I talked to 50 journalists in 2 months, here is what I have learned
When approaching a market from the outside, a lot of effort has to go into understanding the relevant players and people of the field. This takes time and is a steep effort. But it is the only way to understand how your customers think, what their struggles and ambitions are. Only once you know that, you can a) build solutions that are tailored to their needs (expressed and latent) and b) appeal emotionally to the inner drivers of the target audience — which is crucial in the later sales process. To gain exactly such an understanding, user interviews are essential. User interviews, as semi-structured ways to gather detailed information on the executed processes and thought processes around the problem that you intend to solve. We are doing that since the beginning of our journey — and in the last two months, we have intensified these efforts again, in the development process of our new product, Varia Research.
What follows are few key hypotheses that I have learned and in part validated, while speaking with 50 journalists in 2 months.
TL;DR
- The scope of journalism and journalistic processes is vast. While this is no news per se, it has strong implications for product strategy in the field
- There are no established tools for journalistic research. Journalists, as much as their processes, are individualists, lovers of workarounds
- Content Management Systems only play a marginal role in content production, whether publishers like it or not. After all, they are not called CPSs, but CMSs
- Publishers are hesitant with top-down software distribution into their newsrooms. As they are aware of the individualistic nature of journalistic work — and as their tool & software focus is on distribution and top-line impact
- Journalistic research efforts remain underleveraged. While a journalist spends the majority of his time on research, the minority of this effort will manage to make it into the article — and thereby stay away from the bin
- Journalists are experts in not paying for journalist content. Paywall? play ball!
Read on, for more details on each of these hypotheses…
Journalists and journalism are very individualistic
To some extent this is based on the nature of their work — journalism knows many forms and hence draws vastly different people and working habits to the field. Is a sports journalist covering an NHL game for an online site going to use the same tools, like a science journalist writing about the shape of our galaxy, for a science magazine? No. Are these two going to be the same personalities? Neither. They might have a shared customer success goal — to write great articles, but their journeys towards that customer success are going to look quite different.
This vast scope and variety of journalism brings both challenges and opportunities along. For one, you will most likely not build a product that helps all types of journalists. You have to focus — on the type of journalism you want to support with your solution — and also on the aspect of the journalistic process that you want to hone into. In most cases the journalistic process goes from topic discovery — to content distribution. That covers a lot of ground and every process step in between offers further openings for different approaches to it.
The scope of “who” writes content is also broad. Clearly, we are not only talking about journalists. Many journalists go under the job title of “reporter”, “correspondent” or “writer” anyhow. On top of that there are a lot of editors that are writers too and hence journalists at the same time. Especially in science journalism, but also more and more in other fields, external experts are contributing journalistic work. The Conversation, for example, is an open not-for-profit news network, where researchers and scientific experts contribute most of the articles. You can also have even more content from experts — or non-experts. Go to Medium (where you probably read this) or Substack and you will find content that grew out of a journalistic process — but most likely was not written by a journalist. The internet has taken away the content distribution monopoly from newspapers — and in that process, taken away the content creation monopoly from newsroom journalists. I think therefore I create, they say (watch this to get pumped). This applies to video content as much as to text.
There are no truly established journalistic research processes
Largely driven by the diverse nature of journalistic work, as pointed out above, there are few widely established processes. This applies to the whole process, but also to the aspect of journalistic research, where we want to focus on. This makes it a challenge to develop solutions that fit to a “dominant process”. Rather, the goal should be to find common underlying struggles — that are shared, even if the accrued, surrounding processes are different.
When it comes to actual processes and guidelines, there is one dominant document — that seems to serve as basis for many further, internal guidelines. The UNESCO has published a large guideline on story-based inquiry few years back, that appears to inform or inspire many other recommendations for investigative journalists — but also for broader newsrooms. How do we work with confidential sources? At what point is information verified? What does it take to make interview questions meaningful? Many of those and similar questions are addressed in the UNESCO guideline. Worth a read, especially if you are a journalist (and haven’t already). Similarly, the Al Jazeera Media Institute has just published an Investigative Journalism Handbook as well — in case you want a different perspective on the matter.
There are no dominant journalistic research tools
Given the breath of journalistic work, processes and contributors, the scope of tools used, is similarly broad. How do you search information? How do you store information? How do you conduct interviews? How do you draft and write stories? How do you review your content? How do you distribute your content? How do you archive your articles — and why? It. Depends.
The most common tools used, are probably web browsers and phones. Second in line are email programs and search engines. After that, the field of tools grows very broad, very quickly. While for many of the sub-processes and sub-tasks, specialist tools would exist, many such solutions are struggling to find widespread adoption. For example, browser tab organization tools or note taking apps — which as an outsider, I was expecting to be heavily used — are not widespread after all. This has several reasons; in many cases those specialty tools are overkilling a small aspect of the process — while not fitting in nicely in the overarching journalistic process. Also; journalists tend to stick to their routine, once they have found one. They might know very well that their aggregation of workarounds is not ideal, but it’s their habit — and habits are a strong force. Here’s a quote of a journalist that I talked to, and that I have heard in slight variations many times:
“I know my system is chaotic, I could really be better organized. But I love it, it’s my chaos.”
When strong habits are formed, great UX will be key to bring anything new into that accustomed comfort zone.
In the data journalism realm, the situation is a bit different. Driven by dominant way of interacting with data, and tools built on top of that ecosystem, the landscape is much more harmonized. Show a data journalist a Jupyter Notebook & a Tableau/Datawrapper interface and she will be smiling, likely. Also the re-launched Google Data/Journalist Studio seems to be greeted well by the community.
The not so dominant role of CMSs
Hold on a minute, what about Content Management Systems (CMS), you might say — aren’t they a critical part of every journalist’s workflow, did you not forget them in the above paragraph? Well, no. CMS are important in media — but not really in content production and not to all journalists. If at all, journalists interact with a CMS at the end of their process, either to write stories straight into a CMS, or to copy and paste pre-written stories there. In many instances the CMS submission is done by an editor — in which cases the journalists has no touch point with any CMS. This is particularly true for most freelance journalists.
So where do they write, if not in the CMS (where many publishers would in fact want them to write)? Microsoft Word. It was clear that journalists are fans of the written word, but they are also fans of Redmond’s Word (pardon my attempt at humor). MS Word seems to be not only the go-to tool of many journalists, for drafting and writing stories, but also for many other workarounds, like i.e. collecting research information.
Few tools are pushed onto the newsroom workforce top-down
Freelance journalists have to take care of their setup themselves. Choice of tools and also related costs lay in their own hands. The situation for employed journalists is less different than one would assume. Due to the described individualistic nature of journalists, their mindset and their work, news corporates are hesitant with top-down software distribution. CMS yes, since content cannot be managed and distributed through a multitude of tools. But it does not go much further than that. Yes, there are classical communication tools and office related workbench tools around — but especially the research and content creation process remains to large extent in the hands of the individual journalist. Publishers don’t want to interfere with individual workflows, so it seems.
A classic example for this is transcription (speech to text, used mostly for interview transcription) software. At few companies there is a corporate choice for e.g. Trint, Otter, or HappyScribe (to name a few big players). But even those tools are far from a level that all journalists would trust their output and hence fully engage with them. This makes corporates hesitate with top-down adaption and leads to many newsroom journalists getting a transcription software subscription by themselves, to then claim the expenses via the employer.
Publishers’ software and automation focus is elsewhere
These days you can attend a conference with the words journalism, automation, and/or AI in it’s title every week. 2020 has brought you the luxury of not even having to travel anywhere for that. The lowered entry barriers for hosting a conference 2020-style (“Do you have a zoom account?” “Yes.” “Ok, go.”) led to a sharp increase of the supply. But indeed, a lot is going on in the sector, a sort of second wave of digitization is on its way, where not only first experimentation with external services is done, but also significant internal competences are being built up.
However, the focus is not on journalistic research. If at all, then regarding ways to better leverage archives — since these are in their current states mostly bad enough (technical access, searchability, incompleteness, broader UX) to not being properly (or at all) used by a majority of journalists. The automation aspects seem to center around NLG tasks (mostly done via logic-based rules — which however sells less than “AI” in the conference title) for short texts, on recommendation engines, or content curations. By far the easiest ways to open a news company’s door with a software product, in the past & current years, was when your solution was directly impacting top-line figures. Subscription prediction, dynamic paywalls, churn analysis & reduction, or ad-revenue related optimization.
Journalistic research efforts are underleveraged
Journalists spend about 60–75% of their time researching a topic. In that process they not only gather a lot of knowledge themselves, they also collect a large number of relevant documents and information. Be it interviews, online articles, research studies, or own notes. The research that is accumulated during the creation of a feature-sized article can be immense. Only a fraction of it will be used and referenced to in the final piece. And the rest? Two models seem standard: either the rest remains somewhere locally on the laptop of a busy journalist — or gets thrown out a week after the article was published.
This is an immense case of wasted efforts. Being a former business consultant, I know what wasted efforts are, trust me. But it must not be that way. While publishers are “already” waking up to the idea to leverage their archives better, mostly by making them indeed useful to journalists, similar initiatives could be started to better leverage all that research work. This is one of the main goals of Varia Research; to provide a better research organization — so that all these efforts are not lost, even if they don’t all make it into the final edition of the article. The only journalists that already document and archive almost all their research work today, seem to be those working on large investigative cases, in collaborative teams.
What if every journalist could re-visit his/her research later, at any point, for a follow-up piece? or to squeeze out another piece, out of an article that turned into an evergreen story? Or better yet, share the research with a colleague — who would actually understand it, as the research dossier would have an intuitive and clear structure..?
Journalists are experts in not paying for journalist content
This is one of the more surprising findings of our latest user interview efforts. Here’s the problem: journalists rely for their work a lot on secondary research. Interviews and reaching out to people are crucial, but so is seeing what other people already figured out and wrote about a certain topic, or how specifically the benchmark publishers are covering it. Also, to keep track of their topics, journalists spend a lot of time on news sites — which are, with the revival of subscriptions, getting more and more tight, when it comes to free content. So, journalists are running into paywalls rather regularly, behind which, content is hidden that they would need for their research.
For freelance journalists, that is a real problem — of varying significance, depending on the type of stories produced. That freelancers have that issue is not the surprise here, the surprise is that also many corporate journalists have the same issue. Only a minority of publishers provide large scale access to all relevant news sources, for all their journalists. In some cases, exactly such access — to databases like LexisNexis or Pressreader has been reduced, as cost cutting measure. The result is that journalists mostly search for stories and other articles on their topic, in the few subscriptions they have themselves, or that their publisher provides for them. What is few? We are talking about a <10 number of subscriptions in most cases.
So the paywall problem is real — and can impact the quality of research conducted (by narrowing it down), as well as the content subsequently produced. But journalists, by nature, are good at figuring things out, hence they know all the tools of the trade to circumvent paywalls — and not pay for content. Such as: VPN, use a crawler/parser, have a list of friends and their subscriptions at hand, call the author of the piece you need — and ask to send it over, or try to manage several trial subscriptions in parallel. My hunch is; this situation is not ideal.
Wrapping things up, let me leave you with a quote from one of the journalists I spoke with. I think this judgement is rather brilliant:
“The journalistic community is large, unorganized and poor. Two of these attributes speak for your product, the other will make it hard to sell.”
For more insights, feedback & discussion, do not hesitate to reach out, this is supposed to be a summary. Thank you for reading!
This article was written and researched using Varia Research
