Seasoned social media managers have, on one hand, content calendars and optimization tactics. On the other hand, they rely on experimentation to find hidden opportunities. "Best practices" don't exist for all digital marketing activities, like timing, for example. Self-proclaimed gurus will publish infographics and blog posts telling you the best times to post. But that's not how the world works. It takes experimentation to find when the best time to post is for YOUR target audience. There are many more examples like this. Experiment! Try something new! See what happens!
Marketing experts around the globe discuss data-driven decisions on one hand and experimentation on the other. For example, read what panelists at a marketing conference at Boston University had to say about that balance.
You have a few options for finding that balance. One method is to experiment first—take that "leap"—then measure, learn, and refine. Or, you can follow the scientific method: form a hypothesis and test that hypothesis. Whatever you choose to do, having a method gives you a framework for using experimentation and data hand-in-hand to optimize according to the metrics that matter to you.
One thing is sure: you can't plan or make data-driven decisions without some experimentation. Relying on others' data is like taking someone else's blood pressure to determine if you are healthy. Same goes for your business. If a 3pm posting time works for your competitor, that doesn't necessarily mean it will work for your business. Try 3pm, but also try 10am, noon, and 7pm. Try 1am to see if there are any night owls out there who may be checking relevant hashtags.