Contrarian Data Point
State a figure that collides with something the reader already believes, and leave the collision unexplained.
Definition
What it does
The copy states a number, a study result, or a measured outcome that collides head on with a belief the reader holds about their own field. The figure stands on its own, sometimes right after a quick naming of the assumption it is about to break, but it is not attached to a product pitch or a promise. The explanation for the figure does not arrive yet. The reader is left holding a contradiction between the number in front of them and the assumption behind them, and the only way to settle it is to keep reading.
Why it works
A belief the reader has never had to defend feels like knowledge rather than opinion. A hard figure that cuts against it cannot be waved off the way a bare opinion can, so the reader has to reconcile the two, and reconciling takes attention. That attention is the sale. The move also converts the writer from a seller into a source, because someone handing over an uncomfortable number reads as someone reporting rather than pitching. The withheld explanation keeps the tension open long enough for the argument that follows to land inside it.
Where it appears
* Most frequent first, based on materials selected for the Persuasion Taxonomy corpus.
Examples of Contrarian Data Point
“We find that two-thirds of the wayward boys—detained in prison schools—never ate oatmeal at home.”
An early oatmeal advertisement opens by reporting a survey finding about the home diets of boys held in reform institutions.
Why it’s this techniqueThe line opens on a count, not a claim: 'two-thirds of the wayward boys' who 'never ate oatmeal at home.' Attaching a hard number to prison schools forces the reader to connect delinquency with an absent breakfast, a link nobody arrives holding, and the surprise does the persuading before any argument about the food appears. The structural tell is that the product enters only as the thing missing from the finding, stated in the researcher's register of 'We find that,' with no benefit named. Proof of a virtue would count users; this counts non-users among a startling population.
Classification
- Primary technique
- PT-DSR-9279
- Classification confidence
- 0.72
- Source
- Quaker Oats, 1920s print ad (attributed)
“FACT: In Japan, a study found no difference in death rates between people who were vaccinated and those who weren’t.”
A subscription health newsletter mailing opens by citing a national study and the policy change that followed it.
Why it’s this techniqueThe lead stamps a single measured result as settled, 'FACT:', then reports a null finding, 'no difference in death rates between people who were vaccinated and those who weren’t', so the reader's existing belief becomes the thing under refutation before any argument of the writer's own arrives. The tell is that the reversal rides entirely on the shape of the evidence: a study, a named population, an outcome that reads as zero, with no adjective or accusation carrying the weight. An institutional reversal follows, but the government ending its program corroborates the finding, and the finding is what the copy is built on.
Classification
- Primary technique
- PT-DSR-9279
- Classification confidence
- 0.86
- Source
- Second Opinion, 1990s direct mail (attributed)
“A study by professors at various California universities found that the top-rated picks of Wall Street analysts in 2000 resulted in returns 31% below the market! And the stocks Wall Street liked least? They posted returns 49% points above the market!”
An investment advisory mailing opens by reporting how two groups of analyst-rated stocks performed against the broader market in a single year.
Why it’s this techniqueThe opening move is a sourced number that reverses what the reader assumes: 'A study by professors at various California universities found' that the 'top-rated picks of Wall Street analysts' delivered 'returns 31% below the market'. Authority gets cited so it can be turned against itself. The structural tell is the paired inversion, one figure for the favored stocks and a mirrored figure for the ones 'Wall Street liked least' at '49% points above the market', so the reversal lives in the arithmetic instead of in an assertion. The study citation supplies the credential; the flipped ranking supplies the argument.
Classification
- Primary technique
- PT-DSR-9279
- Classification confidence
- 0.93
- Source
- Renegade Investor, 2000s direct mail
“Most people are targeting the men in this market but research shows that WOMEN are flocking to these sites in droves - two or three times as many women as men on every dating site.”
A marketing information product opens by describing who its buyers usually aim at and then reporting who research says is actually there.
Why it’s this techniqueThe line opens by naming the crowd's position, 'Most people are targeting the men in this market', then breaks it with a counted claim, 'research shows that WOMEN are flocking to these sites in droves', and the figure 'two or three times as many women as men on every dating site'. The tell is the order. The received belief comes first and the number arrives as the instrument that overturns it, with the reversal carried by 'but' rather than by intensity. A market-gap appeal sits nearby, yet the copy earns that gap only through the counted contradiction, which is what the sentence is built to deliver.
Classification
- Primary technique
- PT-DSR-9279
- Classification confidence
- 0.88
- Source
- Tina Golden, 2010s web page
“If we include everyone who works on Gumroad, it's 25. If we include full-time employees, it's none. Not even me.”
A software founder's public essay opens by giving the headcount he usually quotes and then the two real numbers behind it.
Why it’s this techniqueThe copy opens with the figure a reader expects, 'ten or so', flags it as accommodation, 'what others expect', then replaces it with two counts that fight each other: everyone who works there is '25', and full-time employees are 'none. Not even me.' The tell is that the number itself carries the argument. Each sentence is an arithmetic answer to one question, and the surprise sits in the count rather than in a story, a claim, or a promise. The confession that the usual answer is a convenience performs candor, yet the candor exists to deliver the figure.
Classification
- Primary technique
- PT-DSR-9279
- Classification confidence
- 0.71
- Source
- Sahil Lavingia Gumroad, 2020s web page
See whether your own copy uses Contrarian Data Point, and what else it is doing: analyze your copy.
Boundary Conditions
When it lands
- The belief being contradicted is one the reader genuinely holds and has never questioned
- The figure is specific and checkable, so it reads as a finding rather than a boast
- The explanation is withheld for a beat, leaving the contradiction standing on its own
- What follows actually resolves the number, so the opening is a door and not a stunt
When it dilutes
- The statistic is merely surprising and does not cut against anything the reader believes
- The number is round, hedged, or unsourced, so it reads as a made up figure
- The copy explains the figure in the same breath, closing the tension before it can pull
- The claim is so extreme the reader rejects the source instead of the belief
Taxonomic Relationships
- PT-DSR-9505Acknowledge-the-Trope Then Reassert
- PT-DSR-0010Belief Violation Open
- PT-DSR-9187Category Dismissal Then Redirect
- PT-DSR-9737Category Reversal Premise
- PT-DSR-9801Era-Shift Obsolescence Frame
- PT-DSR-9824Prior Approach Invalidation
- PT-DSR-9950Reply or Forward Frame
- PT-DSR-9901Setup-Then-Overlooked Pivot
- PT-DSR-9987The Mirror Disrupt
Provenance
- Claude C. Hopkins, Scientific Advertising (1923), on specific figures doing the persuading where general claims are discounted
- Chip Heath and Dan Heath, Made to Stick (2007), on breaking a reader's guessing machine and then filling the gap
- Eugene M. Schwartz, Breakthrough Advertising (1966), on redirecting beliefs the market already holds rather than arguing against them head on