December 28, 2012
When newspapers were print-only, the idea of optimizing headlines, images, and content, or doing audience segmentation, must have seemed impossible. Today, most news organizations that are transitioning their content to online, and mobile platforms haven’t realized the potential that exists for detailed tracking, analytics, and optimization.
I sat down last week with the two founders of a new DC startup focused on solving this problem. OpBandit‘s Blaine Sheldon and Brian Muller both have a background in news organizations (having worked at The Slate Group, Ojo Gringo, and Foreign Policy magazine) as well as a background in technology (Sheldon is an expert in UX, UI, interactive media, and data visualization, and Muller is an engineer and data scientist who built and led the data science team at LivingSocial).
OpBandit is empowering news organizations with a light web service that delivers responsive content to users through real-time audience targeting and page optimization – and I was lucky enough to get a sneak peak. The product helps organizations test headlines and photos for a story, optimize which stories take key placements, decide which content should be shown to individuals coming from various sources, and effectively monetize their audiences.
Online publishers show the same pages to everyone across platforms, speculating which content will best appeal to their wider audience at a given time. While some competitor analytics tools exist, they can overwhelm editors who, instead of focusing on content, currently experience a deluge of information with few easily actionable items. OpBandit is solving this by automating and simplifying the experience.
Rather than run isolated A/B tests, OpBandit is in constant optimization mode. Optimization begins from the very first page load and never stops. Headlines, images, and placements that don’t convert will begin to be shown infrequently and vice versa. For example, content editors can create five headline variations and three image variations for a story and let the system take over to decide which headline and image to show most often in order to drive higher engagement.
One exciting feature of OpBandit is segmentation based on referrer, meaning that publishers can show different headlines or photos to different readers based on what sources they are coming from. For example, a publisher may want to show a different image for a reader who is coming from MSNBC than a reader who comes from Drudge Report; it may be more effective to show an image of Obama for the first, and Romney for the second.
This is not about changing the nature of the news story based on political leanings; rather, it optimizes how the information is presented to readers to increase engagement. Note that organizations aren’t just optimizing for clicks. Sites have the option to optimize against multiple data points including, but not limited to, clicks, time on page, comments, and shares. OpBandit can also make decisions about which stories should be featured at the top of the page, automatically rendering the strongest content to important, designated positions across a page.
OpBandit can help publishers sell their site placements more effectively to advertisers by leveraging detailed data and analytics on influential readers and audience profiles (based on pageviews, time on site, demographics, etc.), helping amplify sponsorship campaigns.
OpBandit is currently in beta testing with a select number of companies. Their technology will surely garner attention for its ease of use and ability to improve site engagement. Lastly, I’m curious how this technology could be applied to other use cases, especially in online retail and flash sales.
Guest author Micah Cohen writes about growth strategies and the LA startup scene. He currently works on growth for the fast-growing social dining startup based in LA, Grubwithus, and previously worked on user acquisition at LivingSocial. He is passionate about using technology to re-imagine and improve lifestyle. You can follow him on Twitter @miccohen.
Did you like this article?
Get more delivered to your inbox just like it!
Sorry about that. Try these articles instead!