UPDATE 2: 9 March 2013
See also our analysis about the statistics of all @Pontifex tweets, plus the content classification of each tweet, in our sister site:
UPDATE: 12 December, 11h39 (CET)
This is the first tweet by Pope Benedict
Then came the first tweets with religious content
The Vatican decided to open an official Twitter account for Pope Benedict XVI for spreading Roman Catholic Church message through the social media, aiming ‘to reach specially young people’.
Pope Benedict twitter account was activated by 3 December 2012. The reference account is @Pontifex (Pontiff, in latin, which refers to the Roman Pontiff, The Pope). Tweets in @Pontifex account will be published in English. It has been launched in eight different languages: English, Spanish, Italian, Portugues, French, German, Polish and Arab.
Spreading the Message: The Messsage, The Followers And The Active Followers
The launching of this new Twitter account has received wide media coverage. As the leading authority of the single most followed religious confession, his active presence in the social media raises attention among social media analysis about the capacity of @Pontifex to attract millions of Twitter followers.
Dalia Lama is currently the religious authority counting with more followers in Twitter. (
@DalaiLama account, whose Twitter bio says ‘Welcome to the official twitter page of the Office of His Holiness the 14th Dalai Lama.’ counts with more than 5.6 million followers by the beginning of December 2012). Tweets by @DalaiLama containing direct messages from Dalia Lama reach an average number of some 13,000 RT (re-tweeted by people following the account to their own Twitter followers). RT allows the content from @DalaiLama to be spread outside the direct circle of people following the Dalai Lama.
So, number of followers of the different @Pontifex Twitter accounts is a critical element in allowing the message from Pope Benedict to reach the social media audience. The next step will depend on how much people following @Pontifex retweet Pope’s messages.
The answer to those questions will start to emerge by 12 December 2012, the day Pope Benedict will publish his first personal 140 character message.
But even before the first tweet is published, many information may already be collected by just monitoring the dynamics of the people starting following @Pontifex.
We propose in this post to present some metrics concerning the first couple of days of life of @Pontifex that, in our understanding, tell us several things not only about the Roman Catholic Church, but also about Twitter dynamics.
This case is somehow one in a kind, and offers a perfect framework for study. This new Twitter account corresponds to a very relevant public personality, counting with many million of (off-line) followers, but also with groups of critics to his religious message. This ensures that if this new channel of communication and conversation is wisely used by Vatican authorities, it may have a serious impact in the social media conversation. We can monitor its impact and evolution since the inception of the account. But this case is also special because it launches eight language accounts. This very rare movement in Twitter will be a high value source of information about Twitter and Catholic Church demographics.
Basic Facts about @Pontifex followers between 3 and 7 December 2012
First result concerns the evolution of the people following any of the 8 accounts of the family @Pontifex.
- Reached 100,000 followers after 3 hours
- Reached 250,000 followers after 8 hours
- Reached 500,000 followers after 24 hours
- Reached 750,000 followers after 3 days
Next figure shows the evolution of the total number of followers, per hour.
In the next figure, we present the evolution of new followers, by blocks of four hours. During the first day, it gained more than 100,000 new followers every 4 hours. It dropped to a path of some 40,000 new followers the day after. Between 5 and 7 December, the number of new followers every four hours oscillates around 10,000 followers.
We can present the same result by days. Pope Benedict got some 380,000 followers by 3 December (CET), 244,000 in December 4, 105,000 in Dec 5, 51,000 in Dec 6 and 38,000 in Dec 7. This means, as reflected in next graph, that in percentage value, Pope Benedict attracted 46.4% of all followers within first 12 hours. 76.2% of all followers were added during the first two days.
Eight Twitter Language accounts. Dynamics Analysis
As mentioned before, the almost unique initiative of lanuching as much as eight twin language accounts will offer a very rich source of quantitative analysis about Twitter behaviour. We present now the initial results we can obtain even before the first tweet.
The following figure shows the distribution of total followers among the different language accounts.
Twitter account in English is by far the leading source of followers, as it attracts more than two thirds of all Pope Benedict Twitter followers. Second most important account is @Pontifex_es, for followers in Spanish. Account in Italian is also a relevant source of followers, as it represents some 9% of all followers. The other five accounts play currently a marginal role, as their combined followers are just 7.7% of all followers.
How stable are these figures? We show in the following graphs the time evolution by hour of the share of some language accounts, in order to capture some trends in this initial stage of life of the Twitter accounts.
We find, concerning followers in English that it is observing a decreasing power since the peak of a 70% share reached in the aftermath of the launching of the account. We will show later that this trend is probably provoked by the fact that @Pontifex plays a role of reference account. This makes that some people writting in some of the other languages with their own account follow the one in English thinking that this is the only existing one.
Concerning the two other main languages, we find an opposite trend, as reflected in the following figure. Account in Spanish is loosing consistently power, moving from initial 17-18% to a mere 15% three days after. Account in Italian reached a record low of some 5.5% at the beginning, but a couple of days after is approaching the 9% mark. Our data suggest that Italian may become soon the second most important account.
Finally, as for the minority language accounts in terms of followers currently, we have that the weight of followers in Portuguese is somehow decreasing (from 2.7% to 2.5%), while both the size of followers in French and German tend to increase.
This time analysis says us that the current share of followers by language is not yet stable. Of course, the number of followers by language will be also highly infuenced after the first tweets are published. This is because people in different languages present a different degree of participation and reaction through tools like RT. And the number of RT disseminated is a key factor in attracting new followers.
We provide in the next two figures the additional information that allow us to expect that account in Italian might become the second most popular one. We showed just before that Italian followers increase steadily their weight, against followers in Spanish. The present gap is nevertheless substantial. This is due to the fact that reaction during the first day was totally favourable to followers in Spanish. We have shown that almost half of current followers were obtained during that first day. But trend below shows that from the third day onwards, new Italian followers are always more numerous than followers in Spanish. If maintained, this trend would put @Pontifex_it as number two by number of followers.
In the next figure we apply the same exercice to the accounts in French, German and Portuguese. Here again initial reaction was completely dominated by followers in Portuguese, but this daily advance was short lived. Since December 6, accounts in French and German receive more followers than the Portuguese account.
The Timing of New Followers
We show in this section the dynamics of new followers within a day. ¿Which are the periods where the Pope receives more new followers within a 24 hours time span?
The answer to this question provides a complement of information ahead from the relative power of each language account. This new data will provide insights about the regions of the globe that generate more Twitter followers.
This is the results, by segments of four hours, based in CET (Paris) hour.
The same information is presented in the following figure. But this time we just count wth six blocks of four hours. It shows how many new followers are gained within each block as percentage of all Twitter followers.
Peak period is between 12h and 16h (CET, Central European Time, UTC+1), as it concrentrates 27% of all followers. The next period 16h to 20h is almost equially relevant: 24%. By contrast, the weakest 4-hour block are between 4h and 12h, where just 7% new followers come in each period.
Twelve consecutive hours between 12h and 24h provide 74% of all followers. So, between 0 and 12h, only 26% of all followers are obtanied.
The period of high traffic of new followers (12 to 24h CET) corresponds to the period of activity of Europe, Africa and American Continent. 12am in Rome is 6am in New York. Low period of activity (0 to 12h CET) corresponds to period of activity of Asia and Oceania, and the beginning of activity in Europe.
Monitoring the time of the day of new followers allows us to measure the importance of the regions of the world in following the Catholic Leader. Considering that Twitter followers from African countries is marginal (as we will show later), we find that Europe and American Continent provide the main source of Twitter followers, whatever the language they choose.
As we have the global picture as reference, we can now study the profile of each language account.
We find that the profile of new followers at @Pontifex in English follow a rather similar path that the combination of all other language accounts. The share of new followers in English is slighly higher than average in the segment 0-8h, the moment of activity in Oceania and Asia.
In the next figure we show the behaviour of the two other main language accounts, Spanish and Italian. Differences in behaviour are here stronger. New followers in the Italian account are much more strongly concentrated in the European segment 8-24h (87% of all followers) than the account in Spanish (76.1% in that segment). Spanish account benefits from its international reach, and new followers come also in the period 0-4h, which is very active in the American Continent.
The account in Portuguese (@Pontifex_pt) follows the same dymanics than the one in Spanish, showing the relevance of the followers from Brazil
By contrast, the accounts in French (@Pontifex_fr), German (@Pontifex_de) and Polish (@Pontifex_pl) follow the same time profile than the Italian one. This is not surprising for the German and Polish cases, as they are European languages like Italian. But it shows that the core number of followers from the French account come mainly from Europe, and not from French speaking countries in America.
Pope Benedict Twitter Followers by Language: Does It Reflect Catholic Presence in Countries or Twitter Presence?
As explained at the beginning of this analysis, one of the attractiveness of this case is that vatican decided to launch eight different language accounts. We have seen that, as expected, the one taking more followers is the one in English, with a share of 68% of all followers. The other two more relevant languages are Spanish (15.5% share) and Italian (8.8%).
These results are quite in line with the share of Catholic followers in the world. At the same time, not all Catholic people count with a Twitter account allowing them to appear as @Pontifex follower. Reports say that there are some 500 million people owning a Twitter account in the world, bu mid 2012. Furthermore, the popularity and diffusion of Twitter is no equal in all countries. So, we can have some countries with many Roman Catholic followers but not following Pope Benedict in Twitter simply because there are few in that country using Twitter as social media chanel.
We will explore in this section to which extent current data on @Pontifex followers reflect both Catholic world demographics and Twitter world geographical demographics.
We present first the list of top 20 countries by number of Twitter followers, as provided by Semiocast, with data by July 2012. In order to provide an insight of the power of Twitter as social media tool in each country, we compare them with the number of Facebook users. Data about Facebook users is provided by Socialbakers.com.
Our aim in this post is to provide just some insights about Twitter in order to understand the results about @Pontifex. Therefore, we will not comment in detail the results we present now about Twitter.
Twitter main market is United States, with some 142 million users. This figure is not far away from Facebook values. Second most important country for Twitter is Brazil, while India is the second one for Facebook. It can be observed that the presence in each market are not similar fro both social media chanels. Linking the results to the different Pope Benedict Twitter accounts, we have that Twitter is apparently strong (by comparison to Facebook) in markets in English like United States and United Kingdom, but it is weak in other markets in English like India and The Philippines. It is also powerful in the Portuguese market of Brazil. In Spanish, Twitter is strong in Venezuela, but weak in Spain, Mexico, Colombia and Argentina. Finally, Twitter is weak in the other languages chosen by the Vatican: France, Germany and Italy. We have no data about Poland.
Which are the key factors behind the number of Twitter users in each country?
The natural answer would be the population of the country. Of course, there is a positive relation between the number of inhabitants in a country and the number of Twitter users. But this relationship is in fact rather weak. We have plotted in the next figure the relationship between both variables for the top 20 countries in terms of Twitter users. The more populates is a country, the more the number of Twitter users in general. But there are many exceptions (big countries with rather few number of Twitter users) that weakens the repationship.
The second natural option to explore is the standard of living, measured by the GDP per capita. It shourl be expected that the more economically developped a country is, the higher the penetration ratio of Twitter users. Twitter access requires computers or smartphone devices, plus access to internet. Both the access to technological infrastructures and the purchasing power to use the services should clearly be linked to the number of Twitter users.
But the figure below showing the relationship between Twitter users and GDP per capita shows again a very weak positive relation.
So, where to find a variable which is really strongly related with the number of Twitter users by country? Paradoxically, the best existing variable to explain the number of Twitter users is the combination of the two variables presented: GDP per capita times the population. This is total GDP of a country, the economic production or the size of the economy of a country.
We find in the figure below that the number of Twitter users can be explained by the size of the economy of each country. In fact, this relationship is stronger for Twitter than for Facebook. Facebook, with is a much more mature and extended social media chanel, has a stronger relationship with population by country than what happnes with Twitter.
So, as GDP is the key factor explaining Twitter users, we will use this information to try to explain the behaviour of the number followers by language of Pope Benedict in Twitter: is it driven by the Catholic population from each language, or is it simply a reflect of Twitter users in each country?
(new content coming)