Thinking about 'Won't Subscribe'

Just a few thoughts on Tim Bray’s recent “Won’t Subscribe” post.

I agree with Tim’s premise, in that: 1) the chances of me subscribing to yet-another-publication is quite small and 2) specifically, there should be some way to pay per-article, rather than needing to purchase an entire subscription, and generally, there should be a better way to pay for journalism.

The chances of me subscribing being small is largely in part due to information overload. I already find it hard to keep up with the various newsletters, publications, and bookmarked reading. There simply isn’t enough time in the day to consume all the content that’s available. The issue isn’t really that I don’t want more information, but that I already have access to amble information and that my usage per dollar spent on an additional subscription will hover just above zero.

However, there are articles I’d like to read à la carte, and I’d like to support the journalist and publication who engaged in writing and publishing the article. Sure, the paywalls are easy enough to get around, but someone needs to pay. I try not to get around paywalls as it feels a bit like stealing, and that effect is multiplied by the financial struggles most publishers are already dealing with.

Bray suggests some sort of per-article payment. It’s an interesting idea, and would fit the trend we’ve seen in industries like entertainment consumption (think, a few years ago paying for a cable bundle with a bunch of channels that you won’t watch vs. now, where you can pay smaller amounts to streaming services that have content you want). It would fix the issue discussed above: readers still want to engage with specific articles, but are less likely to do so due to the cost (both financially and administratively) of purchasing a subscription.

However, I feel like there’s a semi-obvious side-effect. Paying per-article adjusts the incentives quite significantly.

When the primary driver of revenue is a subscription, the value delivered to the subscriber is the whole of the content produced by the publisher on a monthly basis. The goal is to deliver quality content on a consistent basis, month per month. This can take many forms: some publishers prefer short, brief snippets, others specialize in long-form work, and the rest fill in the entire spectrum in-between. Audiences can be selective in what they read based in part on their preference of the format. With a broad range of publishers delivering heterogeneous content, the audience has many options to choose from.1

Once the rules of the game have shifted to users paying per-article, I fear that the focus will not be just on what content is delivered, but how that content is delivered. “How” here doesn’t mean the mode in which it is delivered, like print vs, digital, what data visualization is needed, etc. but rather how to extract the maximum revenue from written content. Publishers need revenue, of course, but the current system of subscriptions + advertisements seems to operate separately from the publishing ecosystem, and don’t seem to have had an observable effect on actual article quality2 for incumbent players3.

The ad-based model is maybe a comparable case to the à la carte method, in that number of views is directly proportional to the revenue success of an article, but it’s not as directly tied as a per-article payment system: longer articles can have more ads (both digitally and in print), and advertisements have this whole additional layer of click-throughs, requiring ad publishers to surface relevant ads to the user, etc. So, it’s a little more involved than a payment-per-article, and I think this complexity weakens the relationship between number of articles and revenue.

Removing any nuance, it’s easy to extrapolate how introducing an à la carte payment method might alter the behavior of publishers. If ad and subscription revenue hold firm if the amount of content produced is consistent, regardless of the number articles, then with a per-article payment method it behooves publishers and writers to break content into as many pieces as possible in order to extract the most revenue from curious non-subscribers. Rather than a single 1000-word story, it’s now broken into two separate 500-word stories. Or maybe three 350-page stories. That’s likely too crude of an analysis and a disservice to journalism - writing isn’t some formulaic commodity that can be cut cleanly n times into atomic pieces - but for the sake of argument, it’s to demonstrate that once revenue is tied to number of articles, it’s hard for a publisher to avoid orienting to where the money is. Once that new payment system is introduced, the incentives change drastically, and a new world order emerges. It may not happen immediately, but as long as revenue from ads and subscriptions are not enough4, publishers will drift toward generating more and more revenue from the new revenue stream; to increases the margins on that revenue stream, once can hold the costs static by not hiring more writers to write more articles while still generating higher article numbers from the current staff by reducing the length of articles).

Maybe one way to disincentivize short, click-baity articles is to have the article cost be correlated with the length of the article. Changing an article fee from a flat-fee to a sliding scale based on length may treat short articles and longer articles more fairly from a revenue perspective. However, article length is a bad proxy for article quality: in fact, with this pricing model in place article length may be inversely proportional to quality, if one prefers brevity.

But, that’s just one alternative that comes to mind.

Overall, I agree with Tim’s point that there should be an alternative way to engage with publishers without subscribing to them. It can be a mutually beneficial relationship: publishers can drive more revenue and users can engage with more content, with the publisher-subscriber relationship currently being the limiting factor for both of those benefits. However, I’m a little concerned about the potential side effects that will emerge alongside a pay-per-article model, and I fear that short, sparse, prosaic journalism will spread once it’s clear that à la carte payments can provide additional revenue, and now all publishers must play the game. I fear that short-term incentives will result in a worse experience on both sides of the table in the long-term. Unfortunately, I don’t have a great alternative, and so maybe trying something is better than doing nothing.

  1. There is a flip side here for publishers: they need to make money, and money-making comes back to subscriptions and/or ads, both which are driven by engagement and thus driven by eyeballs. I don’t have insight into how performance metrics work for publishers - maybe it is based on article views and my point here is moot - but as an outsider looking in, I assume there are many more dimensions to measuring performance, simply because we would see the gravity of revenue slowly pull article lengths to as near-zero as possible. 

  2. This is ignoring the long-term effects of revenue success: less journalists, editors, support staff, resources, etc. Of course, a publisher needs to generate enough revenue to support a well-run organization, and needs to see a return on capital invested for the content in aggregate. If that’s not happening (because the ad- and subscription-business is not generating enough revenue), then quality can definitely decline. 

  3. Some new players in the publishing world likely were born out of the ad-supported model; this brings to mind publications that in the past had been 

  4. A publisher introducing a new revenue stream likely implies that the current revenue streams are not enough, whether “enough” is as dire as supporting the newsroom or just to make shareholders happier.