Everything we need to know about Twitter
We all leave our footprints by working on social networks every day. The hours we spend on social media, the individual pages we secretly visit, everything we write, republish and like are all part of the digital footprint we leave behind.
Technology companies that own these applications have a set of data that can well represent our beliefs, beliefs, values, and patterns of behavior. Some believe that by analyzing this digital footprint, we can know ourselves better than ourselves and even predict the psychological characteristics of each person. For example, in the Facebook-Cambridge Analytica scandal, Cambridge Analytica analyzed the psychological data of 87 million Facebook profiles and analyzed the pages they “liked” during the 2016 US election. In accordance with the psychological characteristics and psychological types of each person, he designed an election campaign related to that characteristic. No matter how influential the Facebook-Cambridge Analytica affair was in Donald Trump’s victory in the 2016 US election, the fact that people can be identified by the digital footprint they leave behind, including through their likes, cannot be overstated. In addition, analyzing the collective behavior of social network users in liking content can give us new information about the mechanisms governing these networks. For example, someone who wants to launch a successful online campaign needs to know what the governing mechanism is for the success of a Twitter storm, and what features the tweets posted in this storm should contain.
What is the method of analysis? What is the analyzed data?
In order to collect the likes of some popular political users on Twitter, the likes of the tweets of 18 user accounts with different political tendencies, including revolutionary and fundamentalist, reformist and revolutionary, in the period from 16 Azar to 19 December of this year, were collected by data mining methods. Was collected. These 951 tweets received a total of more than 1,148,000 likes from 120,742 unique users. This large amount of collected data, ie more than one million and 148 thousand likes, provides us with valuable and reliable findings that reduce errors in conclusions and increase reliability.
What do the likes say about the reasons for a tweet being viewed?
In the process of collecting likes, the time of registration of each like was also recorded. The analysis of the recorded time for each like provided us with interesting findings; Among other things, each tweet has a “golden time” after it is published, which can attract the most attention, and most of the likes of a tweet are for this time period. Our findings show that most likes of a tweet are recorded in the first 12 hours after it is published. This means that 12 hours after the tweet was published, attention to that tweet will probably subside. The chart clearly shows this fact. This chart has been obtained after analyzing the registration time of each of the collected likes (more than one million and 148 thousand likes). The horizontal axis indicates the number of hours after the tweet was published, and the vertical axis indicates what percentage of the total likes of a tweet belong to this time. As you can see, most of the likes of a tweet are related to the early hours after it was published, and almost 36 hours after the tweet was published, that tweet was rarely liked. Many of the tweets reviewed also gained all their likes in the early hours. The first line of the table above shows that each tweet received, on average, a percentage of its total likes in each of these time periods. As it turns out, an average of 95.95% of the total likes of each tweet belong to the first 24 hours after it was published. The second line also shows that in the worst and latest case, a tweet gets a few percent of its likes during this time. For example, a tweet in the first 6 hours after publication gets at least 25% of its likes in the worst case, but on average each tweet will get 67% of its likes in the first 6 hours. These findings show that the early hours after a tweet is published are very important in the success and visibility of that tweet; Because a significant portion of the likes of that tweet are gained in the early hours, and if those early hours are lost for any reason, the tweet, no matter how good, will probably not be seen. This finding is also consistent with what we saw in the trend of the “do not execute” hashtag. Where, just 6 hours after the first tweet containing this hashtag was published, more than 2.5 million likes and 638,000 retweets were allocated to the tweets containing this hashtag, which eventually caused this hashtag to become more global. So every tweet and every Twitter campaign and storm to be seen has to take one fact seriously: the early hours are important.
What time do we tweet?
As you read this note, you may think that what has been said so far is obvious. It is clear that a tweet gets the most likes in the early hours after publication. But based on the fact that our data also confirmed it, the important question that arises is what time should I tweet? The collection of hours of likes helped us to know at what time of the day the most likes were registered; Findings that help us figure out what hours of the day a tweet is most likely to be liked, the results of which are shown in Figure 2. As you can see in this chart, the highest number of likes was recorded around 11 pm. After that, the most likes were recorded at 10 pm, zero in the morning and 9 pm, respectively. So if you want your tweet to be seen better, if you tweet at 11pm, you will probably be more successful than when you tweet at 9pm; In a way, it can be said that the golden interval for liking Twitter is from 10:00 to 12:00. The above findings have been proven experimentally for each Twitter activist, but examining the different hours in the chart above gives us more interesting results. For example, around one o’clock in the morning, more likes were recorded than at 5 or 6 p.m. Likes that are probably related to the tweet recorded in the last hours of the day. This finding has an important meaning. If you tweet at 12 pm (zero in the morning), your tweet will probably get more likes than 5 pm! The fact is that many Twitter users are mostly up until one o’clock in the morning. These findings, along with the fact that a tweet gets the most likes in its early hours, provide us with a simple guide that helps us have a better strategy for publishing our tweets. This means that a tweet with the best content posted at 5pm is less likely to be seen than a tweet with mediocre content posted at 10pm. So in addition to paying attention to the content, pay attention to the time of publication of the tweet, and if you want your tweet to stand out, do your best to republish it in the first few hours.