But with AI, you can effectively run all the same analysis that you perform for your own pages on competitors as well. This gives savvy marketers a huge advantage by being able to see what is working and integrate the best strategies and techniques into their own strategy.
Tools available range from social listening and sentiment analysis like Netbase, to software that analyzes the content itself for qualities that impact performance such as keywords, color, visual objects, or content type. There aren’t many platforms that do all this, but beyond Cortex, Crimson Hexagon, and Simply Measured have some of these capabilities.
Once you understand your audience, the landscape, and your competitors strengths and weaknesses, planning the best content to deliver to them becomes the last piece of a social media strategy. Although social media is continuing to trend more visual, getting the right mix of text, photo, and video for your audience can make a big difference in performance.
Beyond that, what is in the photos and videos makes a huge difference in performance. In a study we did for Visit Utah we found that photos with Fir trees performed 53% better than average, while photos of families performed 30% below average.
90% above average for this photo41% below average for this one
Going into this level of granularity reveals surprising insights into how to optimize your social content, and is practically impossible to get without deep analysis of large data sets.
With Artificial Intelligence saving time on strategic planning and reporting, marketers can focus more of their efforts on creating high quality content. However content marketing is difficult to scale, especially for smaller teams with limited resources. While AI is not at the point of being able to write award winning movies, there is some content that can be completely automated, and tools to aid marketers in creating better content at speed and scale.
Some of the most common automated content creators are chatbots and virtual assistants. Chatbots are services performed by AI that take place over chat interfaces. They have become commonplace in e-commerce sites, in customer service, and on messaging platforms like Facebook Messenger. Some examples include:
Weather bot: Get the weather when you ask
News bot: Find out when interesting things happen
Scheduling bot: Scheduling meetings, reschedule meetings, add participants, whatever you need to keep your calendar up to date.
Although one of the newest applications for AI generating full length publishing content making its way into the mainstream, and from a content scaling standpoint it is one of the most exciting. Machines are already totally automating content such as earnings reports, news headlines, and interview or webinar transcriptions. The most prominent tools in this space are Quill (which does all of the Forbes Earning Reports), and Wordsmith which operates in much the same way.
For content that can’t be completely automated, here are some tools to help make creating it quicker and easily tailoring it to specific audiences.
Visual Search: Use photos to search for similar photos to make creating posts easy, or fill out other assets to be more visually appealing.
Predictive Audiences: Advanced customer data that can predict Lifetime Value, probability of purchase, predict likelihood of customers taking action, alerts for high risk of churn, and more. In the context of social, predictive audiences can inform marketers what content audiences need to see at various stages of the sales funnel, and where best to deliver it to them
Content Recommendations: Want to know what colors, objects, keywords, and hashtags inspire your audience to take action? Tools like Foresight look at trends on your page, your competitor’s pages, and let you know what types of content resonate with your audience.
If you’ve made it this far you now have a deep understanding of what your audience wants, and you’ve created top notch content for them and are eagerly waiting to get it out there. However, to get the most out of all your hard work that content needs to be served to your audience at the right time, on the right channel.
Knowing how often to post and when is one of the most hotly debated topics among social media managers, with as “answers” as possibilities. The truth is that cadence and timing are different for every audience, and the only way to know definitively is to look at the data.
Luckily AI scheduling tools like Marketo, Marketing.AI, and of course Cortex all have specialized tools automatically generate a posting calendar based on audience and industry data.
Once all the content has been scheduled, the final question for marketers is how to allocate their promotional budget. Instead of dividing it up evenly by post or only promoting certain kinds of content, with machine learning you can look holistically at what posts are the most effective to boost based on content, timing, competition, and audience. Then, the computer will intelligently allocate your social spend to maximize performance for your budget.
Marketo and Cortex have this capability baked into their platforms, and Smart Insights, Clearmob, and Data Gran focus more exclusively on this functionality.
While social media marketers won’t be completely turning over the reins to machines anytime in the near future, social media managers who embrace AI and Machine learning tools have a massive competitive advantage over those who don’t. Most importantly, content marketers can finally apply the same level of scientific rigor to their work as their other business and engineering counterparts and generate huge value for their customers and their company in the process.
This article was republished with permission from Cortex.