Increasing visibility on social media has become essential for any good brand that respects itself. For some of them, Twitter is a platform of choice. But with the rules of the game primarily influenced by social bots, how can you make the most of it? In particular, how can we increase the number of retweets? This is the question asked by two marketing researchers who published their results in the International Journal of Research in Marketing. This research is based on the analysis of tweets from major brands that have commercial content in common.
If you only have 30 seconds
- an American market research looked at the factors that influence retweeting
- including specific keywords at the beginning of the tweet increases the probability of retweeting
- the likelihood of a retweet is linked to the reader’s confirmation that the tweet belongs to his or her centres of interest. However, the time available to “scan” each tweet is minimal. It is, therefore, necessary to find a way to allow the reader to confirm his interest very quickly.
- the results show that keywords linked to promotions (free, promo, gifts, sales) are those that have the most significant influence on the probability of retweeting
- repeating keywords related to an event increases the likelihood of retweeting
- repeating call-to-actions reduces the possibility of retweeting
- the further the essential keywords are from the beginning of the tweet, the lower the probability of retweeting
- Can you compose your tweets to provoke retweets?
- Understand how we read information to compose our tweets properly.
- Write your tweets to simplify their understanding
- Results of the research (2 main results)
- Advice for your marketing strategy
The market research in question, published in December 2019, is based on a hypothesis: can the composition of a tweet, the way it is written, influence its ability to be retweeted?
L’hypothèse sous-jacente est que la composition du tweet influence la compréhension du sujet traité, et renforce en retour son attractivité auprès d’une cible particulière. Cette dernière a alors une probabilité plus élevée de le retweeter.
Can the composition of a tweet, the way it is written, influence its capacity to be retweeted?
It would be an illusion to think that we read in a disciplined way, following the lines, word by word. We scan documents, and one of the secrets to improving written communication is to think ahead about how the document will be “scanned” by the human eye and brain.
Reading a document (and therefore a tweet) is a two-step process:
- a quick scan called a saccade: during the whole saccade, words can be ignored. At this stage, the brain does not understand much, if any, of the meaning of what is written.
- a fixation: the eyes “focus” on the word(s) and the brain can process the information.
This jerky – fixation sequence is repeated to form the basis for a complete understanding of the document. Finally, it should be noted that a jerk lasts from 20 to 35 ms (milliseconds) and that fixation takes 150 to 500 ms. This understanding of the functioning of the human eye is made possible by research in eye-tracking. We have already discussed the use of eye-tracking in the field of advertising.
I invite you to watch the video below to better understand how we read.
Eye-tracking, therefore, teaches us that reading a tweet is similar to decoding in several stages. Jalali and Papatla’s research is, thus focused on understanding whether a tweet composed to simplify this process is more likely to be retweeted.
The reasoning is as follows:
- the reader tries, from the very first words, to determine the subject of the tweet
- on this basis, he then expects to find terms related to the topic he thinks he has detected
If the essential keywords are therefore placed at the top of the tweet, the reader can more easily confirm his interest and ensure, thanks to other related keywords, that the tweet is of interest to him.
The essential keywords in commercial tweets
- promotions: free, lucky, sale, commercial, gift, promo, win
- brands: the name of one of the 62 brands studied
- events: game, event, holiday, NCAA, Valentine’s Day, (super) bowl, spring break, St. Patrick’s Day
- time: Saturday, Sunday, Monday, Tuesday, Wednesday, Thursday, Friday, morning, day, night, today, now, year, week, weekend, tomorrow, evening
- call-to-action (CTA): click, watch, join, vote
In order not to burden you with technical details, I will go straight to the conclusions. There are 5 main ones.
- The keywords linked to the “promotions” category are the ones that have the most significant effect on retweets (4x higher than the keywords related to the “brand” category).
- Time or call-to-action keywords are half as effective as keywords related to promotions.
- Repetition of particular keywords can be useful in provoking retweets. Event-related keywords are worth repeating as they are 5 times more effective in causing retweets than brand-related keywords.
- On the other hand, do not abuse call to action. Their repetition reduces the probability of retweets.
- The effectiveness of the different categories of keywords to provoke a retweet decreases with distance. In other words, the further away the keywords are from the beginning of the tweet, the less retweeting they will cause.
User satisfaction on Twitter is, therefore achieved through concise messages and themes that are identified at the beginning of the tweet.
Time is a scarce resource. Attention is even more so in our digital society. Using social networks to communicate, forces us to continually question how to make our messages as effective as possible. User satisfaction on Twitter therefore requires concise messages and identified themes from the very beginning of the tweet.
To make communication on Twitter more effective, you need to analyse your tweets beforehand to;
- to isolate the keywords that have brought you the most retweets
- to include these keywords in the right place in your future tweets
In light of these results, it is also legitimate to question how to communicate on other social networks. LinkedIn is my favourite, but it must be said that the functioning of the algorithm remains quite mysterious. Recent changes have introduced the notion of “dwell time”. The “dwell time” is the direct consequence of the user’s interest in a subject. If the initial scan is positive, the user will take the time (dwell time) to read the topic entirely. The dwell time will, therefore, be positively affected, and the virality will be positively affected.
If you are interested in this topic and want more information (or simply want to think with me about it), contact me!
Illustration : shutterstockTags: consumer behavior, social media