Everyone is setting up topics pages these days; it’s the new fad/holy grail in SEO-ed sites.
It’s not surprising. A lot of traffic comes in via topics pages – and in theory, good topics pages really do serve readers well, by providing context and understanding. The operative word being “good.” Most aren’t, by a long shot. And this post, at The Content Strategies Blog, really skewers the bad ones and offers solid advice for making them better. (Thanks to Yolanda for pointing it out to me.)
Some of it is obvious: Automated topics pages suck. And that topics pages are best when they’re curated by humans, with updated introductions to explain why the page exists. But there are lots of other good tips, on multimedia, the dangers of relying on keywords, ideas on design, and so on.
As well as this one:
Structured data. Here we’re getting closer to the Semantic Web holy grail, but providing structured data can be hugely valuable. For a politician, provide his voting record. For a football player, his latest stats. For a company, stock information. For an actor, a list of his movies. You get the point. Integrating structured data need not be a manual process.
Now that warms my heart.
But there’s more to a topics page than just curating stories well, designing the page attractively, and adding some data. The whole page should be structured. Stories should be structured so readers don’t have to figure out date references; information and value should be extracted from stories so they can be presently easily as metadata and aggregations of metadata, ala Politifact; and with smart timelines, ala Living Stories. And more.
All this could be expensive to do, of course; and expensive topics pages aren’t sustainable. But there’s a fine line – actually not fine at all, just a huge mindset adjustment – between human-curated and written topics pages, automated pages, and structured pages that leverage new newsroom processes that extract elements from daily work on the subject to fit into structures on the page.
Call it the happy medium between expensive human curation and still-imperfect machine logic. And a cost-effective way to serve readers well.