Diana Williams (@D_L_Williams) is a social media associate at the University of Virginia.
Leading up to Final Exercises (aka commencement), the U.Va. Social Media tested out Twitter website cards using content from UVA Today.
Website cards, not to be confused with Twitter cards for developers, allow you to feature specific content from a website which you can use in a tweet. According to Twitter, these cards “drive a higher volume of URL clicks since users are able to preview an image, related context and a clear call to action all from their home timeline.”
I won’t bore you with how to set up a website card since it’s fairly easy once you’ve fully activated your Twitter Ads account. Once finished, a card can attach to a tweet and looks like this in the timeline:
The Test
To help put it in context, here’s some Twitter data from the month of April. I’m choosing April because it’s a typical month—no significant spikes in traffic due to holidays, sporting events or major campaigns —and it was the month immediately before graduation. Here’s a quick snapshot of our Twitter performance for that 30-day period:
- 2.8 million impressions
- 6,000 link clicks
In addition, we hit 30 percent of our follower base (30 percent is necessary to maximize your organic reach) by tweeting from @UVA 169 times during the period or approximately five tweets per day. Therefore, per tweet, we averaged 16,568 impressions and 37 link clicks. Keep this in mind—these numbers are significant.
We attached cards to 20 of our graduate profiles. Here’s how they did:
- 524,000 impressions
- 1,600 link clicks
Each carded tweet averaged 26,180 impressions and 80 link clicks. Pretty good, right? However, we decided to tweet four of the graduation cards again, just to see if we were really getting a boost in impressions as well as link clicks. Here’s how one of the four reposted tweets looked originally:
We then modified the language for the second (mostly to make it fit) with just a shortened link to the content and an image:
Brendan Evans played on Wimbledon's lawn at 18. He’ll walk #UVA's Lawn this spring: http://t.co/8xRenOVUqr #UVAGrad pic.twitter.com/wZcikpcXdj
— UVA (@UVA) May 10, 2015
For the second tweet, you’ll see that the link clicks went down when compared to the first, but impressions went up. Here are the stats:
Carded (first) | Impressions | URL Clicks | Regular (second) | Impressions | URL Clicks |
Brendan Evans | 29175 | 128 | Brendan Evans | 47117 | 96 |
Rafat Khan | 16879 | 188 | Rafat Khan | 38541 | 106 |
Blake Calhoun | 18043 | 111 | Blake Calhoun | 46188 | 79 |
Shantell Bingham | 18893 | 111 | Shantell Bingham | 37969 | 27 |
Total | 82990 | 538 | 169815 | 308 | |
Average | 20748 | 135 | 42454 | 77 |
If you look closely at the second tweet, you can easily guess why the impressions significantly rose– they were shorter and accompanied by images, best practices encouraged by Twitter.
Key Takeaways
Here are three things we learned.
- Website cards do exactly as promised—they increase traffic to a specific page on your website.
- Website cards can be used strategically to add greater context. All of the information on the card is condensed into a 43 character url which you can further compact by shortening it.
- Website cards won’t provide a significant increase in impressions over time. However, observing good Twitter practices—including images whenever possible and tweeting with fewer characters—should bump your institution’s reach.
Website cards worked well for our UVa Today content, and we’ll continue exploring ways to use them in the future. Has your institution used them? What was your experience?
Pingback: Web design and images sizes for social media – Geoff Campbell·
Pingback: Twitter Cards now auto-expand on mobile – Geoff Campbell·