Two Weeks With Edgar: How I Increased Our Referrals From Twitter By 655%
Money

Two Weeks With Edgar: How I Increased Our Referrals From Twitter By 655%

by Kate Finley
Photos Diana Zapata | January 27, 2015

Twitter. Sometimes, it seems like the Twitter feed is where your character-limited quips and hashtag-heavy attempts to engage your audience go to die. Quietly, and in the midst of the six thousand other tweets that were also sent in the same second. That corresponds to over 350,000 tweets sent per minute and 500 million tweets per day.

Amongst that many tweets, your message had better be a veritable rainbow fish—as in: “Baby otter befriended by orphan kittens”—if you want anyone to actually see it.

And if your tweet does indeed manage to be seen, it has to make quick, witty work of rustling up engagement. According to Betaworks, half of a tweet’s total link clicks come within the first four glorious minutes after it’s posted on Twitter.

At Career Contessa, we invest quite a bit of time creating “evergreen” content. Our career profiles in particular require quite a bit of lead time to develop, edit, and produce, so when it comes to distribution on Twitter, it’s dispiriting to see all of that fade away in a single, soon-to-be-forgotten tweet.

Writing, re-writing, and re-re-writing pithy remarks for the same piece of content can quickly fritter away time.

If you’ve ever been involved in a business that creates and distributes high-quality content, you know that trying to distribute that content effectively (and without being super-duper annoying) requires some manual labor. Writing, re-writing, and re-re-writing pithy remarks for the same piece of content can quickly fritter away time.

And nobody likes to fritter when it comes to time! When I first heard about Edgar, marketed as “The only app that stops social media updates from going to waste,” I was intrigued. We at Career Contessa use Buffer as a main part of our content distribution workflow (and I am a huge fan of Buffer), so I figured we were set when it came to social media publishing tools.

Nonetheless, I signed up to get my Edgar invitation. A few steps into Edgar’s automated email marketing flow later, I signed up for a free, month-long trial period.

 

 

 

 

PUTTING THE SOCIAL OCTOPUS TO WORK

Like any other social media management tool, you start using Edgar by connecting your social media accounts. It looks like Edgar currently supports Twitter, Facebook, and LinkedIn. 

 

 

Because we update Facebook and LinkedIn far less than Twitter—and because the lifespan of our Facebook and LinkedIn content is typically much longer—I only connected our Twitter account, at least to start out. I wanted to experiment and track results with one social media channel, rather than switch everything over all at once.

Next, you set up what Edgar calls your “Categories:” different buckets for all of your social media updates. How you organize your updates is up to you. I created 11 different categories specific to both type of content and time of update. Here’s what that looks like: 

 

 

You can add as many categories as you need, and organize them with respect to whatever dimensions you’d like to include. You can then fill each category with relevant content, and choose which social network each update will be applied to. In our case, I selected Career Contessa’s Twitter account for every piece of content that I added to the “Library.”

The library is where all of your updates live, organized by account and category. Here’s what ours looks like:

 

 

You can filter by category on the left column, and you can edit each update individually over on the right. Once you’ve uploaded content into each category, you head on over to the “Schedule” to determine the timing of and frequency at which your updates (grouped by category) are unleashed upon the masses.

Here’s a snapshot of our schedule for the Career Contessa Twitter account:

 

 

In order to build out an optimized schedule, I headed over to Tweriod to check out our followers’ activity patterns, and to determine when the majority of our Twitter audience was most likely to be active and engaged (Tweriod is a great free tool for this analysis!). I could then prioritize our most compelling updates around these times each day to increase the likelihood that more of Career Contessa’s Twitter followers would see our best updates.

Here’s a look at our Tweriod report (showing an hourly graph for Saturday and Sunday):

 

 

After slicing and dicing through the various ways Tweriod segmented our data, I used a Google Doc to sort out the timeframes each day in which I wanted to publish our best content. I also had to decide how many times to post each day—in effect, how to strike the balance between informative and annoying—and ended up settling on a different number of tweets for each day of the week.

I’ll test, measure, and iterate on these numbers before I report back, but for now, go check out Buffer’s Social Media Frequency Guide. It’s a great place to start when you’re trying to decide how often to send out your social media updates (on all channels!).

Back to that Google Doc. On this schedule, pink indicates the peak times when most of our followers are active and engaged, which represents the best times for us to post Twitter updates. Green indicates times that perhaps aren’t peak hours, but are still good times to post content.

 

 

The blue boxes are easy—they’re my trusty time-specific categories. The other categories, however, are based on both recent content, popular content, and external career advice content mentioning or referring back to Career Contessa. “Use Once” works just like it sounds: when you add content to the Use Once category, it is published once and then removed from the library. This is where I added all of the awesome non-Career Contessa content that we’d like to share once (functioning just like the standard Buffer queue).

While we used to operate under the 80/20 Rule when it came to sharing content on Twitter by curating other websites’ content much more than we promoted our own, we recently decided to experiment with tweeting a lot more of our own content. I personally think Twitter is the only place you can get away with doing that, but we’ll see!

To avoid being repetitive, I made sure that the frequency by which the category appeared in the schedule related to the number of updates in that category. The categories with more content will appear more than the categories with fewer pieces of content so that the same update won’t be continuously cycling through our Twitter feed. I keep track of an individual update’s frequency in the columns across the bottom.

To determine which content goes into both of our “Favorite” categories, I jumped into Google Analytics “Behavior” tab and grabbed the top fifty Contessa profiles and blog posts. As we move forward into February, some of the updates from the Dec / Jan categories will move into Favorites, and some will move into a new Jan / Feb category, but I’ll have to keep experimenting to see what works best.

Now, for the important part of this post: the results!

 

HERE'S WHAT HAPPENDED

After setting all of this up, I let Edgar run wild and free with the Career Contessa Twitter account for 14 days: from January 10 to January 23.

Prior to this experiment, I was using Buffer to manually schedule out cyclical Twitter promotion for our content, and was keeping track of the promotion dates for each piece of content in the editorial calendar. I am using the 14-day period immediately preceding our implementation of Edgar (December 27 – January 9) as my control.

First, let’s take a look at Career Contessa’s Twitter analytics. In the first 14 days of using Edgar, the number of impressions that our tweets earned (compared to the previous 14-day period) increased by 122%: 

PRE-EDGAR:

 

 

POST-EDGAR:

 

 

Next, I looked at the basic engagement metrics for our Twitter account, and found that link clicks increased by 168%, retweets by 147%, and favorites by 116%. Check out the full comparison below:

 

 

Finally, I jumped over to Google Analytics to see how Edgar had impacted our referral visits from Twitter. The results were by and far the most compelling of the bunch, as total sessions referred from Twitter increased by a whopping 655%:

 

PRE-EDGAR:

 

 

POST-EDGAR:

 

 

TL;DR

TL;DR: Here's a quick summary snapshot of these results:

 

 

Edgar has been crushing it. While switching to Edgar did require quite a bit of work on the front end to build up our library, I spend about an hour less each day adding updates to our Twitter account.

There is one caveat. With Edgar running this entire time, Career Contessa did tweet more frequently than normal. In fact, we published 88% more tweets during the experimental weeks than we did during the control period. Simply tweeting more frequently obviously had an impact on these results—more tweets, more activity.

Would we have experienced the same acceleration in Twitter activity, engagement, and referral visits had I just sat down and composed 88% more tweets without Edgar? Maybe. However, that would have taken a heck of a lot of time, and none of those updates would be recycled automatically later on.

I do suspect that having our best content published at the times when our followers were most active had a significant impact on the spike in referral traffic from Twitter—but that hypothesis has yet to be tested. In the fast-paced world of social media, a solution that saves time while delivering impactful results works for me. I’ll continue to test and optimize in the coming weeks.

Have you tried using Edgar to up your social media game? Which strategies did I miss? Let me know in the comments!