How Small Business Owners Can A/B Test for the Best Time to Post
Social media can be a fickle beast, since its relative value is entirely determined by the response to what you do and there are no real rules for posting, only guidelines and common sense to help business owners fine tune. Once they are in the swing of posting content, recycling pieces that could be better timed and engaging with other users, then you can begin to compile your own best practices based on the responses that you are getting. One factor that can effect the response that posts you put out receive is their timing. Timing will have a direct effect on where your posts appear in the various feeds of your consumers. Depending on whether you post five minutes before a customer logs in or five hours before, they may or may not ever wind up coming across your post in the first place. That’s where you should engage in some simple testing in order to determine when your content gets the most favorable results.
Create a few pieces of content that are of similar length and on similar topics. While you should avoid creating and posting duplicate content, for the purposes of testing to determine what the timing of your posts has to do with engagement, you should attempt to control for factors like the inherent level of interest customer might have in the piece. A clearly superior piece of content can skew your results and dilute their value.
Extrapolate from what you know about your customers when you think it is likely that they will be online. Depending on the demographic that you are looking to engage with, you may determine a range that you think represents the prime times that these people will be online. From there, it is time to test your theory. Schedule your posts to occur on the same day of the week, with one or two posting times being tested a week. After a few weeks, you can compare the times and their relative engagement levels. Be aware that the value of the data you collect will increase the more testing that you able to engage in, so don’t pronounce something as an absolute truth after only comparing a few results.
Keep experimenting with timing in order to get it down to a science. As already stated, the larger a data pool becomes, the better it can be relied upon as a predictor. Still, experimenting with your timing and putting articles out there in order to get a better sense of what your customers respond to is a good supplement to A/B testing, and over time you begin to see patterns emerge related to the types of content that they prefer and when they want to consume it.
Photo Credit to Bart Hiddink on Flickr