From campaign to post-election cliffhanger: what five weeks of parties’ Facebook posts tell us about election strategies in 2026
In its final report on the 2026 election, the New Zealand Social Media Study looks back at parties’ Facebook posts in one of the most negative election campaigns of the past decade.
Over the past five weeks, the 2026 New Zealand Social Media Study has analysed 2,780 Facebook posts by political parties and their leaders to find out what they reveal about online campaigning during this year’s election.
The posts, published from 5 October to 6 November, were analysed by researchers at the Internet, Social Media and Politics Research Lab, led by Dr Mona Krewel.
Results show Facebook posts were up this election, but one major party seemed to have “missed the memo” about the key topics worrying voters. In one bright spot, disinformation in parties’ posts dropped with only a single instance of fake news flagged by the research team.
More posts by most
Overall, Facebook campaigning in the 2026 election increased slightly compared with 2023.
There was a 3.5 percent rise in the number of posts by parties and party leaders contesting both elections and gaining parliamentary representation in at least one of them.
“All up, 2,363 posts were published during the 2026 campaign by these parties and their leaders. This compares with 2,282 posts in 2023,” says Dr Krewel.
National was the most active Facebook campaigner in this year’s campaign, maintaining its lead in posting activity from the first week of the study through to the final day before the election.
“National's dominance in terms of Facebook activity was remarkably consistent throughout the campaign. From the first week until election eve, National was the party publishing the most content.
“Of course, posting more frequently does not automatically translate into greater electoral success. The unresolved election outcome is a useful reminder that social media activity and electoral performance are two very different things,” she says.
While most parties increased their Facebook activity this year, there were two notable exceptions.
“The most striking was ACT and leader David Seymour. They dramatically reduced their Facebook output from a combined 947 posts in 2023 to just 275 in 2026, a decline of 71 percent,” she says.
This reduction can partly be explained by a shift towards Instagram, where ACT increased its posting activity from 789 posts over the comparable period in 2023 to 1,100 in 2026.
“However, across the two platforms ACT still published 361 fewer posts than in 2023, a drop of about 21 percent. ACT's decline in social media activity stands out because it goes against the broader trend towards more intensive social media campaigning,” says Dr Krewel.
The other exception was The Opportunity Party (TOP).
“TOP published the same number of Facebook posts in both election campaigns, an unusual coincidence given the changes in its electoral fortunes and leadership.”
Analysis also showed there was a small shift in the balance of posts published on parties’ Facebook pages versus posts on party leaders’ pages.
“This year, party accounts contributed a slightly greater share of overall posts compared with 2023. However, these differences were relatively small so should not be given too much weight,” she says.
Figure 1: Overall number of Facebook posts by main parties in 2023 vs 2026

Voters worry about the cost of living; parties talk about the economy in general
The economy and finance were dominant issues in parties’ Facebook communication during this election campaign, featuring in 41.3 percent (1,147) of all posts.
Labour and social issues followed in a distant second place (21.4 percent or 594 posts), with the cost-of-living ranking third (19.2 percent or 535 posts).
Healthcare featured in 14.7 percent of posts (408), followed by domestic politics (11.1 percent or 308 posts), and education and culture (8.3 percent or 231 posts).
“It looks a little like the parties missed the memo about what voters were actually concerned about. Survey research routinely identified the cost of living as voters' number one concern. Yet in parties' Facebook communication, this issue ranked third.
“Broader and much more general economic discourse dominated by a considerable margin. Parties didn’t really manage to break down the economic debate to the household finances of the average Kiwi,” says Dr Krewel.
According to Vote Compass data, the economy was the second most important issue for younger voters, while older voters ranked healthcare second. The Ipsos Issues Monitor also identified healthcare as New Zealanders' second-biggest concern. However, healthcare ranked only fourth in parties' Facebook communication.
Education, the third most important issue for younger voters according to Vote Compass, received even less attention.
“There is clearly some overlap between what parties talk about and what voters care about, but the match is far from perfect. Healthcare and education are good examples. Both feature prominently among voters' concerns, yet they receive considerably less attention in parties' Facebook posts,” she says.
However, digging into the data shows there were marked differences in approach between the two main parties.
“The strong dominance of the economy and finance category was largely driven by National's Facebook campaign, which placed particular emphasis on the economy in general and businesses. In contrast, Labour focused much more heavily on the cost of living, which was its most frequently discussed issue.
“This distinction matters because the overall figures can give the impression that all parties were campaigning on the same issues. In reality, National and Labour approached economic concerns from quite different angles.
“While National emphasised the broader economic situation and businesses, Labour concentrated more directly on cost-of-living pressures facing households. Labour's communication was therefore more closely aligned with what voters identified as their most pressing concern.”
Figure 2: Percentage of Facebook posts on key issues in the 2026 election campaign

Parties ramped up mobilisation but largely overlooked overseas voters
As election day approached, political parties increasingly used Facebook to mobilise voters.
This increase was particularly pronounced in the final two weeks, with calls to vote accelerating around 26 October when advance voting began. A smaller increase was also visible around the start of overseas voting on 21 October.
“The parties clearly had the key dates of the election on their radar and adjusted their Facebook communication accordingly. As election day drew closer, their communication increasingly shifted towards getting people to the ballot box.
“By the end of the campaign, almost three out of four Facebook posts included a call to vote. That is a substantial increase compared with the beginning of the campaign where just a quarter of posts did this,” says Dr Krewel.
Across the campaign, most of these posts (1,255) targeted domestic voters. Only 10 specifically encouraged New Zealanders living overseas to vote. In a small number of additional posts, the intended audience could not be clearly identified.
“Given the election cliffhanger we are now facing, the lack of attention to overseas voters is particularly striking. Overseas votes form part of the special votes that could determine the final election outcome. Yet throughout the campaign, parties made remarkably little effort to mobilise these voters through their regular Facebook posts,” she says.
However, there is an important caveat.
“Our study examined Facebook posts published by political parties and their leaders, rather than their paid political advertisements, which may have targeted overseas voters. We cannot rule out that parties were reaching overseas voters mostly through paid advertising.
“But regular Facebook posts are free, whereas advertising costs money. Even if parties were running targeted ads, they still missed a chance to reach overseas voters through their organic social media communication at no additional advertising cost. In an election that may ultimately be decided by special votes, that must be considered an opportunity left on the table.”
Figure 3: Percentage of posts with a call to vote (5-day moving average)

Leaders in and out of the Facebook spotlight
The Opportunity Party placed its leader at the centre of its Facebook campaign more frequently than other parties chose to showcase their leaders. TOP’s Qiulae Wong featured in 34.6 percent (36) of the party’s posts.
New Zealand First was almost as keen to put its leader on show, with Winston Peters appearing in about 29 percent (56) of the party’s posts.
The rest of the field trailed behind and were much less likely to make their leaders the focus of posts.
“What I find particularly interesting is how little the two Chrises featured in their parties' campaigns. After all, National’s Christopher Luxon and Labour’s Chris Hipkins were the two politicians competing for the job of prime minister.
“You would expect them to be much more prominent in their parties' election communication. The fact that Luxon, as the incumbent PM, featured in only about 7 percent (35) of National's Facebook posts is remarkable.
“That said, there are good reasons why smaller parties tend to run more leader-centred campaigns. Their leaders are often their most recognisable public faces, while voters may know relatively little about the rest of their candidates. National and Labour have many more established political personalities to call on.
“Nevertheless, the results suggest that neither of the two major parties saw its leader as a particular voter magnet, at least judging by how prominently their own parties chose to feature them,” says Dr Krewel.
Figure 4: Percentage of Facebook posts by party that mention party leader

Māori politics absent from major parties' Facebook campaigns
While all parties used some te reo Māori in their Facebook communication during the 2026 election campaign, there were substantial differences in how prominently the language featured.
“At one end of the spectrum was Te Pāti Māori, which incorporated te reo Māori into an overwhelming 95.4 percent (226) of its Facebook posts. At the other end was National, which used te reo Māori in just 8.1 percent (58) of posts. Labour performed marginally better, at 9.7 percent (or 50 posts).
“Apart from Te Pāti Māori, the Greens were the only parliamentary party to use te reo Māori in a substantial proportion of Facebook communication, with the language appearing in 52.6 percent (102) of posts. However, even this was considerably below the level recorded for Te Pāti Māori,” says Dr Krewel.
There were also marked differences in how frequently parties addressed Māori political issues.
“Te Pāti Māori clearly demonstrated its role as an advocacy party for Māori rights and interests, addressing Māori political issues in 71.7 percent (170) of Facebook posts. The Greens ranked second, but at just 8.8 percent (17 posts) were a considerable distance behind.
“What is particularly remarkable is how little attention Māori politics received from the two major parties. Both Labour and National virtually ignored Māori political issues in their Facebook communication. ACT (2.9 percent or 8 posts) and NZ First (3.1 percent or 9 posts) addressed Māori political issues, although not often.
“We haven’t measured whether specific issues were discussed in a supportive or critical way. But given the very different political positions of Te Pāti Māori and the Greens on the one hand, and ACT and NZ First on the other, we would expect their discussions of Māori politics to look quite different,” she says.
Figure 5: Percentage of posts by parties and their leaders that used te reo and covered Māori issues

Disinformation almost disappears from parties' Facebook campaigns
Despite widespread concerns about disinformation in election campaigns, the study found little evidence of disinformation in parties' Facebook communication.
“The proportion of posts containing either fake news or half-truths fluctuated between about 1 to nearly 3 percent, with a minor downward trend towards election day. Importantly, we only identified a single instance of fake news during the entire 2026 campaign,” says Dr Krewel.
This instance involved National's claim that more than 2,100 additional nurses had been employed. The figure included part-time nurses and attributed employment increases to changes the government had made even though much of the recruitment effort had begun before it took office.
Compared with the 2023 election campaign, there was less disinformation in posts.
“In 2023, the proportion of posts containing fake news fluctuated between 0.7 and 3.8 percent. Half-truths were more common, reaching approximately 6.5 percent towards the end of the campaign. In 2026, by contrast, the combined proportion of fake news and half-truths remained below 3 percent throughout the campaign,” she says.
There is a rider to this finding. The study examines posts published by political parties and their leaders, not posts by unofficial political groups or other actors. Analysis also does not capture disinformation spread through paid political advertising.
“We can’t generalise our results to the entire online information environment. Disinformation may still circulate widely through other accounts and channels. But our findings do not support the idea that political parties themselves were flooding Facebook with disinformation during the 2026 election campaign.
“Quite the opposite: based on what we can observe in their official Facebook communication, outright falsehoods were rare. This might have been an extremely negative election campaign that was often perceived as unfair but it wasn’t an extremely false campaign.
“And we should be careful not to confuse negative campaigning with disinformation. These are different concepts that often go hand-in-hand but they do not necessarily always co-occur and should not be conflated or considered interchangeable,” she says.
Dr Krewel attributes much of the decline in disinformation in 2026 to the decreasing activity of fringe parties on the right of the political spectrum.
“These parties were important sources of disinformation in previous election campaigns, but their presence and activity on social media have declined considerably.”
Figure 6: Percentage of posts with fake news or half-truths (5-day moving average)

About the New Zealand Social Media Study
The New Zealand Social Media Study was run in both the 2020 and 2023 elections, and ran again in 2026. The study tracks Facebook posts by political parties and party leaders in the five weeks leading up to the general election. It sources Facebook data from the Meta Content Library.
Nine parties are included in the 2026 study:
- six parties currently in parliament—National, Labour, Green Party, ACT, New Zealand First (NZF), and Te Pāti Māori (TPM)
- three parties outside parliament—The Opportunity Party (TOP), Conservative Party, and NZ Outdoors & Freedom Party.
In 2026, the research team used a hybrid approach to analyse Facebooks posts. The approach combined artificial intelligence (AI) for most coding tasks with independent validation by human coders. The main AI model used was Mistral Large 4, supported by Xiaomi MiMo-V2.6-Flash and Qwen3-Max.
The research team provided the AI model with detailed instructions based on a coding framework. This framework, used in the 2020 and 2023 studies for manual content analysis of Facebook posts, defines what to look for and how to classify it. Analysis considers written text as well as images and videos.
All AI classifications were automatically checked for consistency and compliance with the coding rules. Where claims needed to be fact-checked, additional external sources were consulted and evaluated, and human data annotators further validated the LLM classifications.
The research team included three postdoctoral scholars and three postgraduate students. Dr Krewel thanks the team for its hard work and invaluable contributions to the study throughout the 2026 election campaign.