Archives for posts with tag: seo

This session was aimed at print editors who need to make a transition into online editing, but I thought I’d come away from it with some ideas for what we should be teaching. Some of the most interesting information I got was about search-engine-optimizing headlines.

Dan Gaines of LATimes.com, again, was one of the three panelists. LAT is part of Tribune Co., which employs one of the more prominent SEO experts working in journalism, Brent D. Payne. LAT customizes each Web story’s page title, which is the title that appears on the bar at the top of the browser window, and that title is different from the main headline appearing in the content area above the story. (For those of you who know HTML, we’re talking about the <title> tag and the <h1> tag.) LAT adds literal search-term keywords at the beginning of the title, in a way that they might not appear in the headline. An example Dan showed was this story live now:

  • Headline: U.S. warships launch airstrikes on Libya
  • Title: Libya attack: U.S. attacks targets in Libya – latimes.com

Tribune’s theory is that the title is much more important to Google than the headline — and that front-loading the keywords is advantageous. So assuming “Libya attack” is a good SEO keyword, that gets prepended to a workable headline for the story — and the site name is appended, which I presume happens automatically — and we’ve got super-optimized display type. I’d argue (and Dan did, too) that the title here is unnecessarily repetitive; it could have been something like “Libya attack: U.S. warships launch airstrikes.” But you see how it works. We can’t specify a separate title in the Missourian’s CMS, but it’s a feature we could easily implement.

Another bit of information that I found useful came from Doris Truong, who’s an editor at The Washington Post and a Mizzou alum. She reminded me about Google Trends, which is an easy way to see what keywords are important right now for searchers using Google. I usually recommend that our copy editors dip into our own Google Analytics account to see what keywords are important to the Missourian, but that information is only good for stories that have been in the news for several hours at least. Google Trends is telling us what is hot at this minute, and it can show you national trending as well as regional trending. It also allows you to compare different terms. A cool tool.

This item has been cross-posted at the Reynolds Journalism Institute’s blog.

Much has been said and written by editors about the need for headline writers to understand search-engine optimization. This is certainly important, although it is really just a different way of thinking about something that headline writers have always thought about: keywords. I’ve heard much less about another aspect of online headline writing that we talk about here at the Missourian. I’ll call it “reader-optimization.” First, though, an overview of what we’re teaching and doing.

All Missourian copy editors get access to our Google Analytics account, where they can see what terms our readers are searching for, both via major search engines and via our own site search. In class, we cover SEO for headline writing, particularly the importance of proper names in headlines. And we learn on the job. For example, early in the fall semester, we figured out that our stories about Missouri tailback and newsmaker Derrick Washington were much better off when we included his first name along with his last in headlines. Later, when an underage Columbia College student fell off the stairs at a downtown bar, we began using her name in headlines when it became clear it was a popular search term. We insist on including “Missouri” or “Columbia” in almost every headline where that’s appropriate, way more often than we’re naturally inclined to.

But writing headlines for the web can’t be just about SEO. About 45 percent of our readers at ColumbiaMissourian.com come from search engines. Another 35 percent are referrals, meaning they clicked a link to our content from another website, or from an e-mail or a tweet. The remaining 20 percent — a number we’d like to increase — type our address or click a personal bookmark. These are our regular readers. They probably live in Columbia or have some tie to it, and they probably came in the front door, the home page, to poke around for the news that interests them. These readers aren’t so different from the print readers who pick up the paper and start on Page 1A, looking for the headlines that pique their interest. It follows that the headlines we write for these home-page readers shouldn’t be so different than the ones we try to write for 1A. Intriguing. Terse. Story-selling.

It’s common sense, really, and I wondered how many other sites were “reader-optimizing” their home-page headlines. I assigned my copy editing students to each choose a newspaper, find a story printed on Page 1 recently, and compare that headline to the headline on the home page for that story and the headline on the story itself on the website. Their findings? Most newspapers write snappy headlines for their print front pages and prosaic, keyword-laden headlines — SEO headlines — for their websites. But most don’t bother to rewrite their headlines for the home page.

Granted, some online content management systems make this impossible. If that’s the case in your CMS, you should complain to its developers and request a feature that allows you to override the story headline on your home page, or make sure that feature is on your list when you go shopping for a new system. Our CMS at the Missourian — a internally developed system based on the Django Web framework — allows our editors to rewrite all the display type for the four stories that are featured on our home page. We’re not consistent about doing this, but it’s our goal. We want the headlines on our home page to speak to readers, not search engines.

To the list of things Joy has suggested are at the core of interactive copy editing/web production (prediting? proditing?), I’ll add this: optimizing the home page.

Often through the day, editors should spend a minute or two staring at the home page and asking themselves: If I didn’t work here, would I be interested in any of this? Do the headlines and excerpts do the best possible job of selling these stories to me? Are the right stories featured?

Each weekday this summer, the home page received about 6,000 page views on average. That’s by far the most trafficked page on our site, and it represents a lot of opportunity to engage our community. This graph shows the fairly consistent pattern of daily home-page views for the summer.

Daily page views on the home page

However, the bounce rate from the home page — the percentage of people who visit only that page of our site — hovers around 40 percent. (The graph below shows the weekly bounce rate for the summer.) In other words, two out of every five people who visit the home page find nothing worth clicking on.

Weekly bounce rate from the home page

Is the site that dull? I’m going start with the hypothesis that it is not — that the problem is mostly that we’re not doing enough on the home page to put our best foot forward to fully engage readers in our work.

How do we do that? It’s starts with interactive copy editors and news editors critically asking the questions above. It’s getting the right stories on the home page and the best stories featured most prominently. It’s about taking as much care in optimizing the home page headlines and excerpts as we take in crafting story-level headlines optimized for search. On the home page, the optimization is not about satisfying Google’s spiders and search-engine users; it’s about satisfying our local readers. More often than not, we should be overriding the SEO headline and writing engaging, conversational headlines for the home page. That’s easily done at the layout level in the Django system.

What else does the home page need to liven it up? What routine will we work into the ICE shifts to make it happen?