What You Need Is a News Peg

Ben Welsh on making data journalism work with that time-honored strategy, the news peg

Screenshot of the Los Angeles Times election map

In the news industry, disruption is an understatement. In recent years, the news media has traveled a course similar to the massive ocean-liners that, ballasts emptied and leaking polluted effluent, are suicidally run aground onto the beaches of East Asia, where their once-prized hulls are stripped down by Pakistani shipbreakers and sold for scrap.

The time-tested enterprise of financing an independent check on government power with print advertising—also known as “the newspaper”—is in collapse. Thousands of journalists have lost their jobs. Formerly vital features, from the paperboy to the pica pole, now seem quaint accouterments of an ancient trade. Those of us who remain must adapt to compete on a crowded digital playing field.

But, now that we’ve got all that Internet crap out of the way, let’s be real. Some things ain’t changed. What Walter Williams professed more than 100 years ago is still true, clear thinking and clear statement remain fundamental to good journalism. What Joseph Pulitzer telegrammed his staff in 1897 is still instructive, selling out your readers by presenting ads as news remains wrong:

News head-line type should never be used in advertisements, 
and advertising type should never be used for head-lines.
Tammany advertisement on fourth page very bad I forbid any political advertisement ... 
unless marked by the word advertisement fully spelled out.

And, though the pace is faster and the audience more disorderly, all the rules have changed and blah, blah, blah, if you want anybody to find your seedlings in the vast, young forests of online news, here’s one trick you should borrow from the unfashionable obsolescence of old media: catch a news peg.

A news peg is jargon for something out there in the world—a dominant news story, a ritual event, a long-anticipated development—that people are interested in. There are news pegs for virtually every topic people care about and they provide an opportunity to hang your news item on something bigger and convince people it’s worth serious consideration. While tailoring news to catch a news peg is an age-old practice of the industry, it’s something I think programmer types like you, dear reader, may too often neglect.

Allow me to be literal. Here are some things that are not news pegs:

  • The morning after your hack day
  • The deadline for your grant’s “deliverable”
  • The end of your work week on Friday afternoon

Now here are some things that are actual news pegs:

  • Your local election day
  • A televised spectacle—like the Superbowl or the Oscars—that millions of people gather to watch
  • The hatching of a billion-strong cicada brood

The difference should be clear. The first list is made up of events driven by you and your organization’s production cycle. You make things when it fits around everything else you are doing. The second list is drawn from the rhythms and accents of the news cycle, events beyond your control that are of societal significance and coincide with when the public tunes in to find information.

If you organize your designs and releases around the first, you will find fewer readers and your work will have less impact. If you identify news pegs on the horizon and tailor your work to catch them as they arrive, you stand a better chance.

It Isn’t That Hard!

Even this chump can do it. I didn’t sleep last night. Yesterday, Los Angeles voters cast their ballots to decide the city’s next mayor. As our team at the Los Angeles Times scrambled to keep up with the latest developments, I helped out on a few tasks in my role as one of the nerds in the newsroom. But, more than anything, I waited.

My moment didn’t come until 3:40 a.m. That’s when the city clerk’s office finished counting votes and published a data file online that reports the vote totals in each of the hundreds of precincts where ballots are cast across the city.

Once it arrived I rushed to convert the data file into a map, which you can now find it featured on latimes.com.

Despite missing some sleep, I didn’t even have to work that hard. I didn’t have to sort out quirks in the data. I didn’t have to argue with coworkers over which colors to use. And, best of all, I didn’t have to fix Internet Explorer bugs. That’s because the whole thing was finished before most of our readers woke up and checked the news.

We were able to do all that in advance because we knew election day was our news peg and, months in advance, we got prepared to catch it. Most of the code was written weeks ago. I was in close contact with the clerk’s office to know when the data would be released and arrange the hand off. The map’s user interface was published months before election day, developed using historic results that we knew would be structured the same as the new numbers.

If we hadn’t put in all that preparation and had instead started fresh this morning, it might have been days before we built and finished a map from scratch. It probably wouldn’t be as good. By then the number of readers still interested in clicking would have been severely diminished. And we might have been beaten to the job by a competitor.

Not all news events can be planned. Unexpected events like the Boston Marathon bombings or the Christopher Dorner manhunt will continue to flare up and quickly dominate news coverage. However, there are many other days on the calendar that traditional news producers plan ahead for. Almost all of these pegs present an opportunity for online pioneers to continue the reinvention of the media—and get the biggest audience—by borrowing a simple tactic from its past.

Five Steps to the News Peg Way

  1. Make a calendar. Identify major news events in the coming months. Is there a data release on a beat your outlet covers that’s going underutilized?
  2. Find a rabbi. Contact the people responsible for publishing the data you’re interested in. Ask them questions. Learn the dataset’s quirks and limitations. Establish a relationship so you will have their ear when new data are released and any pressing questions arise.
  3. Let old data be your guide. Do previous releases of the data you’re targeting for development exist? If so, you might be able to prep the basic outlines of a story or build an entire application using old data and slot in the new records as soon as they are available.
  4. Prepare integrity checks in advance. Write tests in your code that will evaluate that the new data are legitimate. If you’re going to be quickly flipping a large set that has just been published for the first time, you need to check to make sure you loaded it properly and that the source didn’t make errors.
  5. Pre-release the boring stuff. Even if you never alert the public, fully developing and deploying an application in advance lets you finish technical work, like caching, browser bugs and static file hosting. Then you can focus on what matters and frees up your brainpower to respond to any unexpected events that happen when the news peg finally comes.





Current page