This week in my Social Media Theory & Practice class, we read several articles about understanding how to get the most out of posts on social media and understanding metrics. The first article, “10 Questions To Ask About Your Twitter Reach & The Free Twitter Tools To Answer Them” by Gerry Moran, provides useful advice on tools to better measure the effectiveness of Twitter, concentrating on amplifying messages, engage followers and conversion. The 10 questions included: (1) What do I know about my Twitter followers? (2) Where are my Twitter followers? (3) What part of my follower reach is really not worth following? (4) Is my messaging reaching an audience beyond my followers? (5) When is the best time to tweet, so it reaches most of my followers? (6) Is my messaging reaching an influential and geo-targeted audience? (7) How can I find people who should be following me? (8) Who are the most influential people on a subject? (9) How healthy was my social network today? (10) Am I positioned as an expert so I can increase my message reach through retweets and Twitter listings. In reading about the different tools that can assist in provide data and metrics to these questions, the most robust two that stood out to me were Ads.Twitter.com and FriendorFollow.com for the kinds of data they can provide.
Working at a university, our Office of Communications & Marketing frequently receives requests from other offices and academic departments that want to jump feet first into social media, usually establishing a Facebook or Twitter profile. Oftentimes, they’ve already launched them and are floundering. I wish they could understand that social media is not just the latest way to get in front of the high school market or even current students. They usually want a page/profile because everyone else has one. They do not ask the kind of questions mentioned in this article. These questions go to the heart of creating a strategy for engagement and being able to track that. What we usually see is, for example, an academic unit establishing a page, which gains less than 100 followers and they can’t figure out why they are not growing. Or the other scenario is that someone from their department has established a page, and then abandoned it, so the page has lost all relevance. We do strongly encourage all units to assign the updates on their social media to a full-time employee versus a student. Student workers may be the easiest solution in terms of knowing how to do it, but they also come and go each semester, and there are security concerns. We have created Social Media Guidelines to bring a level of consistency campus-wide to TTU’s social media presence — http://www.tntech.edu/web/socialmedia/.
The second article for this week was “A Scientific Guide To Writing Great Headlines on Twitter, Facebook, And Your Blog” by Leo Widrich. This was a really a good review for writing traditional headlines with some helpful tips for writing for online and social media. I thought it was interesting on the emphasis on numbers in headlines and making them more “how to” or DIY. I think this help make the content evergreen/timeless when you write lists that can remain relevant. And of course, the tip on using action verbs also holds true for print media. I thought it was good to that the writer strongly advised that when posting, be specific about the action you want your readers to take — download, visit link, retweet, etc. So be declarative in your request for action — I think this harkens back to positioning oneself as an expert online and via social media. I also think it is very true that photos are the attention-grabbers on Facebook. People respond to them and share them. This reminded me of how traditional papers always have photo above the fold, sometimes just a standalone photo with a cutline to attract reader attention. Another good tip from this article was when writing for your blog and you are searching for a topic — try tweeting the topic first and see what traction it gets. This will be an indicator of how successful your blog post can be.
The third article for this week was “5 social media metrics that your business should be tracking” by Luke Chitwood. As someone who works in marketing, I understand the importance of metrics and ROI. Some of the metrics that stood out to me here were: reach, engagement and conversion. On reach, I agree that the growth rate of your audience depicts your social media momentum. But perhaps the most important metric is engagement. This is the metric we look at for the university’s social media — how many people liked a post, commented on it, shared it, retweeted it, favorited it, posted original content back to you, etc. Engagement is measuring what is resonating with your established base. Of course, conversion is the metric where you expect to see an action. It would be ideal at the university if we could capture the true conversion rate. However, this is a big challenge because so many offices are working to recruit a student (Communications & Marketing, Admissions, specific Academic Department, Student Orientation, etc.). Each office collects data to the best of its ability, but we have never been able to funnel that data together to get a true conversion rate or even an accurate number of what it costs to recruit a student. So, on our end, the best metric I can show for the university marketing I do is reach and engagement. At my level, I can only show I brought so many eyes to the page and how many inquiries were generated. Marketing can only bring them to the front door, the other offices I listed have to bring them through it. One metric mentioned here activity and the customer service savings was a new concept to me to measure in a formal way. We have currently started measuring number of inquires at the campus help desk and the number of inquiries received through the university web master email. I think I will start tracking the number of questions answered through social media because honestly that is where most students connect.
The final article for this week went deeper into metrics with a specific case study. The article, “How Promotion Affects Pageviews On The New York Times Website” by Brian Abelson. This article detailed a research experiment that looked at the New York Times’ promotion efforts in relation to the success of an article (how many pageviews it received). I do agree that page views are probably not the best measurement, but as an advertiser that juggles multiple media when I am doing a campaign, they are an easy go-to metric to let me see what kind of reach my ad is receiving. I think this was valuable research in using PAR (page views above replacement) since it created a predictive model for articles and correlated their method of promotion to predict how many views it will receive — in other words “a clear relationship in the data between promotional energies and page views.” The predictive model follows logic in that the more an article is promoted (either by time spent on the NYT homepage or if it is promoted on NYT Twitter account, then the more page views that article will receive). Probably the most interesting fact here is that the model can explain almost 90% of the variance in page views. The one thing not addressed here is that as an advertiser, I would want to be able to pick the best combo for my ad to be seen on — so I would want to always be seen with the articles that are most likely to come out on top in terms of page view. And now with a predictive model, I wonder if this might hurt other areas of the news — like sports articles being more popular than arts articles? Could this be detrimental to less popular sections of the news?
I was not able to read the full article on Al Jazeera English — only the abstract. So, there may be an update to this reading section if I can access the full article.
UPDATE:
Dr. Brown found a new link to the full Al Jazeera English (AJE) article. Below are my take aways from this article, “Al Jazeera English Online: Understanding Web Metrics and News Production when a Quantified Audience is not a Commodified Audience” by Nikki Usher. This scholarly article focused on AJE as the Arab’s world largest English-language news site to reach an international audience and showed how, although freed from economic forces, journalists at AJE still use metrics to a certain extent to gauge audience reach. The author addresses upfront the finance issue at work with AJE — it is creation of the Qatari royal family and supported by oil revenues. And even though the family insists the news organization is independent, I don’t know if I fully believe that.
The article focused on the use, or lack thereof, of Web metrics in the AJE’s newsroom. Launched in 2006 to counter the lack of Western coverage of the Middle East, some notable metrics mentioned for the site include: available in 220 million households across more than 100 countries, averages 8 million visitors a month, and more than 50% of web traffic comes from the U.S. The author’s research focused on absent the pressure to show metrics to support financial decisions, she wanted to know whether editorial judgements in the newsroom were affected by metrics when journalists decided which stories to cover/write about.
The author here found that at AJE, although individual journalists had access to basic metrics, using Chartbeat (and upper management used Google Analytics), the use of these tools did not appear to influence any editorial judgments on what to cover. It was interesting to note that the journalists used the tools as a method of personal validation — i.e. personal info on how successful my story was based on readership. Many journalists were aware that upper management used Google Analytics, but none of that historic info was shared from the top down, which indicates to me that you want to keep your people in the dark versus maintaining editorial integrity.
Reading this article, I couldn’t help but think of the old journalism cliche of “if it bleeds, it leads.” Even before the web was in wide use, newspapers in print have been making decisions about what to include on the front page above the fold to help sell papers and get a wider circulation. So, that is why a murder makes the front page versus a new public school opening — it sells more papers. That is why newspapers, especially those found in small towns, still do non-traditional pubs/special tabloids — not because the content is particularly interesting, but because it sells ads to boost overall revenue. Some examples of special sections — Best of the Best for local businesses, home improvement, back to school, etc. I remember writing a paper about the history of my hometown paper, the Herald-Citizen in Cookeville, TN. I interviewed the managing editor at that time in 1998 and asked what was the biggest story they ever covered. The answer was the assassination of local state senator Tommy Burks by the local property assessor who was running against him. He judged that based on selling out of newspapers that featured the story on the front cover.
Overall, I think an awareness of metrics is key in today’s data-driven world. Even at the university level where I work — we are not paid by the story, but we do have to show number of clippings for our press releases and number of clicks on the webpages. We have to give account for our web traffic and explain how that influences our website’s site map. Social media is another tool we track data on to see how successfully we can engage our stakeholders/audience.