Photo by Andreas Zieroth/Flickr
Photo by Andreas Zieroth/Flickr

I’ve noticed something over the past year: We’re using hashtags less often in tweets. By “we” I don’t just mean AJ+. I mean everyone. In recent months when I’ve checked Topsy or Twitter trends before tweeting, I’ve often seen a keyword(s) for a particular story trending higher than a hashtag.

I’ve wondered: Are hashtags growing obsolete?

Solely based on observation, I’d answer my own question with “no.” But, it does appear now more than ever that people and the media are becoming more selective about how and when to use hashtags — meaning sometimes not at all. At the same time, when we do use hashtags for certain stories, we’re finding ourselves grappling with the ethical implications of using community-generated classifications to enter existing conversations.

Think #BaltimoreUprising versus #BaltimoreRiots. We used both of those user-generated hashtags on Twitter to enter the conversation, to the dismay of some journalists in our office. I’ll get to that. But first, I’m going to explore why at AJ+ we’re shifting toward using fewer (and sometimes no) hashtags when tweeting, and what we debate when it comes to their use.

Hashtag origins and etiquette

In this cute comic, 20px shows what the # sign was before Internet chatrooms and later Twitter turned the symbol into an online taxonomy system for conversation. The first recorded use of a hashtag in a tweet is credited to designer Chris Messina in 2007, when he asked his followers how they felt about using # to group conversations. Props.

Twitter officially added hashtags to their platform two years later, and Instagram, Vine, Google+ and Facebook followed suit. Soon, everyone was using and abusing # to join or start conversation streams. Annoying newbies continue to drive us all nuts #withsentencelongpointlesshashtags. Don’t get me started on Instagramers who use 25 hashtags per photo.

One of our two resident social media queens at AJ+, Stephanie Ivy Whiteside, describes bad hashtag etiquette by the media like this, “A lot of media organizations jumped in, and lot of people who didn’t understand hashtags were using them thinking it was a magic bullet to get ideas across. Now, people have gotten a lot closer to using them for their original purpose — classification used to group overarching topics with lasting impact, like #BlackLivesMatter. It’s really about cataloguing conversation.”

Or creating jokes (users, not media). Think #SorryNotSorry, one of Katrine Dermody’s favorites. She‘s our social media editor. “Using fewer hashtags may be a statement — we’re backing away from the editorial questions that hashtags bring up,” Dermody says. “But when big events come up and we want to jump in on the conversation, we still need hashtags.”

Hashtags = engagement?

That’s because hashtags tend to garner engagement. Adweek has even published a guide on best uses for the “almighty” hashtag, claiming individuals “can see a 100% increase in engagement by using hashtags, while brands can see a 50% increase.” Eh, I’m not so sure about that. I know I’m not the only one who has tweeted something with a hashtag and …. crickets.

You might have thought that by tweeting #CallMeCaitlyn last week you were guaranteed engagement. Perhaps, as the hashtag was trending up. But a quick glance at Topsy near the end of the day showed “Caitlyn Jenner” as a keyword set on Twitter was trending higher than the hashtag (700,000+ tweets vs. 200,000+). We sent out two tweets for that story: One with a hashtag at the very start of the trend, and one without at the end of the day.

That same day, our top tweet was also sans hashtag: Muslim schoolgirls in Belgium who’d been sent home for wearing long skirts. Twitter users had not generated a hashtag for that story, and it’s likely we were one of the only media organizations reporting it. At AJ+ we do not generate our own distinctive hashtags unless we’re actively engaging our audience (i.e. a live Twitter chat using a branded hashtag). We leave that to the activists and tweeps. We trusted the content and kept our tweet clean. It worked.

Our second top tweet that day used #BlackLivesMatter. That’s because the activists in the video were being referred to online as members of that movement. “It was a title, like calling them a ‘civil rights activist,’” says Dermody, who crafted our tweet. “In this case, the hashtag gave people context. And, it was trending over any other keyword for that story.”

By using the one hashtag, we consciously entered an existing, catalogued online conversation with a long shelf life, where the agenda had already been set by users — knowing an actively engaged audience would be likely to retweet.

But for most stories this is now the exception for us rather than the rule.

The ethical dilemma of hashtags

All that said, there are certain hashtags that deliver an ethical dilemma. We encountered this most recently when covering the Baltimore unrest with our use of both #BaltimoreUprising and #BaltimoreRiots.

We had sent a team to report from the ground in Baltimore using mobile phones. At first, there wasn’t a distinctive hashtag for the protests beyond #FreddieGray. We simply used that and #Baltimore.

But as unrest erupted the day of Freddie Gray’s funeral, #BaltimoreUprising began to trend. So we used it for our tweets and Facebook posts to join the conversation. As the unrest continued, the tone online changed: #BaltimoreRiots swung upward. We used both hashtags when there was space. When there wasn’t, #BaltimoreRiots. As with any other hashtag of an editorial nature, we were careful to place it at the end of tweets, or to use it in context when reporting online conversation. But that distinction was lost on some, understandably so. Most don’t read deeply into tweet structure.

Using #BaltimoreProtests, for example, would have been more “journalistically sound,” as Whiteside reflected later. But that wasn’t trending or generated by users. Using either “uprising” or “riots” as a media organization meant we appeared to be aligning ourselves with a viewpoint. That was a debate we had among the social media team just before tweeting. A few others in the office raised that point amid the height of our reporting. We defended our decision by explaining that we were entering existing conversation responsibly, the tone of which had not been set by us.

But that was not the end of the debate. I’ve since returned to those in our office who expressed concern to hear more about what they would have done. Perhaps we will do it differently when presented with the same dilemma in the future. Perhaps we won’t. The rules and troubles of journalism are compounded in the breaking platform of social media. Reporters and editors have always battled to perfect the language to describe an event accurately and fairly. What is an “uprising”? What is a “riot”? Who determines that? How and why did we in U.S. media choose to call the events in the Arab world in 2011 an “uprising” or a “revolution”? Do we use those same determinations (if any) closer to home? Why not?

“In the past, the goal on Twitter was to incorporate hashtags into your copy to save space and streamline. But as time went on, you wondered, ‘Am I editorializing by doing this?’ Do we sound like activists? Placement matters. Most of the time, we aim editorially to go down the center. #CallMeCaitlyn can be thought of as an opinion. So can #JusticeforFreddieGray. Most of the time, we want to let the content speak for itself,” Dermody says.

Amid the Baltimore unrest, I’d pushed one of our reporters to clarify whether an event he was at could be described as a “protest” or “rally” or “community gathering” or “group of people,” etc., before I tweeted on his behalf. But when it comes to hashtags, the agenda had already been set by users. As media, it’s up to us to decide how and whether to use them.

And whether we do, or not, we should be able to defend that decision.

Shadi Rahimi is a deputy producer for AJ+. This is her first monthly column for Poynter. You can follow her at @shadirahimi.