I’ve noticed several new blog posts on journalism and its future over the last couple of months. Couple this with listening to This Week In Google regularly along with following Jeff Jarvis’s blog, and the question of what journalism is going to evolve into (and the journalism that makes it to several years from now will have changed dramatically from how they operate today) has been on my mind of late. The more I consider it, the less I think journalism’s ultimate destination will be recognizable as what we have available today.
First and foremost, any attempt to predict what journalism will look like years needs to accurately capture what journalism will intend to be all those years out. For the sake of this discussion (and because there’s no reason to assume otherwise), let’s assume that journalistic enterprises are still trying to create a populace that’s informed about the issues impacting their community. That means that we can limit our focus and predictions to methods for achieving that goal. In my opinion, the biggest mistake that people make in trying to guess how journalism will work in the future is assuming that future journalism will look like a modified version of journalism today. The more try to think about how journalism will work in the future, the less convinced I become that “journalist” will even be a standalone profession.
New Output
One of the things that struck me is that the ideal output for local journalism isn’t the article and/or report that we’re used to. Take for example Jeff Jarvis’s article about journalism as a service, using Hurricane Sandy as an example. Jarvis lists things that would have been great to know, and how that would have been right in the wheelhouse focused, local journalism trying to serve a well-defined geographic community (for those of you who are too lazy to read his blog post, which you should, he was saying that he wanted to know exactly what streets were closed, exactly what areas were without power, and exactly where the power company said that it was working). The blog post seems to assume that the primary means of delivering this information would still be as an article. To his credit, he did mention that an annotated map would also be good, but if you read his post, he keeps going to back to a “report” that has up-to-date lists.
The thought that sticks out to me is that there’s very little reason for any of that data to ever be in article form. Road closures would best be served as reports to GPS services like Google Maps, Waze, Apple Maps, etc. Why? Because the only reason you would want to know them is so you can figure out how to get around town. But GPS programs already know how to dynamically adjust directions, so all that’s missing is the signal that a road is closed. Limiting that data to lists, or even just an interactive map on a website means users have to go to the website to find this information and then manually plot out a route themselves. That’s not intuitive at all – what do you do when you need to figure out how to get somewhere? Open up your GPS. That’s where road closure information would be most impactful and useful – and that should be the first, primary, and really only place you’re pushing it. Now the information people need is in the service that does the complicated task they’re trying to solve for them, which is (not coincidentally) probably the first place they’d think to turn to for this type of problem.
As for things like power outages and where the power company is working, an article still isn’t the best vehicle for delivering this information. Here the annotated map Jarvis mentions would be the best tool, assuming that the map tool has the ability to push data to the local power company (e.g. someone reporting an outage or confirming the power just came back on) and that the power company has the ability to push updates to it as well (e.g. there’s a truck at this location doing this work, along with a time estimate when feasible). Jarvis’s blog post also brought up the issue that the comments were often far-more useful than the article. That’s hardly surprising – things like disaster clean-up are constantly developing and the situation is (in theory) changing regularly. That means any contextual data needs to be in a format that reflects the temporal nature of the information being conveyed. We don’t think of articles as being temporal in nature, but we have no problem viewing comments that way. Combine that with some logic that “expires” comments that are no longer relevant (e.g. comments about an outage in an area lose visibility after that power is restored), and you have a very workable reference for the electricity status for an area.
This also helps resolve the issue of following up on the user-generated reports. You have people on the scene reporting this type of stuff, they’re the best source of follow-up. GPS applications like Waze already handle this by letting users flag data on their maps as inaccurate, and can deduce when closed roads re-open by noting when drivers start driving down the “closed” road. You could achieve the same thing by pushing a notification to people who live on/near the roads marked as closed saying that it appears open and asking them to confirm that. Same with power outages – they’re getting reported by people that live and work in the area, odds are most of them are going to notice instantly because the lights are going to randomly come on – no need to pay someone to drive out there and keep double-checking these claims – the people who reported it all in the first place are likely to do that for you already.
You can do the same thing with box scores for local sports (think the area high school team). You don’t need to have the school call in the results to the local newspaper, the fans can post the main points from their smartphones. For the other details, coaches/statisticians can fill out a webform, or (because it’s only a matter of time before things are done this way anyhow) have whatever app they use for stats be configured to upload the data to the appropriate server endpoint when they close out the game. College/pro games would be even easier because television gives you larger audiences to use for outsourcing this type of work.
There are probably plenty of other examples of how you can use people who are already in a geographic area of interest to give you the most useful raw information from a scenario. For the sake of discussion, let’s call this API-based journalism. The output varies based on the nature of the data being conveyed, but the data is all generated by the individuals who are already “on the scene” with no real need to dispatch a reporter.
Subscriptions aren’t going to work
Another major change for journalism is probably going to be in the compensation models. Right now, there are 2 major components to paying for journalism – subscriptions and advertising. Most media outlets are focusing on the former, which is probably going to prove to be a mistake. Let’s go back to our premise for the purpose of journalism – creating a populace that’s informed about the issues impacting their community. Subscription-based business models undermine that goal by creating a subset of a populace that’s informed about the issues impacting their community – specifically the subset that’s willing to pay. I understand that journalistic enterprises need money to keep running, but nobody’s going to make money with paywalls, not when you have news articles online and available for free from major publishers like CNN, the BBC, etc.
The closest I think you’re ever going to see journalism keeping a subscription model is probably going to be Patreon campaigns. For those of you who’ve never seen anything supported by Patreon (most of my experience has been through podcasts), basically you commit to pay some amount (you pick the amount) every time the organization you’re supporting produces output. For podcasts, that means you pay $X per episode released. From a journalism standpoint, that could be $X per article (assuming an article is the appropriate medium for whatever it is you’re doing).
While I’m not completely certain of what the future of the journalism business is going to look like, I think predicting that subscriptions aren’t going to be the bulk of the business model is a safe bet, especially in instances where the data isn’t being generated by a reporter directly (because you’re going to have a hard time convincing people to pay a subscription for content they’re generating) or because the work isn’t showing up on your site (think back to discussion about the road closure examples – how do you convince a subscriber base that you’re responsible for the improved maps after a natural disaster?).
Will advertising work?
Advertising seems to be the funding model that has the best chance of succeeding, but also seems to be the least desirable to a lot of news outlets, which I find confusing. This is the model that lets you keep your site free, lowering the barrier to entry for people to get informed about the issues impacting their community (the purpose of journalism, remember?). The good news is, ad-supported content is pretty much considered a norm for most people, and it’s a major funding source for a lot of existing media outlets, so letting it be the primary economic plan for news wouldn’t be that big of a stretch.
The biggest problem with advertising is going to be the fact that it’s not an option when you’re publishing to external services (see back to our map example again). The good news is that those services have enough utility that they can probably thrive on a Patreon model (particularly if the initial developer can self-host at first). You could also partner with local industries benefitting from these utilities (e.g. a web app that lets you report power outages being supported by the power company since it makes it easier for them to get notified about and track outages, and get confirmation that the problems are fixed).
Another thing to keep in mind with advertising is how where you publish your content relates to the advertising network behind it. Advertising and marketing used to be batch-and-blast, just because there was no other option. Thanks to technological advances and the ability to track behaviors on a per-user level we’re now seeing ads that are much more suited to us and our behaviors online and in our email (I should probably note here that I work for an email marketing company that uses – and encourages – such targeted marketing). As a result, businesses that focus on advertising aren’t the same businesses putting out journalistic content, a trend noted by Ben Thompson in his Stratechery post.
That said, I wouldn’t expect publishing anything on Facebook, YouTube, or any other site that both optimizes ads and hosts content to be a sustainable business model. I don’t know a whole lot about YouTube’s compensation model, but the little bit of hearsay I’ve heard referenced indicate that the revenue sharing arrangement does not make “posting videos on YouTube” an easy career. Facebook already limits the visibility of posts made by organizations, and then charges the publisher to increase their reach. The companies behind those mediums own the network, the infrastructure, and advertising that gets attached to the content. And thanks to the free hosting and sharing capabilities of those mediums, there’s plenty of hobbyists uploading things they do without feeling any pressure to make money that those platforms don’t need journalism content. In other words, don’t expect the revenue sharing on these platforms to improve any time soon.
Funding models
Originally I was pretty sure advertising was going to be the funding model to go with, but the more I thought about it the less on board with that idea I was. I’m beginning to think that successful funding models for journalism will look more like the funding models of freemium apps and services – a small percentage of the consumer base is the source of an overwhelming percentage of the revenue. Patreon-supported journalism could work like this too. Either way, successful journalistic funding is going to have the artifacts freely available to everyone, while a small subset of the consumer base is paying, either as patrons or for premium content.
Here’s a potential example of where this could go. Your regular, day-to-day news (sports scores, basic recap of city council meetings, store openings/closings, etc.) is free for all. Think the sort of general information I’ve discussed earlier in this post, where just about anyone can post for other people to access. This is useful information, but it isn’t the type of things we’re thinking about when we consider the value professional journalism can bring. For that, we typically think of things like investigative reports, or in-depth (heavy depth, as in everything a person could want to know the topic) coverage of a story.
The problem with this is that these types of deliverables are ready for distribution so infrequently that you can’t reliably predict when the next one of these reports is ready for consumption. The time, effort, and resources that go into these reports far outweighs that going into the day-to-day updates people get between these stories. Moving the regular updates to a system that just anyone can post for you lets you focus your resources on the things that provide the most value to the public at large. This output has clear value (making it easier to justify putting a price on it), and the profits from the premium content can support the “regular” news output in-between major reports.
You could, in theory, even charge for the API access needed to report outages/road closures directly to the appropriate services and service providers. These types of things tend to impact broad geographic areas, so you only need a small subset to get adequate coverage for these types of issues.
Wikitribune
Personally, I think WikiTribune is the best guess for what the “species link” between current journalism and what journalism will look like in the future. They’re still assuming a news article is the final deliverable, but Jimmy Wales is smart, and open to innovation, so I’m sure he’ll have no problem experimenting with other forms of output that users suggest. There’s a reason they describe what they’re building as a platform, and it’s to give themselves precisely that kind of flexibility for the future. WikiTribune uses community contributions to supplement the work of reporters, is designed so that the funding scales better than the readership counts, and is likely to support multiple output formats as time goes on. In other words, it’s in the perfect position to let the community focus on the minor updates while dedicated reporters focus on the investigative, in-depth coverage that most of us aren’t equipped to produce.
As popular as it is to say that journalism is changing, a lot of people still seem to be operating under the assumption that it will still appear largely recognizable compared to what we’re used to now – articles being put out somewhere that will probably have some ads attached to them and (for many organizations) require subscriptions. The heavy lifting for creating this content will be done by official, specially-trained “journalists.” There are some organizations that are experimenting with structures that will probably allow them to break away from this “traditional” mindset and start experimenting with other output options, and that will ultimately lead to a period where journalism is more of a public API than a set of professionals. There will still be a need for in-depth, heavily-researched documentation of major issues, but that will finally start getting treated like the premium content that it is. Most importantly, the people we need to build that deep-dive content will have been freed from the time-consuming, easily-outsourceable work to focus on the most valuable parts of journalism. That, in my opinion, is the true future of journalism, and I think it’s likely to start coming sooner than we think.