The most quoted brand trust statistic in circulation is six years old and almost nobody citing it says so.
That figure, the one about only a third of people trusting the brands they buy, comes from Edelman’s 2019 brand trust study. It appears in decks and articles dated this year with no date attached, which is fine until someone in the room checks. The current numbers are different, in some cases substantially, and the 2026 picture is more interesting than the stale version anyway.
What follows is the set of brand trust statistics actually worth citing, each with its source and year, plus a note on which three should change what you do rather than just what you say.
Trust Now Sits Level With Quality
The headline finding from Edelman’s 2026 Trust and Brands report is a ranking rather than a single number. Asked which factors are important or critical purchase criteria, respondents put quality at 89%, trusting the brand at 88%, and value at 88%.
That is the number to use in a budget conversation, because it puts trust on the same tier as the two things every business already measures and already funds. A standalone trust statistic invites the response that trust is soft. A statistic showing trust matching value, measured in the same question, does not.
The methodology matters if you intend to cite it. That report surveyed 17,688 people across 15 countries, around 1,200 per country, through 25-minute online interviews conducted between 23 April and 11 May 2026. It is the eighth annual edition of Edelman’s brand-specific study, which is what makes the year-over-year movement in it usable rather than anecdotal.
One caution about how these are framed. Every figure here is a stated attitude collected in a survey, not an observed purchase. People overstate how much principle drives their buying and understate how much price does, consistently, in every study of this kind. That does not make the numbers useless, because the comparisons within a single question are still informative, but it does mean the right claim is “people say trust is as important as value” rather than “trust is as important as value.” The first survives scrutiny. The second invites a fight you will lose.
The Headline Numbers for 2026
Four figures from the 2026 report carry most of the weight.
Trust as a purchase criterion sits at 88%, level with value and a point behind quality. This is the anchor statistic.
Sixty-six percent of respondents said they are hesitant or unwilling to trust someone different from them, where different covers core values, the facts and sources a person accepts, their approach to societal problems, or their background. That is two thirds of the sample reporting a default position of distrust toward people unlike themselves.
Within that group, 30% said they are unwilling to use brands that are used by people who differ from them. Read that slowly, because it is the most commercially consequential number in the report. Roughly a third of the two thirds are treating brand choice as a boundary marker rather than a purchase.
And where a brand has earned both trust and relevance, respondents reported almost twice the willingness to support that brand expanding into new and different audiences. Trust alone does not produce that permission. Relevance alone does not either.
Trust Is Fragmenting by Identity, Not by Category
The 66% and 30% figures together describe something different from a general decline in trust, and the distinction changes what you do about it.
A general decline would mean everyone trusts everything less, which is a problem solved by proving yourself. Fragmentation by identity means trust is being allocated along group lines, so the same brand is trusted and distrusted simultaneously by people who have identical information about it. Proving yourself does not help, because the question being answered is not whether you are good.
There is a second-order effect here that the figures hint at without measuring. If a third of a distrustful two thirds treats brand choice as a group signal, then a brand’s existing customer base becomes part of its product description to everyone considering it. That is not a new idea in luxury or in politics, but it is new as a mass-market condition, and it means customer acquisition and brand positioning stop being separable activities. Who already buys you is now a reason to buy you or not.
The uncomfortable implication is that neutrality is not available. A brand that declines to signal anything still signals something to a reader looking for a signal, and what it signals is usually read from its most visible customers rather than from its own communication.
Edelman frames the response in terms of relevance, which it breaks into utility, identity, community, emotional connection and pop culture, and reports that utility is on average the most effective route globally. The report also says brands are falling short on all of these dimensions.
The reading worth taking from that is practical. Utility is the one dimension that works across groups, because usefulness does not require shared values. A brand reaching for identity or community as its relevance strategy in a fragmented market is choosing to be adopted by one group and rejected by another, which is sometimes correct and should at least be deliberate.
Where a Brand Is Based Now Counts
More than two thirds of 2026 respondents said a brand being headquartered in their own country is important or a critical purchase deal breaker. That is up five points since 2023.
A five-point move across a 15-country sample over three years is a trend rather than noise, and it is the statistic most likely to be dismissed internally because it sounds like something that applies to other markets. It applies to yours. For multinational brands it raises a question about which entity customers believe they are buying from, and for domestic brands it is an advantage most are not claiming out loud.
The practical version is simple enough to test. Can a customer tell where you are based from your site in under ten seconds? For a surprising number of companies the answer is no, because the information lives in a footer or a legal page.
What Trust Is Worth, in Behaviour
The behavioural numbers are older, from Edelman’s 2019 brand trust report, which surveyed 16,000 people across eight countries. They are worth citing with that date attached because nothing more recent measures the same gaps as cleanly.
People who trusted a brand were far more likely to act. On buying new products first, 53% against 25% for brands they did not trust. On staying loyal when a trendy or innovative competitor appeared, 62% against 29%. On advocating for the brand, 51% against 24%. On defending it when things went wrong, 43% against 22%.
Each of those is roughly a doubling, and the last one is the one to put in front of a board. Defence during a problem is the capability that reputation work buys, and it is invisible until the day it is needed.
From the same study, 81% said brand trust was a deal breaker or deciding factor in a purchase. That figure pairs usefully with the 2026 one at 88%, as long as you present them as two studies rather than as a trend line, because the question wording differs.
The Advertising Numbers Nobody Quotes
Two more from the 2019 report, both of which undercut the usual conclusion drawn from trust research.
Seventy-four percent said they find ways to avoid advertising. That is the context for every trust campaign ever proposed, and it is rarely in the deck proposing one.
And 87% reported strong trust in a brand message after encountering it across six channels, against 13% after a single viewing. The gap is the finding. Trust is not produced by a message, it is produced by repetition across independent surfaces, which is why a single excellent campaign moves it so little.
It also explains a pattern most communications teams notice and cannot account for. A campaign runs, awareness moves, and trust does not. The six-channel figure says why: a campaign is one voice repeated, and repetition of one voice produces recognition rather than belief. Six independent sources saying a compatible thing produces belief. Those are different purchases and they come out of different budgets, which is why the second one rarely gets made.
Two further figures round out the picture. Fifty-six percent said too many brands use societal issues as a marketing ploy, and only 21% said they know from personal experience that the brands they use keep society’s interests in mind. The distance between those two is where most purpose marketing fails.
The Earned-Voice Ratio
Here is the framework we use to turn these brand trust statistics into a plan, and it comes directly out of the six-channel finding. Call it the Earned-Voice Ratio.
Count every surface where a prospective customer could encounter a claim about you. Then split those surfaces into two groups: ones you control, meaning your site, your social accounts, your advertising and your own newsroom, and ones you do not, meaning press coverage, review platforms, analyst notes, customer posts and the answers AI assistants give about you.
The ratio of the second group to the first is the number that predicts whether the six-channel effect can happen for you. A brand with twelve owned surfaces and one earned surface cannot produce repetition across independent sources, no matter how much it spends on the twelve, because a reader who sees the same claim in twelve places you own has seen it once.
Edelman’s 2026 report puts this plainly in its own language: earned voices matter more than brand messaging, with unpaid advocates, customers and peers carrying the greatest weight. The ratio just makes that measurable. Count both groups this week and the number will probably embarrass you, which is the point.
The target is not a particular figure. It is movement, and the only way to move it is to add surfaces you do not control, which means media work, review generation and getting into the reference sources that AI systems read.
How to Cite These Without Getting Caught Out
Three rules, learned the hard way.
Attach the year and the study name to every figure, in the sentence rather than in a footnote. “Edelman’s 2026 Trust and Brands report” takes six words and removes the entire category of objection.
Never present figures from different studies as a trend. The 34% from 2019 and the 88% from 2026 measure different questions, and lining them up implies a change that the data does not support. Someone in your audience will know.
And check whether a statistic you have inherited is primary. A large share of trust figures circulating in marketing content trace back to a roundup citing a roundup, and the original either says something narrower or does not exist. If you cannot reach the primary source in two clicks, do not use the number.
What to Do With This in the Next Month
Three concrete moves, in order of how quickly they pay.
Run the Earned-Voice Ratio count. One spreadsheet, two columns, an hour. If earned surfaces are fewer than three, that is your diagnosis and it outranks everything else on your marketing plan.
Make your location legible. Put the country you are headquartered in somewhere a customer sees it without looking, given that more than two thirds now treat it as important or a deal breaker. This is a copy change, not a strategy.
Then pick utility as your relevance claim and write it down in one sentence: what you are useful for, to whom, better than the alternative. Edelman’s data says utility travels across groups while identity and community do not, so in a market fragmenting along identity lines, usefulness is the claim with the widest reach. Test the sentence on someone outside your category and see whether they can repeat it back.