By now you should know that social media marketers aren’t only responsible for community management, content creation and building brand awareness. Social media data analysis is a huge part of the job. So much so that some companies are increasingly adding social media data analysts to fortify their marketing teams, answer questions about larger business objectives and optimize social media strategies and campaigns using the scores of data at their disposal.
However, in the 2020 Sprout Social Index, we found that social marketers aren’t necessarily using social data to its full potential. Measuring ROI, second only to identifying and reaching target audiences, is the top challenge social marketers face today. And yet, only 23% use social data to measure ROI. Even fewer (16%) use it for competitor insights, which can be critical in understanding your brand’s share of voice and optimizing the position of your brand in the marketplace.
Strong social media data analysis enables marketers to translate raw numbers and qualitative data into a story that answers the questions:
- What happened?
- Why did something happen?
- What is likely to happen?
- What action should be taken?
Regular reporting on KPIs helps you decide if your strategy is on track and performing as planned. But deeper analysis helps illuminate the greater impact of social media. In this article, we’ll outline how marketers can unleash their inner social media data analyst using Sprout’s Premium Analytics and social listening offerings.
If you haven’t tried Sprout’s reporting options yet, start a 30-day free trial to get hands-on experience.
Descriptive analytics: What happened?
Let’s start by talking about descriptive analytics, which is the interpretation of historical data to understand changes that have occurred. Put simply: what happened? Most commonly, social media marketers use descriptive analytics in their monthly, quarterly or yearly reports to answer questions like:
- What content was most engaging?
- How many followers did we gain?
- How many traffic referrals were driven by social?
- How do our metrics compare to the previous reporting period?
Data analysts are required to prepare, transcribe and “cleanse” data before analyzing it. This ensures that the data and answers you come up with are accurate and relevant. However, when you’re using Sprout Social, Premium Analytics takes on that tedious, time-consuming part of the process for marketers and displays clean, colorful, presentation-ready social media marketing data. The percentage change in metrics performance period-over-period is automatically calculated as well. If you prefer to work with raw numbers, build your own visualizations and do the math yourself, you always have the option to export data from Sprout as a CSV file.
Compare the performance of all your social platforms at once in a cross-channel report, or go platform by platform. In each Premium Analytics report, users can customize the Performance Summary and choose which of the available social media metrics to highlight.
Additionally, you can view charts that give further context around those high-level metrics. For example, if engagements are one of your KPIs for Facebook, Sprout gives you a full breakdown of each engagement type and maps out where spikes and drops occurred.
Looking closer: Why did something happen?
Figuring out “what happened,” is arguably the easy part of reporting. Identifying why something happened, also referred to as diagnostic analytics, is an opportunity to really hone your data analyst skills. It’s also essential to investigate the why before presenting any findings to leadership.
First, identify any anomalies or emerging trends you find in your data. For instance, was there an unexpected spike in impressions and engagements on a specific day? Was there a sudden change in traffic for no obvious reason?
The answer to the former question may be simple to come by if you know you’ve increased your posting cadence or you had a major campaign launch. But if it’s less obvious, look at the charts in Sprout and on the days you see anything out of the ordinary, cross-reference your Post Performance Report. With Premium Analytics, you can choose the metric by which you’d like to sort your post performance. Or, sort by date, so you can determine if the post published on the day you noticed the anomaly occurring is the one driving that change.
Also be aware if you’ve tried something new. For instance, if you typically always post with a photo, but posed a question to your community, just text, that blew up and garnered a massive response, take note.
If you can’t seem to determine why something happened, there might be something larger at play. Look beyond analytics to social listening.
Listen to gain a deeper understanding of your audience, industry and more
Brands should always be running a listening topic on brand health so you have a sense of how people think, talk and feel about your products, services and brand overall. This ensures that you’re tapped into the conversations that you aren’t necessarily tagged in.
With Sprout’s social listening tool, social media marketers have more qualitative data to inform their social performance. Sentiment analytics are a good place to start and provide a gauge on whether your audience feels positively, neutrally or negatively about your brand. If sentiment takes a nosedive, you can click to view those messages and get to the bottom of it.
Listening can help you understand how things outside of your control may be dictating your content performance. For example, we’ve seen the pandemic’s ongoing effects directly impact social media engagement. Continually use listening to track how world and local events, breaking news, industry trends, emerging crises and more impact your brand.
Predicting outcomes: What is likely to happen?
The ability to predict trends and future data is often dependent on past results. Each time you’re reporting on your data, take note of recurring patterns and the content types that consistently engage your audience.
Do you know that your audience loves GIFs? Are polls or community questions always slam dunk for engagement? Are Twitter users consistently driving more traffic to your website than other users on platforms? Using Sprout’s Tag Report, you can keep track of all these themes and considerations and use past performance to inform future content.
To form a prediction or forecast outcomes you will need to look at more than one data set. We did this recently at Sprout for our data report on retail trends for back-to-school 2020. We analyzed messages to and from 9,100 retailers’ social profiles across Facebook, Twitter, Instagram, Pinterest and LinkedIn and leveraged social listening for more insight into consumer behavior, expectations and concerns around back-to-school shopping.
Looking back at data from 2019, we were able to determine that June and July were the busiest months for retailers for both publishing and engagement, and thus, we predicted that the same would be true in 2020. We then used that insight to further investigate the kinds of content that retailers shared in those months. June and July aren’t the peak of back-to-school shopping months, but we did find that brands use those months to connect with their audience and lay the foundation for back-to-school campaigns.
Social listening is another way to predict trends. In the back-to-school example, we retrospectively looked at listening data to identify the biggest trends in 2019 to determine if they would carry over or change come 2020.
For example, we looked back on data from 2019 and found that social media users frequently mentioned the keywords “donate” or “donation” when talking about back-to-school and that those messages were highly engaging. With that precedence in mind, we then looked at data from the first half of 2020 and noticed an uptick in mentions of those same keywords. That data coupled with our understanding of the economic impact of COVID-19 led us to believe people were looking for ways to help others.
We predicted the trend around donations would continue as back-to-school season approached and made the recommendation for retailers to share philanthropic incentives and be transparent on social about where proceeds will go.
Turns out we were right! Since June 1, 2020, social media users mentioned those same keywords in relation to back-to-school over 18,700 times.
John Maynard Keynes once said, “It’s better to be roughly right than precisely wrong.” As long as you can back your predictions with real data, if they don’t come out exactly correct, that’s okay! It’s an opportunity to try, try again.
Now that you have all this data, you have the information you need to act and inform decisions about your social campaigns and content strategies. If you need to, package up your findings and use your storytelling skills to translate your data for leadership.
Sprout’s custom reporting options allow you to decide the order in which you present your data and story points. The Report Builder also has the functionality to add notes so you can provide further context, pull in any relevant listening findings and highlight the most important takeaways.
Turning social data into actionable insights should happen more than once a month. Regularly reflecting on your data, even if you aren’t doing a formal report, will help you continue to sharpen your quantitative and qualitative data analysis skills, and ultimately make a greater impact on your entire marketing team.
Learn more about Premium Analytics
This post Unlock success by embracing social media data analysis originally appeared on Sprout Social.