in my life. I really valued the comment section, and I do have strong opinions, but I don’t write comments just because of time and, I don’t know, effort. So, we wanted to make it a one-click solution for users to share their opinion. And there are other providers who do this as well, for example, PlayBuzz or Apester, two Israeli startups, are building tools that also provide polling or voting.

Related:  EP 7 - Publishing As A Side Gig With Jasmine Watts Of Miss Millennia Magazine

There is Raw, for example, which is I think is an Austrian or a Swiss startup. And there are different ways. I think there’s one, there are many different trade-outs to ask. And one is that because we were born in a newsroom and I was facing the newsroom reality every day where I saw many great engagement stuff and tools and article enhancements available, but also very limited resources to use them.

So, we built a technology that enables us, or enables our partners, to automatically distribute and drive these conversations in their articles by an LP-based algorithm that places polls and questions in articles without an editor who has to do anything. Which is, I think, very much in line with the future of any CMS where the dumb work and the copy paste work is not supposed to be done anymore by the valuable time of smart journalists and editors. But the CMS is doing as much as it can.

So, we have our own newsroom that creates polls and questions and drives debates, and a very smart algorithm, which places these conversations on publishers’ sites and on articles. And this makes it scalable and easier for partners to drive these conversations and to focus on the insights part. Because that’s what is relevant. And that’s what I’ve always found very irritating, from a newsgroup perspective, that the users’ opinion is …

You know a lot about users. You know how old they are, you know what they’ve read before, you know where they’re all coming from, etc. But you never know what they’re thinking, and you don’t know when you write an opinionated, moody article about the German rhetoric party, which is read by millions, you don’t know are they reading it because they like it? Or are they reading it because they hate it?

And I think journalists and editors time should be focused on understanding these opinion trends and understanding how their coverage impacts the public’s opinion of their ends by distributing, creating polls ourselves with our own newsroom distributing it automatically enables our partners to fully focus on the insights part, and on the conversion part, and on the impact level of it.

Vahe Arabian: Do you think, though, that journalists themselves find enough time to collect feedback? Or do you think they might rely on people who are focusing on audience engagement within their team, the cross-disciplinary team? What are your thoughts?

Pia Frey: I think it’s in progress. I think it’s in progress. I think there is one problem, and there is one cultural shift going on. In print era, the newsroom was the epitome of wisdom, and they were just telling the world out there how things are. And now having a direct stream of conversation and a direct access to an audience through a website or in distributed content, and being much more forced to show the value of your product when you want to drive user-generated revenues, you have to become user-centric. And this user-centric thinking is something that is more and more spreading across publishers, but far from being there. And showing a newsroom what their audiences think, and how their audience compares to other audience, and how their coverage influences opinion trends in different ways than other publishers do, is a strong driver in this user-centric approach.

But I know that audience development teams that are established across many publishers over the recent years are still having a tough time sometimes to do this fight. When it comes to not the insights part but the loyalty part, I think publishers still have the problem of lack of in-house expertise of real customer relationships management.

And many of them just expected to do great content, do quality journalism, and the users will come, and the subscribers will come. But it’s not that easy. And as a user, you have expectations based on how you are treated as a user for any service you’re using. And building this expertise of how to manage an audience and customers, and how to keep loyalty, is an expertise that is still on a very early stage, let’s say, across publishers.

And we are happy to help them by just getting users into the funnel and contributing to retention, and showing very valuable data about who these users are willing to subscribe for.

Pia Frey: So, I think I didn’t explain it, how and why we are driving subscriptions or sign-ups, right?

Vahe Arabian: Let’s go through that, yeah.

Pia Frey: I think that it’s hard to get users into the funnel, into subscription funnel, through the inventory that is available on site right now, which are ad spaces. Because users know to ignore these ad spaces for a very long time, being very invasive advertising based on that. But now these exact places are used for self-promotional on products, which are easy to ignore from a user perspective.

And we saw that, when you create a receptive moment by asking a user a question that is related to what he is currently reading, you have a momentum that is very valuable for a conversion aspect. And asking, giving a user a reward once he or she shared their opinion, and by saying thank you for your votes and we value our engaged users. And that’s why we give you four weeks free access to our premium section, for example, is very efficient when you want to get users into the funnel.

And by interacting with users across many sites, we know quite well where users have subscribed already. If a user is more intrigued by subscribing to a sports newsletter than to a politics newsletter, and the user has declined an offer to free app downloads or free trial for subscription and wouldn’t ask him or her again. So, this knowledge, across publishers, helps us to sift users through these funnels with very great conversion rates.

That actually surprised me, how well it worked. That’s fun, stumbling like that.

Vahe Arabian: So, let’s go through an example of one that you’ve recently done, or that you’ve recently seen from your publishers. What’s a campaign, what was the conversion, what was the performance like? And, I guess, what did they learn out of the process that made them result in using it again, potentially?

Pia Frey: I would give you one example from a German publisher who just refused to do the funnel, and just sent us the target users with the highest barrier that we have, which is our $20 subscription. And they make, since more than three months, more than 400 subscriptions a month, sometimes up to 80 subscriptions a day. And it’s a general news site. The hotter the topic, and the more, and there is a correlation between how relevant a news or a topic is and how willing the users are to subscribe. Which is nothing new, and the Trump bump has shown it quite a lot.

On another end, for example, one vertical from NBC is driving newsletter subscriptions and Twitter followers. So, it’s a lower barrier, from a user perspective, and they continuously have a click-through rate of more than 10%. More than 10% want to learn more about the newsletter and on the signup page after voting.

And these are the numbers that are just, we are just very proud of.

Vahe Arabian: So, these are very high numbers. For example, even with the NBC case, I see in the article that you might have, how would they, the integral team, design the questions to then drive them to the newsletter. What would be some of the types of questions? How would they approach that?

Pia Frey: So, we are doing a lot for them here. Our editorial team is creating polls and questions for them that match their content and their topic range. They and these polls are connected with cross voting actions that are also set up by us, approved by them. And what they do while integration is just copy paste these readymade questions with conversion call to actions that are created by our team into their article template.

They wouldn’t have to do this, they can also just let our algorithm do the placement. But yeah, it’s quite simple. It’s an approval process, and then we do the work.

Vahe Arabian: So, what kind of questions, question techniques, and polling, do you usually take to drive a question?

Pia Frey: That’s a very interesting question because it differs a lot across markets. And we learned that in the US, people are way more responsive and engaged by questions that directly address them, and ask for a subjective opinion. For example, a question like “do you think that Donald Trump should be impeached?” typically addresses high engagement than the question “should Donald Trump be impeached?”, the objective question.

Whereas in Germany, the should question, the objective question, would drive higher engagement than the personal addressing question. And then there is no difference in engagement across news content and special interests content. Or there is a slight difference because we do see on lifestyle sides or on local publishers often have a higher engagement in general use. But there is a higher engagement with the special interest sites.

And we, on the question side, predictive questions, will question, will Donald Trump be impeached, typically has lower engagement rates than an opinionated question like should Donald Trump be impeached. So, there are similarities, differences, and quite interesting data that we see by asking hundred million users for their opinion every month.

Related:  EP 9 - The State of UK Local Journalism With Marc Reeves

Vahe Arabian: But what’re some of the ways journalists can learn about how they can ask questions to their audience?

Pia Frey: In the very first place, I think it’s a mind shift that is supposed to happen. And I think, I always find it surprising that journalists, and I include myself as a journalist, are extremely good at asking questions. It’s their job, when they’re doing an interview, for example. But somehow, when it comes to the users, I feel very hesitant to speak to the users.

So, first of all, there must be the ambition and the openness to speak with the users. And not to disregard them, or to look down on them, but to treat them at eye level. I think an error of trolling has done some bad there, in this culture and trust between users and newsrooms. But when there is the openness and the willingness to ask users questions, my advice for newsrooms is always to think about when you write an article and when you create a poll for this article.

And users can create their own poll. Think about what is the question that will be discussed by someone who read this article after he read this article. When he is reading your article before his lunch break, and then he goes to lunch, and then he has your story in the back of his head, and then he starts a discussion with someone else. What is this discussion about? And almost any story must have this potential for discussion.

The only type of content or article where we say no, there’s no question to be asked to the user, is when it comes through that someone has died because he was old. Something death related questions. But when it comes to terror, for example, of course, there are discussions happening. And you should be afraid of, and the opinion from the users.

Still, because we don’t do representative polling, we ask users for their opinion on, or we ask, we mirror communities opinion trends that are not representative in itself. The publisher or a museum has to be okay with any potential outcome of the poll. So, when you ask a question like do you think Trump should throw a bomb on North Korea and when your community says yes, and you don’t like this result, you shouldn’t ask the question in the first place, when you fear a shit storm based on a result.

So, that’s one piece of advice that we give our partners, and that we consider when creating questions.

Vahe Arabian: How can they determine a performance prediction, and how can they determine what kind of questions they can ask the audiences that no, let’s not get that sort of results, access that they’re not expecting. Well, not expecting, but they don’t want to portray, or they don’t want to, yeah, portray themselves as a brand.

Pia Frey: It differs from brands and publishers. And when it comes to brand safety, of course, there is a different set of things to consider when asking users a question that is branded by, let’s say, Toyota. Typically, or in general, we see that every community is very different. And there are similarities or general insights that we see as a tendency that Americans respond more to personal questions, etc.

But when it comes to what topics drive engagement, and what are the things you should do and you shouldn’t do, we are hesitant in posing our learnings and data to our partners because there have been so many cases where we saw no, that’s a very unique community, again, that has a very unique and individual way to respond to questions or to polls.

But there are technical things like, for example, when you ask users to join a conversation with the poll in the first 30% of an article, that the conversion rates, the scroll depth, the sharing rate, all these loyalty metrics, tend to be better and higher than when you place a poll at the end of an article. I think that’s one mistake of the comment section that a conversation where users are asked what they think is happening in the very basement of an article. It’s not a too pleasant place there, down there.

And we encourage publishers and brands to make this conversation with the users more in the living room of the story, which also shows the value of how a publisher or how a brand values a user.

Vahe Arabian: The thing is, well, and again, I agree, it’s basically from that, it’s getting it, taking a unique approach to every single poll or question that you guys place in the conversation. But your expertise comes also from the technical placement and the stats around that. Does something that I read the other day, and I probably should do some outside around experiments to publishers are doing today, and someone, the John knight fellowship group that is based in Stanford.

They did a coffee table where they’re trying to engage and do local journalism engagement and they’re asking people questions and doing the concierge service, and one of the findings from that was that people couldn’t connect to what they’re trying to do to the call to action, that they’re trying to present to them. So, the thing that I see that is, potentially you might have a poll there, but how do you make sure that you can connect the conversation to the call to action? Like subscriptions, or the objective that publishers are trying to achieve? How can you do that naturally?

Pia Frey: Do you mean how this workflow works? How the user journey is working or how we are setting this up for a partner?

Vahe Arabian: I think it would be good to answer both. So, how can publishers consult the user first so that it eventually connects to the conversion and how can you benefit also the user?

Pia Frey: Okay, so let’s imagine you’re a publisher, and you have certain goals. Like, for example, you want to generate leads, and you want to generate subscriptions. And you have a portfolio of newsletters that do have a value in themselves but have a KPI of driving sign-ups and generating leads for you.

So, what we would do is tagging these newsletters on categories like sports, politics, Brexit, lifestyle. And the poll that goes out to your site is also tagged with sports, politics, Brexit. And any question on Brexit will be connected with a post lodging call to action telling the user thank you for your vote, you obviously have a strong opinion on what’s happening in the course of Brexit. And you might like to get access to our fresh new opinion or analysis of Brexit in your inbox every day.

So, we give you free access to our Brexit newsletter, for example. And this call to action and this creative that converts a user after he or she votes is set up by us, approved by you, and then just contextually matches any hold that goes out on your site. And when it comes to the funnel, which user gets which call to action, we know a lot from a user by a user’s history.

So, we know, for example, that user is already subscribed to this Brexit newsletter. So, he wouldn’t get a call to action on this Brexit newsletter again. But is ready for the next stage, like, in your case, the subscription. But if the user refuses to subscribe for three times, and says no, I don’t want to get this present from you, he wouldn’t get this call to action again.

So, it’s a combination of contextual fitting and user history that drives these calls to action. That’s how it has been working so far. What we are coming up with next is extending the conversation chatbot like mode post voting where the user gets access to more insights into his opinion profile, which is where we hope to drive even higher conversion rates.

So, for example, after voting, you would get alliancing, and interesting, you have a very different view than other 29-year-olds. And then, as I would be intrigued and want to learn more, where I can enter my email address to get the full opinion profile that relates me and my voting history to my demographic peers like other people from Berlin, other people, other 29 year olds, other people of my income class, other people from my age, etc.

And we know this demographic information also by post voting call to action where users are asked to thank you for your vote and do you want to see how other 29-year-olds think about this? Do you wanna know how old versus young people think about this? Tell us your agenda. Do you wanna see how people from Munich versus people from Hamburg are thinking about this as your location? etc.

So, and by extending this conversation, we generate these first party, in sense, data, that we then in the future use to incentivize users to share their email address. And this email address will be, which is very relevant in the course of GVPR, not stored on our side but on our partner, in our partners GMP.

Related:  EP 10 - The State of Food Media with Paul Barron

Vahe Arabian: That’s exciting to hear. Is that something, is that the trend that you see moving forward, is that something that that’s the trend that’s going to be moving forward with engagement on the site, having these user profiles there? And are there gonna be any other content engagement techniques and polling techniques that you think might fit into the process?

Pia Frey: That’s a good question. I think the incentive of giving the user information about him or herself is an untapped field that has a huge potential. Because everybody is very interested in many things but most interested in themselves. And using this to unlock these insights that you have as a content platform about a user with a barrier, with a key, to share your email address or to sign up for something that is not actively tackled yet. It’s quite new.

Also, in the way of usability, how this conversation and how this follow up the conversation is done with a chatbot like mode, is not done too many times yet. I know this New York Times chatbot, which’s driving the comment section but so far chats and messenger mode was something that existed in standalone apps like this great calls app, for example. There’s a German app, Raisey, which does the same. But driving conversations in content is much more natural and less isolated to me. That’s a direction I think that we’ll spread, but we don’t see too many other examples out there yet.

But that’s our mission, to do new things. But not the mission itself, but we like to do and try new things. And even the way how our tools look like always provoke the first comment oh, what is this, I hadn’t seen this. I used to think that’s a bad thing, but I think when it comes to UX, or to usability, reinforcing confusion to a certain degree is driving curiosity and the high engagement rates that we have.

Vahe Arabian: But that’s your training in status quo, is that what people want to get used to it?

Pia Frey: Yeah, so I’m passionate about this because we had many discussions about our tools where you people said I don’t understand it in the first glance when I look at it, and because I hadn’t seen it yet. But our engagement rates prove that people do engage with it. And then, with this first momentum moment of oh, what is this, I feel backed by, for example, by the memory of how the iPhone got released. Where there was real confusion, there’s a phone with no keys and just one button? How are you supposed to use it? But it was a huge change maker in the way our devices look like.

And so, yeah. We dare to confuse people.

Vahe Arabian: I’d say challenge people, for sure.

Pia Frey: Yes! Exactly, exactly. Exactly. Yeah.

Vahe Arabian: Yeah. 100%. I read that you recently received venture capital funding, congratulations on that.

Pia Frey: Thank you.

Vahe Arabian: So, with that, what is your initiative for VC, your primary initiative for this year, and for beyond?

Pia Frey: So, for this next eighteen months, we are very much focused on strengthening our presence in the US and building the team there. And also we are in this very typical stage from where an early-stage startup, where many passionate juniors are working hard to get things done, actually getting things done, and to a structure with more experienced and specialized set of people to scale our growth.

And that’s what happening right now, we’re hiring at all ends. In sales, on the US side. And our partnership management, we’ve been hiring on the editorial side more than on the product side because getting people on the ground, especially in the US, is what drives our growth goals for this next eighteen months where we aim to get from the hundred users to a ten times larger user base.

On the product side, it’s very much about the challenge to extend the conversation. And we’ve been proving quite well that we’re good at starting conversations with users, and to converting these conversations into loyalty. But we sit on such a valuable and insightful database, knowing what all these people across topics, across demographic groups, think, but don’t make it accessible to users yet. And what I just described that this chatbot like extension of the conversation is one direction where we aim to improve our conversion and by making these insights available to users.

When we got started four years back, as I said, on a very experimental level, one point we were thinking, “what is our mission and what are we doing here?”. And these mission statements always have a touch off bullshit, especially for our German minds, but we don’t like them. Germans don’t like these big-thinking statements so much, but still. We had to define what are we doing here?

And what we came up with, or what we thought out, yeah, we went to all the users in debates and empowered them to join debates. Orienting empowerment, orientation and empowerment stick with us, and we want to become even better on the orientation part, and to make these new set of sentiment data accessible. Not only to our partners but also to users themselves.

Vahe Arabian: That’s a big challenge. It’s a big area. Yeah, so it’s gonna definitely take about eighteen months, I think, and I wish you the utmost success on this.

Pia Frey: I can’t plan ahead for too long, to be honest. My brother, so I have two co-founders. My brother, who is managing our investors and finance. He is taking the ownership and thinking five years ahead. While I, for me, one and a half years are enough. And my other co-founder, Maxi, is managing the brand side, and I think he’s more my end, thinking one and a half years ahead.

But, I mean, basically, it’s just about what fascinates me about Opinary again and again and again, every day is just how this very basic idea, how you could improve engagement outside these Facebooks of the world, just on your own site. How can you do this? And this basic idea of visualizing debates, making your one-click thing for users to share their opinion and to convert it from there, which existed on paper, very poorly drawn by me. How, and this passionately discussed with my brother. How this kept growing and keeps growing, and becomes bigger and bigger and bigger.

And I think I have a bit of a traditionalist’s mindset of thinking of the next step that is needed. And the next big step is to actually create a real presence and drive our value on the west side. That’s what my brain is busy with, and how the team has to look like for that. We’re hiring.

Vahe Arabian: Good.

Pia Frey: And my brother’s thinking about the five years stuff.

Vahe Arabian: That makes sense, there’s a big market hitting the US always.

Pia Frey: Oh, yeah.

Vahe Arabian: Pia, just to wrap things up, I just want to come back to pre-progression of advice as well. So, for people who aren’t on the tech side and the general side, what advice would you give them to come to the position that you were in when you were working for a publisher, and in the position where you are now?

Pia Frey: I remember when I quit my job, many people said oh, I would never dare to do this, leaving a safe job. And I think nobody has to be afraid when there’s a good idea. Just go after it. It will find its way. There is no safety in anything. So, it’s boring, but I just say, just do things and just don’t get affected by too many routines and too much comfort.

I think curiosity is a great driver for everything. And curiosity is typically stronger than fear. And just going after, focusing on what you’re curious about, and what you want to learn and find out is a better driver for anything than thinking about what am I, what are the risks, and what are the things you can be afraid of.

Vahe Arabian: I understand. And for journalists as well, for people who want to succeed in journalism and in the editorial, in publications. What’s some of your advice on that?

Pia Frey: Oh, even more, curiosity for journalism, I think.

Vahe Arabian: Curiosity applies to both the sides, yeah.

Pia Frey: No, and for journalists, let’s say, journalism and content, in the end, is a product for users. And when you want to succeed with it, put your users into focus.

Vahe Arabian: For a mental advice, thanks so much.

Pia Frey: Think about your users, and talk to them. They’re not dangerous.

Vahe Arabian: Thanks for listening to episode 13 of the State of Digital Publishing podcast. Be sure to listen to past and upcoming episodes and all major podcast networks – iTunes, Spotify, etc. You can also join up on social media on our Slack group. And for exclusive news and insights, look for to visit our website and our subscription on stateofdigitalpublishing.com. Until next time!