How The Daily Dot Hones Its Facebook Strategy Through Analytics & Experimentation
Today’s consumers are discovering more content on mobile and on social platforms than ever before. That puts publishers in a bind because instead of readers going to their homepage to discover new content, they’re relying on third-party platforms and apps to surface the most relevant stories for them. Born-digital media companies like Buzzfeed and Refinery29 have excelled in this new era, distributing content across multiple platforms and engaging readers where they are, whether that’s on Snapchat, Instagram, or Facebook.
Distributing content on platforms has inherent risks -- like loss of audience and first-party data -- but there are ways to mitigate these threats, said David Johnson, the former assistant director of business intelligence and data science at The Daily Dot and now director of digital analytics at Haymarket Media. The online publication, also known as “The Hometown Newspaper of the World Wide Web,” has found that constant experimentation and careful measurement can help drive more readers from Facebook back to its website. Johnson offered tips to publishers looking to get the most out of their Facebook strategy during a presentation at the 2016 FUSE Media technology summit. Watch his full presentation here.
Maximize Post Frequency
Johnson wanted to determine if The Daily Dot was posting the right amount of content on Facebook and if a change in frequency could improve referral traffic. He suspected that post frequency was too high, so his team looked at the six months’ worth of Facebook activity and took into account the number of posts during certain time slots and days of the week. What they found was that on certain days and times The Daily Dot was over-posting on Facebook, while at other times it under-posted.
While the site did increase posting during some time slots, overall The Daily Dot reduced posting frequency 13% on weekdays and 36% on weekends. As a result, traffic from Facebook increased 54%.
Optimize Content With Multivariate Testing
Another way that The Daily Dot optimized Facebook activity was through multivariate testing of headlines in its ads. Multivariate testing creates several versions of an ad, experimenting with different headlines and images. The version that earns the most engagement becomes the new ad. Johnson cautioned that this kind of testing is labor intensive and must be done manually. Publishers should also factor in the cost of using Facebook’s ad platform, said Johnson. Due to the expense, The Daily Dot only implemented multivariate testing on evergreen pieces of content and video, because they earned enough traffic to make the experiments ROI positive. As a result of headline optimization, The Daily Dot saw a 36% increase in clickthrough rates from Facebook.
Just Keep Iterating
When it comes to optimizing activity on third-party platforms like Facebook, iteration is key, said Johnson. Because reader habits and algorithms often change, publishers must regularly assess their social activity and test new tactics. “You constantly want to be testing this and maneuvering to see if you can improve these rates,” said Johnson. “. . . It’s something I think you can benefit from by constantly evolving.”
To make regular testing a more natural part of the publishing process, Johnson said publishers should provide accessible and actionable data to editorial and marketing teams. “It’s not just something that shows up in the inbox -- [reporting] has to be something that’s tied to the review process and something that’s openly communicated in the organization for it to be effective.”