I'm an avid user of RSS/Atom feeds.
Several years ago, I developed a system for downloading full-text RSS/Atom feeds from 40 online magazines that allowed me to craft targeted questions for my interviews. I also developed a platform for journalism conservation that featured a serious full-text search engine.
Currently, I have 399 technical blogs in a feed reader for everything from game design to automated testing.
I've encountered bad practices aplenty, such as integration issues that regularly knock feeds offline or effectively offline, or paying zero attention to the design and formatting of feed content. But there are two bad practices among the worst that are nearly universal:
- Delivering excerpts-only content; and
- Delivering a limited number of recent articles.
Editorial conceits: For this piece, I will use the term "publisher" to refer to both amateur bloggers and commercial publishers. I will use the term "business" to refer to the organization that extracts value from content. I will use a first-person voice to address publishers specifically. I will also assume that publishers share analytics and advertising goals.
Bad Practice: Excerpts Only
I understand why you might be delivering excerpts-only content. You want to drive traffic to your website where you can gather analytics and display advertising. But you don't need to force readers along a roundabout path to achieve those goals.
- Analytics: There are solutions tailored for feed analytics, including FeedBurner and FeedPress. You don't need to funnel every reader into Google Analytics.
- Advertising: You can show ads directly in feeds. The industry term is in-feed native advertising. There is, in fact, an IAB standard for native ads. In-feed native ad placements can be added to your existing online advertising menu.
You may also believe that forcing readers to click "read more" encourages participation in your online community, which improves your revenue and visibility.
First of all, comments are bad for technology, science, policy, and many other topics.
Uncivil comments not only polarized readers, but they often changed a participant's interpretation of the news story itself. […] Simply including an ad hominem attack in a reader comment was enough to make study participants think the downside of the reported technology was greater than they'd previously thought.
Second, great content encourages participation; frustrating users does not.
You're probably a Facebook user, so consider this: Facebook is a feed reader and aggregator. As a Facebook user, think about your browsing habits. You don't engage with every post you read as you scroll down your news feed, but when you do engage with a post, you had something to say. What made that post different from the rest with which you didn't engage?
BEST PRACTICE: Set up feeds to deliver full-text content and give readers something to say. Make finding the comments from that content easy, and readers will participate.
Bad Practice: Limited Articles
Two words: dead content. Dead content has no value and earns no revenue. You know this already. You've organized your business around deliberately aging content, so that live content is surfaced to the top and dead content is pushed down and away.
However, to a searchable database, such as a feed reader, there is no dead content. All content has value in perpetuity. Limiting the number of retrievable articles actually turns good, valuable, and revenue-worthy content into dead content. Why do you want more dead content?
You might think the limit encourages readers to visit your website for the same reasons as above (i.e., readers will have to visit your website), but you're missing a crucial fact: there is a metric ton more competition for your readers' attention in a feed reader than there is on your website.
After a few days of operation, I have around 7,000 articles across 399 feeds in my reader. Do you really believe that delivering only the 5, 10, or 20 most recent articles every hour will make you competitive? You're working against yourself by hiding content from your feed.
And when you hide content from feeds and expect feed users to visit your website to find more content, you're hoping feed users will happily try to determine where older articles can be found on your website, identify which articles on your website they've already read in your feed, and then use your website to read the older articles that might interest them.
In effect, you're expecting feed users to behave in a way that runs precisely counter to why they're feed users in the first place, and you're expecting feed users to follow a wholly unintuitive trail of breadcrumbs to content they don't know exists.
BEST PRACTICE: Open the gates to all your free content. You may also find that doing so alleviates your search costs since full-text search is offloaded to feed readers.
These bad practices are ultimately pointless. A determined RSS/Atom feed user can easily scrape your website and fetch the full text for your feed anyway—and, in doing so, strip out in-feed native ads and circumvent feed analytics collection.
Don't give readers a reason to change how you deliver content. If you do, and they change how you deliver content—and they will—recognize the demand and meet those needs.
Treat your feeds as standalone publications with their own markets and consumer profiles. Do not treat them as merely a way to make your website subscribable.
Devote time to understanding, designing, and operating feeds just as you would any other publication. Integrate feeds into your business model.
And please add feed validation to your unit tests. (You are testing, right?) Just like your website, your feed should never break, render poorly, deliver empty articles, or go offline.