Between the 2024 and 2025 General elections, the main parties increased their follower base by 20 per cent and the candidates by 55 per cent, with TikTok registering the largest relative growth. In 2025, the candidates will overtake the parties in total number of followers (2.5 vs 2.2 million). Chega and André Ventura lead in number of followers on all social networks, with the exception of X, where the Liberal Initiative leads among the parties.
As in previous General elections, Facebook was the social network where candidates posted the most. On TikTok, the number of posts tripled, reflecting the shift from three to six active accounts among the candidates. Among the parties, the PCP was the most active on the networks, followed by Chega and PS. Among the candidates, André Ventura posted the most, followed by Mariana Mortágua and Pedro Nuno Santos. Facebook led the way in terms of interactions generated (reactions, comments and shares), unlike in the 2024 legislative elections, where it was Instagram. Overall, TikTok was the most effective network, i.e. the one that generated the most interactions per post considering its follower base, registering strong growth in total interactions and surpassing X (formerly
Twitter).
Chega on Facebook generated 1 in 3 total party interactions and had the best performance per post, with an average of 6,500 interactions per post, six times more than PSD, PS, IL or BE. André Ventura had an average of 18,500 interactions per post, six times more than Mariana Mortágua or Luís Montenegro, and thirteen times more than Pedro Nuno Santos. André Ventura’s most viral posts coincided with his episode of indisposition in the final days of the campaign.
In terms of reach, Chega and Ventura are undisputed leaders, with Chega having an average of 57 per cent of the total views of party publications and Ventura 78 per cent. By comparison, Luís Montenegro’s and Pedro Nuno Santos’ posts together gathered only 10% of the total views among the candidates.
When it comes to posts by other social media users who refer to the candidates in their posts, X clearly emerges as the network of choice for political discussion, garnering more than 70 per cent of all posts on the four social networks that expressly mention the candidates’ names. But the situation is reversed when we look at interactions. In this case, it’s Facebook that gets the most attention from users (46 per cent), followed by X (30 per cent). It should be noted that, although residual in terms of the number of posts about candidates, Instagram and TikTok still manage to capture 18.4 per cent and 6.2 per cent of user interactions, respectively.
On Facebook, it is mainly posts from the main media and political parties that drive the discussion about the candidates. But on the other networks other dynamics emerge in the attention paid to political content that mentions party leaders. On Instagram, the media vie for leadership with other unconventional actors, while on X and TikTok it is clearly individual users and alternative media outlets that discuss candidate-related topics the most and capture the most user attention with these posts.
Analysing the posts and interactions about the candidates in the four weeks leading up to the election reveals an upward pattern: both the number of posts and interactions grew on practically all social networks until election day. But the effect of André Ventura’s health problems in the final days of the campaign is also notable. This had an impact on the number of posts and interactions on social media, particularly Facebook, Instagram and X.
Comparing the prevalence of narratives on corruption and immigration, we found that immigration became the dominant theme of the electoral debate on social media in 2025, far outstripping the visibility of corruption, which had been central to previous campaigns. Although the total number of posts on immigration grew exponentially, what was more relevant was the volume of views: almost 21 million, around five times more than for corruption, indicating a practically universal reach among Portuguese users on the networks.
The main narratives of these elections were promoted by actors from the radical right, feeding three major meta-narratives: the idea that “Portugal is being invaded” (including false claims about 2 million immigrants), the “Islamisation of Portugal” (through the instrumentalisation of the fake “Islamic Party” satire) and the narrative that “50 years of democratic rule have been 50 years of corruption”. These meta-narratives were amplified with disinformative content, exploiting fear, anger and mistrust, shared by political figures and pages with high reach.
During this period, some cases of coordination between anonymous accounts and networks were detected, but without evidence of automated control or external operations. Rather, a hybrid dynamic was identified, with national profiles – often anonymous or pseudonymous – operating in networks to amplify content aligned with the Chega party’s discourse. These accounts, with a strong overlap of followers and synchronised sharing, managed to reach hundreds of thousands of views at key moments, such as the day of refection.
The disinformation linked to the blackout also deserves a mention, as it occurred during the pre-election campaign and had obvious repercussions on it. Thus, the circulation of a text falsely attributed to CNN, about a Russian cyber-attack on the European electricity grid, revealed the ease with which fabricated content can gain traction on platforms such as WhatsApp. Subsequently, a report by the Israeli company Cyabra – widely reported but without due transparency in the methodology used – generated misunderstandings about the very presence of disinformation, by confusing anonymous accounts with fake accounts and fuelling misinterpretations about digital support for Chega.