Yesterday, I lead a Webinar for GSA’s Web Manager’s University in which I covered the basics of Twitter of Facebook.
Alice Lipowicz of Federal Computer Week was in attendance and wrote a synopsis of the session, which itself sparked a conversation between myself, Ari Herzog, Alex Howard, and a few others on Twitter.
There were two main issues that Ari and Alex raised, and I’d like to explore them further in the format of a blog post rather than on Twitter. And before I even bring up the issues, I should say that the conversation on Twitter helped me to refine my thinking on the subjects we covered, and will certainly have an impact on how I discuss those topics in the future.
In that light, I’m thankful to Alex, Ari, and the other participants, and I’m very interested in reading further responses either here on GovLoop or in other venues.
Here are the issues that were raised on Twitter:
- Hashtags: I said they should be used “pretty much always in every tweet.” Ari (and others, including @mchronister and @ribock) disagreed.
- Broadcasting: I said that applications like TweetDeck and HootSuite “allow for advance arrangement of multiple tweets throughout the day, and of tweets that are simultaneously broadcasted to Facebook and other platforms.” Alex strongly disagrees that agencies should think in terms of “broadcasting” to “audiences” when they turn to social media.
Let’s take these one at a time.
Hashtags
My discussion of hashtags centered around what I consider to be the two most important benefits of using them: first, they help the author of the tweet keep the message focused and its readers in mind. Using a hashtag (or two) helps authors answer the important questions: what is the essence of what I’m saying, and who are the people who would find this most interesting? Second, hashtags help people who are not following the tweet stream to find the tweet based on its content, not on the author. I gave the example of #gov20, #tech, #data, #health, and #justice as examples of hashtags that government tweet streams might use.
Informing my strident assertion of “pretty much always in every tweet,” was a poll that was conducted during the Webinar that asked participants how often they used hashtags. Of the 260 respondents, 42% said “0% of the time;” 29% said “1-25%;” 8% said “26-50%;” 4% said “51-75%;” and 16% said “76-100%” That is, nearly half never used hashtags, and nearly a quarter used them between 0 and 25% of the time. Exhorting them to include hashtags with every tweet was perhaps an over-endorsement, but I think it was warranted given the audience. The larger point, however, was tagging tweets with metadata, as I said, both for the sake of the author and the reader.
As an example of good tweeting, let’s take a look at Alex’s feed. On October 30, he posted 50 tweets, of which 14 (28%) had hashtags. That said, only a single tweet of Alex’s had neither a hashtag nor a mention (e.g. @AntDeRosa I thought that was the right interpretation.) . I do believe that at another point in the Webinar, I said that a mention essentially “counts” as a hashtag (again, since both are “metadata”). So, looking at both hashtags and mentions, 98% of Alex’s tweets were tagged with metadata. Of Ari’s 18 tweets on October 30, every one had either a hashtag or a mention.
Looking back, if I could amend my suggestion (and I will do so in future presentations) I would say that “pretty much every tweet should have metadata–either a hashtag or a mention.”
Broadcasting
This was the more contentious issue. I spoke, more than once, about “broadcasting” tweets or Facebook status updates. Alex’s objections to the use of the terms “broadcasting” and “audiences” are captured in two of his tweets (and this longer item):
“broadcasting” is one way communication, one->many. Radio & TV exemplify the form. Social media is by definition beyond broadcast. permalinkIf you advise an entity to “engage” the “audience” by broadcasting TO them, are they going to listen? permalink
I responded that one can certainly broadcast through twitter, especially if tweets include a hashtag. For example, when the USGS tweets“No #tsunami warning advisory after 6.9 earthquake off coast of #Peru: http://bit.ly/vd30uP” or when they tweet “Links to #earthquake photo collections may be found here:http://on.doi.gov/tHRuSJ Check it out, students & teachers!” they are talking to their 160K+ followers, of course, but also to people interested in earthquakes, tsunamis, and Peru. Is that not example of one-to-many communication? And is that not a good use of the USGS twitter feed?
I would argue that is both broadcasting and a good use of the feed. Further, I would argue that they have identified an “audience,” in people who are interested in tsunamis, earthquakes, and Peru. In reference to Alex’s tweets, I would say that social media does allow for broadcasting, and that broadcasting can be either the beginning of engagement (to which many people would listen) or a good in itself (don’t worry about that #tsunami!).
As a benefit of using Twitter, either of these posts can also spark true engagement: other followers can tweet back to USGS with their own photos of the earthquake, adding to the discussion, or they share the link from the first tweet about Peru. But just because twitter enables efficient two-way (or really multi-directional) communication, it does not follow that agencies or people should not turn to Twitter as an effective medium for disseminating information quickly to interested people (i.e. an “audience”).
In the future, I’ll certainly be more conscientious to point out that broadcasting is only the smallest part of what we can and should do on social media, and that it will ideally act as the beginning of an engagement.
Those are my expanded, though still evolving, thoughts on this subject. Feel free to respond (or tweet!) your own.
One element I’d add to the discussion is “what is realistic”
With shrinking staff especially at local level, I really think it’s just hard for agencies to have the time to truly engage on social media . So I think using as regular broadcast that is scheduled to be consistent while dipping in to engage when have time…is a good realistic recommendation.
I think another piece may be that the deep engagement is needed at special times – like a snow emergency, crisis communication (or like how USAJOBS is using Facebook now at deep engagement when issues)
Responded, in a blog post that I refused to tag out of principle. https://www.govloop.com/profiles/blogs/organizing-the-commons
Waitasec Gadi. If a mention on Twitter is metadata (which is news to me), then wouldn’t Twitter itself be metadata too?
Step back and imagine you’re sending the same message to multiple media. Wouldn’t each medium be part of that messages’s meta?
Ari, metadata is “information about the information.” A hashtag is metadata because when you see #gov20 or #tech or #data, you know what the tweet is about (or at least what it purports to be about). Likewise, if you start a tweet with a mention, @gbyehuda, a reader could surmise that your tweet is addressed to me, perhaps either a reply to a prior tweet or a question or something.
In either case, a hashtag or a mention, readers scanning your tweet stream, algorithms looking for trends, and applications performing searches can deduce information about the information of your tweet. Simply knowing that your tweet was posted on twitter, or that your update was posted on facebook, does not reveal information about the information. So it would not be metadata.
We can also see metadata applied in terms of location, timestamp, etc. Are you arguing for syntactic or semantic metadata? If syntactic, than Ari is correct – that is embedded or implied by the medium. If semantic, then my cogent, unaddressed blog says it all… 🙂
Hashtags are not metadata. They are content just like the tweet itself. They are a poor way of trying to organize all of the information that comes out via twitter. When you have one big giant chat room (which is all twitter is) then the community will come up with some order out of that chaos. That is how hashtags were born.
One day Twitter will break apart into topics, rooms, subjects, etc. – whatever. They will have to come up with a way to organize and group the information. Then twitter will start working like IRC rooms have worked for decades. Then real metadata will become part of each tweet.
I have observed that certain hashtags are commonly used among the people following certain topics (#smem, #gov20 #opengov for example) and in effect they have become metadata, and are fulfilling a need for organizing material on Twitter. I see advantages to using those hashtags for those constituencies. If other people are interested, they presumably will find the content, too, through other types of searches.
So as various Twitter communities are coalescing, I think we will continue to see hashtags as metadata for those communities. Some will be ephemeral but some will be longer term in nature.
You would be able to see the tsunami tweet whether you search for tsunami or #tsunami. I think the more interesting question is how can these hashtags – or channels – can mature.A lot of the time they get sorted, renamed and matured relatively quickly and organically by the community.
The snowstorm from two years ago was a perfect example – you didn’t put #snow #dc, you started listening to what eventually became the #snowpocalypse or #snowmageddon channels – the top 2 during the storms.
The point is if you are a government communications expert, you need to monitor how these channels emerge, develop, and mature to truly reach your audience in an effective way. And to be able to do this you need to be listening and engaged, not just broadcasting.
Just slapping on #snow #rain or #dc isn’t the best strategy.
And for those who do not know, this is one of the top reasons why you might get follow requests from people. I keep my account closed and must approve anyone requesting to follow me. Mostly it’s from friends, associates, and friends of my friends/associates. But, I sometimes get requests from people and wonder who they are and how they found me. It’s largely from hashtags. The more you use them the more people will find you. If you’re interested in getting more people, such as those in business or those interested in connecting to folks for socially-conscious or other non-personal reasons, then hashtags are great. But generally as an ordinary, average citizen, I don’t think they have much value.
Some people, though, seem obsessed with them. They use hashtags anytime they write something online, whether on Twitter, a blog, Facebook, email, or wherever. That’s a little extreme.
Mike: you can certainly have a search running for “tsunami” instead of #tsunami,” but consider the difference between running a search for, say, “tech” instead of “#tech” or “data” instead of “#data.” Lots people might mention “tech” in a post that doesn’t really have to do with the subject of tech. Or data in a post that doesn’t really have to do with data. So the difference is, are you looking for mentions of a word, or for people who are ascribing to their posts a general theme.
Now, a danger is looking for “#health,” as an example, is that people try to game the system–promoting a book or a pill or some other product with a popular hashtag–but I’d still recommend that government agencies use hashtags so that people can follow their conversations without having to subscribe to their feed. As an example, I think FEMA’s “#safetytips” is a good use of hashtags.
Gadi,
I agree hashtags are important. But you are considering them as search terms instead of channels. For example, #gov20 is a curated channel that people monitor regularly and if people take advantage of the channel – spam it, fill it with unrelated content, etc. – the community will call those people out and correct their behavior. To use your example on healthcare, there are already strong communities using #FDasm, #hcr, #hcsm, etc. You would have more success reaching out to those groups first than you ever would with #pill.
Again, my point is use hashtags all you want, but you will reach influencers, ambassadors and stakeholders in the established or emerging channels, not by arbitrarily picking potentially related hashtags.
Gadi,
I love using Hootsuite to monitor hashtag conversations, twitter chats and mentions. But I don’t like auto-connecting/broadcasting Twitter and Facebook. They are two distinct networks with unique sets of etiquette and norms. To encourage engagement, tweets need to be structured differently than Facebook updates (ie, use of @, RT, MT, #hashtags are Twitter specific). Also, Twitter is a high volume network where tweeting several times a day is norm vs Facebook. Thoughts?
Thanks for the comment, Gladys.
I completely agree that Twitter and Facebook have different conventions and require different writing styles. My suggestion to use an application (whether TweetDeck or HootSuite) was simply to streamline listening and posting–so that people don’t have to switch between Facebook and Twitter all day. I was also careful to mention that while Twitter.com was a waste of time (in my opinion) and that third-party applcations presented the information that Twitter.com does in a better way, Facebook was not so bad, and might even be far superior to third-party apps, especially in presenting ongoing discussion on a single topic (as an example).
I think you’re also right regarding posting volume on Facebook and Twitter.
Twitter vs. Facebook vs. “both” looms as a battle in our organization. It’s policy-from-above here to not identify oursevles personally, which reduces opportunities for engagement and leaves us with broadcasting as our primary way of using social media. I’ve found that there’s not much overlap between our Twitter followers and those who “like” us on Facebook, so using HootSuite to post to both makes sense for us — unless we’re writing a post that fits into the style of one service more than the other (i.e. @ mentions or hashtags).
I do find the concept of metadata in tweets to be useful and will begin employing that tactic with greater frequency.
Dave – I think you’ve hit upon an important point: the ‘audiences’ for each medium are different. That is one reason why it’s so important to post information on both platforms. But it’s also critical to heed what Gladys said about each platform requiring its own style of writing and engagement.
Out of curiosity why do hashtags have to have one particular definition or meaning? They clearly have a different purpose between each type of user. While brands might use hashtags to create a trend or associate themselves to it, another brand might use it to reach a particular audience. Personally I use it as both a meta tag, and a way to reach the hidden communities on Twitter.
@Eliiot: it’s true that the same hashtag may mean slightly different things to different people (Gov20 chief among them); there’s a Wittgensteinian problem here, no doubt.
All I would say is that Twitter communities have ways of understanding hashtags and, as Michael pointed out below, there are gradations within interest-communities: #smem, an example.