There is one thing almost every study on polarization says and almost nobody quotes.
The divide between left and right is not the majority. It is the loudest minority.
A study published in September 2026, based on two decades of Brazilian data, shows that 69% of voters are not polarized in the strict sense, meaning they do not combine affection for their own camp with active rejection of the opposite camp. Strict affective polarization grew from 19% to 31% between 2006 and 2026, a 12 percentage point increase. It is real growth, but still less than a third of the electorate.
What grew most was not militancy. It was political negativity: indifference, fatigue and rejection without allegiance to either side.
This distinction is the starting point for everything that follows. Because most companies are making strategic decisions based on the minority that shouts, not the majority that buys.
What the numbers say about the real divide
The map is worth reading, because intensity varies far more than the discourse suggests.
According to data compiled by Statista, between July 2025 and June 2026, 15% of Americans placed themselves on the far left and 20% on the far right. In Germany, only 8% and 7% respectively, with 24% identifying as firmly centrist. In France, centrism is the lowest of the compared countries, at just 11%, while 13% sit on the far left and a notable 21% on the far right.
In other words: the United States and France are the cases of greatest fracture. Germany is the case of greatest centrist stability.
And there is a figure that contradicts the dominant narrative. Research by More in Common, cited by Valor in November 2026, indicates that around two thirds of people do not define themselves by political orientation and that the majority is more disengaged and willing to change their vote. The organisation’s co-founder, Tim Dixon, describes the dominant feeling as exhaustion.
The average voter is not at war. They are tired of the war.
Why perception differs so much from reality
If the majority is not polarized, why does it look like it is?
The answer has three layers, and none of them is political.
First, the architecture of platforms. In October 2026, Pew Research Center published a survey in which the share of people saying the internet and social media have made people more divided politically rose in almost every country compared. In the United States, it went from 79% in 2022 to 81% in 2026. In several European countries, it approaches 80%.
Second, individual choice. An important and often ignored point: quantitative studies show that even when algorithms offer unbiased information, users choose information similar to their own. Individual choices weigh more in limiting exposure to diverse content than algorithmic decisions. The algorithm does not create the bubble alone. It amplifies a preference that already exists.
Third, the attention economy. More interactions and more time on social media lead to higher degrees of polarization. Content that irritates generates more clicks than content that informs. The system was not designed to divide, but it was designed to retain, and retaining is easier with conflict.
The result is a dangerous asymmetry: the polarized minority has far more visibility than the moderate majority. And anyone who communicates professionally reacts to visibility, not to demography.
The cost nobody measures: the company in the middle
Here is the part that matters to anyone running a business.
A study published in the Journal of Global Information Management in January 2026 investigated how perceived political polarization on social media affects client retention in B2B firms. The conclusion is direct: perceived polarization negatively affects client retention, primarily through trust erosion.
Note the detail. The effect does not depend on the client agreeing with the company’s position. It depends on them perceiving that polarization exists. Merely associating a brand with a polarized environment already costs trust.
There is a second data point, from IBPAD, in a 2026 study on Brazilian consumption. Average purchase intent among those who approve of the government was 7.3 out of 10, against 6.8 among those who disapprove. The difference exists, but it is small: 0.5 points. In other words, a consumer’s political orientation explains far less of the purchase decision than common sense suggests.
And there is a third figure, from Ipsos, showing the shift in expectations. In February 2026, 56% of Americans said brands should remain neutral on political issues, down from 63% the previous year. The demand for neutrality is falling. But note the other side: 62% of Democrats and 53% of Republicans say they would punish a brand that took a position they disagree with.
The result is a double trap. Taking a stand alienates one part. Staying silent is read as omission. And punishment is real on both sides.
What corporate communication has already understood
The 11th annual USC Center for Public Relations survey, published in April 2026, measures what professionals are doing about it.
81% of public relations professionals rate the level of polarization in the United States as high or extremely high, against 69% of the general public. Those inside the sector feel more pressure than the population.
And support for companies taking positions on social issues unrelated to their business has dropped 38% since the killing of George Floyd. Only 55% of respondents agree that companies have a responsibility to advocate for social issues.
The practical translation: brand activism has stopped being a sign of modernity and become a risk calculation.
An Aberje study with 145 organisations shows the structural response: 96% of communication departments already participate in institutional positioning decisions and 91% participate in crisis management. Communication is no longer the department that writes the statement after the decision. It has become the department that assesses the decision before it exists.
The most expensive mistake: confusing noise with market
Here is the conclusion worth retaining, and it is counterintuitive.
If 69% of the electorate is not polarized, and if the difference in purchase intent between opposing camps is half a point on a ten-point scale, then the biggest market opportunity is not in choosing a side. It is in serving the majority that does not want to choose.
But that majority is silent, disengaged and hard to measure. The polarized minority is loud, highly engaged and easy to measure. Anyone looking at vanity metrics sees the minority. Anyone looking at revenue sees the majority.
It is exactly the same logic behind a well-built digital structure: the decision is not made by what makes the most noise, it is made by what the data shows. And the data, in this case, says the centre is larger than the sum of the edges.
What to do, in practice
Five concrete actions, in order of usefulness.
1. Separate values positioning from political positioning. A company can have clear values about how it treats clients, suppliers and teams without pronouncing on partisan disputes. Confusing the two is the origin of most boycotts.
2. Measure your base before deciding. If 80% of your clients are in one camp and 20% in the other, the maths is different from 50/50. The decision to take a stand should start with a map of your own customer base, not a personal conviction.
3. Treat polarization as operational risk, not a marketing topic. The Journal of Global Information Management study shows that trust is the channel of damage. Trust is lost quickly and recovered slowly. That is risk management, not communication.
4. Automate the monitoring. Brand mentions, sentiment by segment, spikes in negative reaction. No team can read this manually every day, and a reputational crisis is resolved in hours. It is the kind of routine intelligent automation sustains without depending on someone being on duty.
5. Use data to leave the bubble, not to confirm it. The most hopeful data point in this analysis comes from a study published in Nature Communications: when people were shown accurate figures about what the other side actually believes on controversial issues, their dislike of the other side softened. Accurate information reduces affective polarization. That holds for societies and for companies: those who know the real customer decide better than those who imagine the ideal one.
This is the kind of reading that data science applied to business makes possible: replacing impressions about the market with measurement of the market.
The limits of this reading
Four caveats, so that analysis does not become prophecy.
First, the Brazilian data is not universal. The 69% study is about the Brazilian electorate and the Lula-Bolsonaro polarization. The structure may repeat elsewhere, but the numbers do not transfer automatically.
Second, affective polarization did grow. It went from 19% to 31% in twenty years. It is a minority, but it has almost doubled, and it is the one that votes in primaries, comments on social media and organises boycotts.
Third, neutrality is losing support. The drop from 63% to 56% in one year, in the United States, is a sign that the demand for silence may be reversing. What works today may not work in 2027.
Fourth, perception is what counts. The B2B retention study is clear: the damage comes from perceived polarization, not the actual position. A company can be neutral and still be read as militant, if the environment around it is toxic.
Conclusion: the majority that does not shout
The divide between left and right is real, but smaller than perception. In the United States, 15% place themselves on the far left and 20% on the far right. In Brazil, 31% live strict affective polarization, against 69% who do not. Two thirds of people do not define themselves by politics.
At the same time, 81% of communication professionals consider polarization high, client trust erodes when a brand is associated with polarized environments, and the demand for neutrality fell seven points in a year.
What remains is a simple and dangerous asymmetry: the polarized minority sets the tone, the silent majority sets the revenue.
The question that remains for anyone running a business:
If your next positioning decision drove away 20% of clients to win 5% more engaged ones, would the numbers work?
It is worth doing the maths before the market does it for you.
