Statistics What marketeers tend to forget May 28, 20144 minutes Geir There is a lot of talk about the value of social media. Not only what value is, but how to measure it. Experts, bloggers, analysts and random people with a Twitter account seem to jump into the ROI discussion with their opinions and formulas. How much is a like worth? What’s the value of a page post? Why is my organic reach screwed? What if I told you that the entire discussion is based on false premises? It is. 24SevenSocial decided to dig more into the life of page posts. We manually analyzed the posts of lots of pages, and manually tagged them into categories. Then we analyzed the key metrics with the help from statistical experts from Østfold University College. The page post are all from May 1, 2012 to December 31, 2013 and range from smaller to bigger pages. All pages are Norwegian medium to large brands with a decent marketing budget. From retailers to online shops and media companies. The results? The start of a new theory of ROI in social media. Before I present some of the key findings, I have to put Facebook marketing and page posts into perspective. In general, when communicating, brands have always and will always follow the same funnel. Brands have to Get people to talk to.. ..engage with them.. ..and convert them to a given goal. This gives us three main individual communication objectives. All Facebook page posts may be divided using these three groups (in our study we also used sub groups, but let’s simplify it and stick to the main groups). Key findings The standard deviation (individual differences) on Facebook page posts is high. This means the results vary a lot. A lot! But we still found some statistically significant results. The objective of a Facebook post has a major impact on expected reach, engagement and negative feedback. Facebook page posts with only one objective give more effect than trying to address two or more objectives. 49,35 % of the posts had conversion as main objective, 27,14% engagement and 23,51 % branding. The conversion-posts with best effect, were those that followed engagement posts with above average engagement % and fewer previous conversion posts. Engagement posts have 1,705x the reach compared to branding posts and 1,214x compared to conversion posts, but also 1,293x and 1,124x the negative feedback. Average frequency of an organic post is 2,54 (branding) , 2,66 (engagement) and 2,46 (conversion). (This has changed over time, I will come back to that.) Key learnings Even though there are big individual differences both on page, time and post level, there are certain learnings the analysis tells us with small error margin: Have one objective per page post. If you have multiple objectives, create more posts. Categorise your page posts based on their objective (branding, engagement, conversion) Based on the objectives, use a separate set of KPIs for each group of page posts. (Benchmark engagement posts against other engagement posts – not conversion post) Make sure you have a good mix of posts from all categories. As a rule of thumb, do not have more than 20 % conversion posts. There is no one single Facebook KPI. You need both branding, engagement and conversions. They should all be measured with their own KPIs, but with the understanding that they have an effect on the others. Thus they should also be seen as a sum of activities strongly related to each other. This means that the “what is ROI of social and Facebook”-discussion today is oversimplified trying to find one golden metric, but still overcomplicated in the hunt for the magic formula. ps. This blog post is way shorter than it should have been. I know. But it’s a compromise trying to please both the inner geek and the outer marketing dude. Reach out to me for more in-depth geekiness and ask questions below.