Don’t take our word for it. Here’s every fact on this site, with its evidence grade, its caveats, and a link to the source, so you can check us.
We start you on the strong ones: findings that have been replicated or come from one large, well-controlled study. Moderate and weak facts are here too, labeled for what they are. Use “Link to this fact” to share a single card.
Indoor air / Other evidence
At 2,500 ppm CO2 one scale of the decision test went up, not down: focused activity rose about 20%, which the authors read as possible over-concentration.
Weak evidence: Scale-specific mixed pattern in 22 people; the 'overconcentration' reading is the authors' interpretation.
Source: Satish 2012, Environmental Health Perspectives Link to this fact
Indoor air / Other evidence
At constant ventilation and low VOCs, injecting pure CO2 to about 945 ppm lowered decision-making scores by about 15%, and to about 1,400 ppm by about 50%, compared with the low-CO2 Green+ days in the same 24 workers.
Weak evidence: Small within-subject study, fixed day order (CO2 days were days 2 and 3), one test per day, industry-gift funded, later large randomized trial found no systematic decrement.
Source: Allen 2016, Environmental Health Perspectives Industry-funded Link to this fact
Indoor air / Other evidence
Across the six days, each 20 cfm/person more outdoor air was associated with about 18% higher decision-making scores, each 400 ppm more CO2 with about 21% lower, and each 500 µg/m³ more TVOC with about 13% lower.
Weak evidence: Confidence intervals 'data not shown'; exposures are partly collinear with study day; these per-unit estimates are far smaller than the day-level contrasts.
Source: Allen 2016, Environmental Health Perspectives Industry-funded Link to this fact
Indoor air / Other evidence
In a lab test on human airway cells (not people), one hour of breathing-zone-level exposure to citrus-scent limonene reacting with ozone, at the upper end of office air levels and with or without printer exhaust, did not kill the cells or cause measurable oxidative stress or inflammation (no change in IL-8 or other markers versus clean air).
Weak evidence.
Caveats
Evidence/in-vitro: single lab, one cell line (less sensitive than primary cells by the authors' own comparison), 1 h only, 3 runs, no barrier or sensitive-person endpoint; a null is not proof of safety.
Source: Verstraelen 2024 (J Environ Sci Health A), Journal of Environmental Science and Health, Part A Link to this fact
Indoor air / Other evidence
Running an inkjet printer in the same chamber added micron-sized particles (PM10 average about 10 micrograms per cubic meter, maximum about 30) and raised ultrafine particles to about 6,600 per cm3 on average (maximum about 11,900) once limonene and ozone were also present.
Weak evidence: One inkjet model and one run; particles began about 20 min after printing started; the authors infer, without testing, that printer VOCs reacted with ozone to form aerosol.
Source: Verstraelen 2024 (J Environ Sci Health A), Journal of Environmental Science and Health, Part A Link to this fact
Indoor air / Other evidence
A 2025 narrative review of school air quality (215 records screened down to 17 included studies, no registered protocol and no bias assessment) concludes that better ventilation, HEPA filtration and CO2/PM2.5 monitoring in classrooms are widely recommended, but its authors admit the evidence is correlational.
Weak evidence: Low-barrier journal; the 17 studies are not identified; numbers often attach to references that do not contain them; not fragrance-related.
Indoor air / Other evidence
The review reports that each 2 cfm/person rise in classroom ventilation (from 2 to 15 cfm/person) went with about 3% more fifth graders passing state math and reading tests, and doubling from 7.5 to 15 cfm with about 8% better academic performance.
Weak evidence.
Caveats
Secondary restatement that omits the null full-sample result and bell-shaped fit in the primary study; the percentages are really percentage points of pass rate; verify against Haverinen-Shaughnessy 2011.
Indoor air / Other evidence
Low-cost real-time sensors (PM2.5, CO2, temperature, humidity), IoT networks (329 German classrooms for six months) and community monitors such as PurpleAir let schools track air quality continuously, but sensors need calibration against reference equipment.
Weak evidence: Descriptive; no accuracy data given.
Indoor air / Other evidence
The classroom CO2 and cognitive-score association was weaker for average CO2 than for peak CO2: only 1 of 6 mean-CO2 estimates was statistically significant versus 5 of 6 for peak CO2.
Weak evidence: Small n; confidence intervals mostly span zero for mean CO2; Stroop throughput at peak CO2 was null (-1.13, CI -3.4 to 1.13).
Source: Dedesko 2025, Journal of Exposure Science & Environmental Epidemiology Link to this fact
Indoor air / Other evidence
Raising a solvent-paint VOC level from about 100 to about 1,000 micrograms per cubic meter (TVOC) did not consistently change cognitive test performance over about an hour in a well-ventilated chamber; only 3 of 20 outcomes differed and they pointed toward better, not worse, scores.
Weak evidence.
Caveats
The VOCs were petroleum-solvent alkanes and xylene, not fragrance compounds; one hour; different people in each VOC group; the test battery may be insensitive; 20 tests uncorrected. A null here does not clear fragranced products.
Source: Zhao 2025, Science of The Total Environment Link to this fact
Indoor air / Other evidence
The study found no interaction between temperature and VOC level: the heat effect on accuracy was not significantly larger at about 1,000 micrograms per cubic meter of paint-solvent TVOC than at about 100, though within-group slopes (5.2% vs 2.4% more errors) looked different.
Weak evidence: The slope comparison is a difference in significance, not a tested difference; the study was not powered for interactions; one paint, one hour.
Source: Zhao 2025, Science of The Total Environment Link to this fact
Indoor air / Other evidence
Dosing a 12-compound VOC mixture (mostly ethanol and propanol, plus limonene, toluene, aldehydes), with total VOCs measured up to about 2,100 µg/m3, did not impair cognitive test performance in healthy adults over about 2.75 hours.
Weak evidence.
Caveats
Only one mixture (not scented products, not building materials, no ozone products); TVOC levels missed targets (the no-dose group had the highest); aerospace-industry funded; short exposure.
Source: Herbig 2026, International Journal of Hygiene and Environmental Health Industry-funded Link to this fact
Indoor air / Other evidence
The air in this cabin experiment was rated good or very good by most people in every group, and odor was barely noticed, even at the highest measured total VOC level.
Weak evidence: Low humidity (10-15%) and olfactory adaptation may have blunted perception; one mixture; short exposure.
Source: Herbig 2026, International Journal of Hygiene and Environmental Health Industry-funded Link to this fact
Indoor air / Other evidence
Earlier reports that raised indoor CO2 or VOCs impair thinking come mostly from small crossover studies; a large randomized test found no consistent effect, so those findings should be treated as unsettled.
Weak evidence: The new study did not use the strategic management simulation test and did not test material-emitted VOCs, so it does not refute the earlier work; it is a scope conflict.
Source: Herbig 2026, International Journal of Hygiene and Environmental Health Industry-funded Link to this fact
Indoor air / Other evidence
The estimated mass of new particles in the 0.1-0.3 um range was about 27-40 ug/m3 with injected limonene and about 3-24 ug/m3 with a lemon-scented solid air freshener.
Weak evidence: Estimates from counts at 0.1 um and above by assumption, not weighed; chamber at 60-100 ppb ozone, well above typical indoor levels outside summer peaks.
Source: Wainman 2000, Environmental Health Perspectives Link to this fact
Indoor air / Test-chamber measurement
A 15-second spray of lemon-scented furniture polish on one coffee table in a 25 m3 chamber raised limonene to 175 ppb (about 975 ug/m3), and it took about 3 hours to clear.
Weak evidence: One run, one product; the run had no added ozone and no particle measurement, so the particle risk from polish is the authors' inference.
Source: Wainman 2000, Environmental Health Perspectives Link to this fact
Indoor air / Other evidence
Indoor ozone in New Jersey offices and homes tracked outdoor levels in a 150-day summer record and exceeded 120 ppb on 4 days (80 ppb on 17), so the 60-100 ppb ozone pulses in the chamber runs fall within levels seen indoors on some summer days.
Weak evidence: Cited from earlier studies of the same group, 1989 to 1999 era, New Jersey only.
Source: Wainman 2000, Environmental Health Perspectives Link to this fact
Indoor air / Other evidence
A rough estimate says the airborne products of indoor ozone chemistry are present at one-third to twice the molar level of ozone itself, so people breathe in roughly 0.33-2 times as many oxidation-product molecules as ozone molecules indoors.
Weak evidence.
Caveats
A back-of-envelope model built on three small lab and chamber surface studies (F 0.1-0.7); it counts molecules, not toxicity, covers surface chemistry only, and was not tested in homes.
Source: Weschler 2006, Environmental Health Perspectives Link to this fact
Indoor air / Other evidence
Someone doing their own house cleaning in a region with moderate outdoor ozone was estimated to inhale about 20 micrograms a day of formaldehyde and 35 micrograms a day of new particles from ozone reacting with cleaning-product and air-freshener ingredients.
Weak evidence.
Caveats
A modeled estimate borrowed from another review, not a measurement, for cleaning products and air fresheners together, not for perfume. It adds to formaldehyde and particles from other sources.
Source: Weschler 2006, Environmental Health Perspectives Link to this fact
Indoor air / Other evidence
Personal-use scented products (perfumes, colognes, hair treatments) release their ozone-reactive chemicals close to the breathing zone, so inhaled doses of the reaction products can be larger than room-average models predict.
Weak evidence: A single supporting chamber citation, stated as a qualitative point; no dose given here.
Source: Weschler 2006, Environmental Health Perspectives Link to this fact
Indoor air / Other evidence
In the same model, OH radicals convert about as much total VOC as ozone does, and more when a gas stove is running, because stove NO scavenges ozone; OH attacks alcohols, aldehydes and aromatics as well as terpenes.
Weak evidence: Single simplified model, no field validation; OH inputs from literature; only about a quarter of OH conversion is limonene.
Source: Waring & Wells 2015, Atmospheric Environment Link to this fact
Indoor air / Other evidence
In the model, limonene is the largest single VOC loss for each of the three oxidants (ozone 68%, NO3 59%, OH 24% of each oxidant's VOC loss), but OH also attacks ethanol, formaldehyde and other VOCs, so limonene is only about a quarter of OH conversion; this is a model result, not a measurement.
Weak evidence: Model with median VOC concentrations; limonene is not specific to fragrance (also wood and cleaning products); products formed were not computed.
Source: Waring & Wells 2015, Atmospheric Environment Link to this fact
Indoor air / Other evidence
Modeled NO3 radicals convert about ten times less VOC than ozone or OH in most homes, except with gas-stove NOx and high outdoor ozone.
Weak evidence: NO3 is lower than other published estimates (Weschler 2006, Nojgaard 2010), which the authors note; model only.
Source: Waring & Wells 2015, Atmospheric Environment Link to this fact
Indoor air / Other evidence
In the model, a gas stove can make HONO and NO2 that light turns into OH radicals, making this the main OH source in stove homes; the authors call for field tests.
Weak evidence: Rests on a single classroom photolysis data set and scenarios that include assumed high light; not tested in homes; gas stoves are not a fragrance source.
Source: Waring & Wells 2015, Atmospheric Environment Link to this fact
Indoor air / Other evidence
Plants kept in water gave off far more limonene in a lab chamber (about 810 vs 74 micrograms an hour) than plants in soil, but in real rooms the extra limonene was only about 1,000 vs 530 micrograms over 3 hours, about 14% of what the lab rate predicted.
Weak evidence.
Caveats
Lab rate assumes all monoterpene signal is limonene; number of plants in the chamber not stated; reasons for the gap (ozone reactions, air exchange, sorption) were named but not measured.
Indoor air / Other evidence
Over 3 hours the fragrant plant removed essentially no benzene, toluene, ethylbenzene or xylenes (concentrations changed by less than 1.2 times).
Weak evidence: Single species, 3-h window; consistent with, not independent of, wider reviews showing plants do little in real rooms.
Indoor air / Other evidence
Two products in the same category can behave very differently: with seed particles present, one deodorant turned about 17% of its evaporated mass into aerosol and the other about 6% (about, read from Fig. 2), and the two cleaning liquids ran about 5% and 12%.
Weak evidence: Approximate values read from Fig. 2; two products per category; the authors say the products are not representative of the market.
Indoor air / Other evidence
Limonene alone made only a small share of the aerosol from these products: the authors estimate about 0.4-2.8% of the seeded and 1.9-26.5% of the unseeded aerosol, so other product ingredients mattered as much or more.
Weak evidence: Calculated from limonene content and pure-limonene yield; the other precursors are not identified; chamber at 6.5 ppm ozone.
Indoor air / Other evidence
Adding seed particles raised aerosol yields 1.5 to 14 times, but pure limonene made about 52-53% aerosol with or without seed, so how much aerosol a scented product makes depends on the particles already in the air.
Weak evidence: Single lab, chamber at 6.5 ppm ozone; ammonium sulfate seed only; mechanisms not tested (no product speciation).
The industry / Journalism
The first press coverage of a 2011 study finding 133 undisclosed chemicals in 25 best-selling scented products, in the US government's environmental-health journal, quoted no manufacturer or industry group; the underlying numbers should be cited from the study itself (Steinemann 2011).
Weak evidence: Media-coverage note only; all product numbers belong to the primary src-2011-fragranced-consumer-products-chemicals-emitted-ingredients-unlisted.
Source: Potera 2011 (EHP news), Environmental Health Perspectives Link to this fact
The industry / Review of other studies
The fragrance industry's answer to autoxidation is to add antioxidants to linalool and linalool-rich natural extracts, but the Gothenburg researchers say that protection is hard to control.
Weak evidence: Industry practice via personal communication from a RIFM scientist (A. M. Api) in a RIFM co-funded paper; no data on effectiveness.
Source: Hagvall 2008, Contact Dermatitis Industry-funded Link to this fact
The industry / Animal study
Two common fragrance ingredients, hexyl and benzyl salicylate, may test positive in the mouse skin-allergy assay not because they are true human sensitizers, but because of a mouse-specific irritant/pharmacological effect, similar to a known false-positive detergent (sodium lauryl sulfate).
Weak evidence.
Caveats
A hypothesis proposed by structural-alert absence and analogy only; the paper offers no direct evidence (e.g., a biomarker distinguishing true haptenation from irritant-driven lymph node activity) and the underlying data are RIFM's own unpublished proprietary dataset.
Source: Roberts & Api 2018, Regulatory Toxicology and Pharmacology Industry-funded Link to this fact
The industry / Review of other studies
The detergent industry's own risk assessment for esterquat fabric softeners sets a margin of exposure of about 8,100 (NOAEL 300 mg/kg bw/day vs. an aggregate consumer exposure estimate of 36.9 micrograms/kg bw/day) and concludes esterquats in fabric conditioner are safe for consumer use.
Weak evidence.
Caveats
Every number comes from the industry's own unpublished, unnamed studies with no independent citation trail; the exposure model covers only dermal and minor oral routes (no inhalation route at all, despite fabric conditioner being used in heated dryers); no carcinogenicity study and no multigeneration reproductive study exist for esterquats — the 'safe' conclusion infers these endpoints are unnecessary from adjacent negative results rather than testing them directly.
Source: HERA Project 2008 (A.I.S.E./CEFIC) Industry-funded Link to this fact
The industry / Review of other studies
The author of a 2023 clinical review arguing that chemical sensitivity is primarily psychogenic (not caused by product chemistry) previously received partial, later-withdrawn funding for an uncompleted IEI genetics study via an unrestricted grant from the Research Institute for Fragrance Materials, the fragrance industry's own research and trade body.
Weak evidence.
Caveats
The funding is disclosed by the author, was withdrawn before the study was completed (due to poor enrollment, not a documented data or conclusions issue), and does not by itself invalidate the paper's clinical arguments; it is a conflict-of-interest fact worth weighing, not proof of bias in this specific manuscript.
Source: Binkley 2023, Journal of Allergy and Clinical Immunology: In Practice Industry-funded Link to this fact
The industry / Review of other studies
This commentary on benzalkonium chloride's adverse effects on the nasal mucosa was funded by an educational grant from AstraZeneca USA, a pharmaceutical company with a commercial interest in intranasal corticosteroid products, which the same commentary recommends as the preferred alternative to decongestant sprays for allergic rhinitis.
Weak evidence.
Caveats
Funding disclosed in the acknowledgment; the underlying rhinostereometry trials (1995-1996) predate this 1999 funded commentary, so only the commentary's framing/synthesis, not the trial data, is implicated.
Source: Graf 1999, Clinical Therapeutics Industry-funded Link to this fact
The industry / Other evidence
This review of BAK's harms to the trabecular meshwork was partly funded by Merck, Inc., and its senior author, Paul Kaufman, discloses paid consultant, honoraria, and financial-support relationships with Alcon and Allergan, both manufacturers of BAK-free alternative eye-drop preservatives (SofZia and Purite respectively) that the review discusses favorably.
Weak evidence.
Caveats
This is a conflict-of-interest disclosure, not an evidentiary finding. The funding pattern runs opposite to a typical 'industry defends its own product' case: the paper argues against a preservative (BAK) while its senior author has financial ties to companies selling competing, BAK-free alternatives.
Source: Rasmussen et al. 2014, Journal of Ocular Pharmacology and Therapeutics Industry-funded Link to this fact
The industry / Review of other studies
This benzalkonium chloride pulmonary-toxicity review was written by a single author with no coauthors and no named funding source, who is the principal of a private toxicology consultancy (Pharma Toxicology Safety Solutions LLC) whose business is toxicology/regulatory consulting typically for pharmaceutical clients; the article declares no competing financial interests but gives no funding statement either way.
Weak evidence.
Caveats
This is a characterization of authorship and potential conflict of interest, not a toxicology finding; the review's consistent framing (reinterpreting adverse animal/in vitro findings as clinically inapplicable, and concluding regulatory tightening is unwarranted) matches the argumentative pattern common in industry-oriented safety literature, but no direct pharma-company funding or engagement was found or disclosed.
Source: Johnson 2018, Journal of Aerosol Medicine and Pulmonary Drug Delivery Link to this fact
The industry / Review of other studies
The study authors cite an earlier analysis reporting that some commercially available PTFE powders contained more than 400 ppb of PFOA — 16 times the EU's 25 ppb regulatory limit — indicating PFOA contamination has been found in the market, not only in controlled lab experiments.
Weak evidence.
Caveats
This is a citation within the primary paper (attributed to Larsen et al. 2005, Analyst) that this evidence base has not independently read or verified; no product names, brands, or cosmetic use are identified.
Source: Oshima et al. 2020, Scientific Reports Industry-funded Link to this fact
The industry / Review of other studies
According to an industry report cited in this review, about 75% of scented candles are fragranced with artificial scents because they cost less than natural fragrance materials.
Weak evidence: Cited without an independently checkable primary source in this review; industry-report figure (ref. 48).
Source: Adamowicz 2019, Cancer Prevention Research Link to this fact
The industry / Review of other studies
One of this 2024 axillary-dermatitis review's three authors discloses paid consulting, honoraria, or equity relationships with multiple cosmetic- and pharmaceutical-adjacent companies, including the industry-funded Cosmetic Ingredient Review panel.
Weak evidence.
Caveats
Disclosed conflict of interest is at the individual-author level, not a study funding source; the review's factual content was not evaluated differently on this basis, but the affiliation is named for transparency.
Source: Musicante 2024, Cutis Industry-funded Link to this fact
The industry / Other evidence
The US market for scented candles is estimated to be worth approximately $2 billion annually.
Weak evidence: Abstract-only/background figure; sourced from an industry-characterization paper (Ahn et al. 2015) not independently verified in this evidence base.
Source: Nazir et al. 2024, Annals of Medicine & Surgery Link to this fact
The industry / Measured in people or real products
A 2022 Korean study proposes prioritizing the specific hormone-disrupting chemical combinations that most often turn up together in household products, rather than testing chemicals one at a time, for future mixture-toxicity testing and regulatory management.
Weak evidence: This is a testing-prioritization proposal based on ingredient co-occurrence, not new toxicity data showing that any of the proposed mixtures cause harm.
Source: Lee & Ji 2022, Ecotoxicology and Environmental Safety Link to this fact
The industry / Other evidence
A study by Petry et al. (2014) concluded that, under normal conditions of use, scented candles do not pose known health risks to consumers.
Weak evidence.
Caveats
Petry et al. 2014 is now a evidence base primary (src-2014-health-risk-evaluation-selected-voc-svoc-particulate). Its Conflict of Interest statement reads: "Financial support for this work has been provided by a consortium of companies organized by the Research Institute for Fragrance Materials Inc. (RIFM)." All eight authors are affiliated with ToxMinds, RIFM, SC Johnson, Firmenich, Procter & Gamble or Reckitt Benckiser. It is one study among several in this same Viewpoint reaching the opposite conclusion (Adamowicz 2019, Derudi 2014, Chandrasekaran 2021), and its own 'no known health risk' conclusion did not test acrolein or NO2 and used a PAH detection limit too coarse to find the BaP levels a later, comparably funded sibling study (Salthammer 2021) measured exceeding WHO guidance.
Source: Singh 2023, Environmental Science & Technology Link to this fact
The industry / Other evidence
An earlier FDA-lab study found that musk xylol, a synthetic musk fragrance ingredient, did move out of a 24-hour skin reservoir and into the bloodstream over time, unlike the self-tanner and hair-dye chemicals in the 2004 follow-up study.
Weak evidence: Cited secondhand from Yourick 2004's discussion; the Hood 1996 primary source has not been independently read and ingested into this evidence base.
Source: Hood 1996, cited in Yourick 2004, Toxicology and Applied Pharmacology Link to this fact
The industry / Self-reported survey
Regular users of the tested brand reported more relief overall than users of other brands, and the gap was largest on the unbranded placebo (2.03 vs 1.53), which the authors read as brand confidence or more placebo responders among brand users.
Weak evidence.
Caveats
Brand-user vs non-user comparison was not randomized; the brand-specific t-test was at the 6% level (one-tail); alternative explanation of different placebo responsiveness is acknowledged by the authors.
The industry / Other evidence
In a sample of 49 Dutch sedative and antidepressant products, sedatives were more often green, blue, or purple (9 of 31) than antidepressants (1 of 18); Fisher exact P = 0.029.
Weak evidence: One sample per manufacturer, one country in the 1990s, borderline P; white was the most common color in both groups and colors were not consistent within either.
Source: de Craen 1996, BMJ Link to this fact
The industry / Other evidence
Retailers scent stores on purpose: an early marketing experiment found students rated a mock store, its environment, and its merchandise higher when an essential-oil scent was diffused, and the presence of the scent mattered more than which scent.
Weak evidence: One lab simulation with students and four essential oils, unscented cell size not stated, no field data or purchases; marketing outcomes only, no exposure or health data.
Source: Spangenberg 1996, Journal of Marketing Link to this fact
The industry / Other evidence
Scented mock-store shoppers did not stay longer (about 578 s vs 586 s) but felt they had stayed a shorter time: unscented shoppers overestimated their visit by about 75 s, scented ones by under 2 s.
Weak evidence: Single simulated-store study with students; the 'flow' explanation is the authors' untested interpretation.
Source: Spangenberg 1996, Journal of Marketing Link to this fact
The industry / Other evidence
In the same lab study, scent raised intention to visit the store (4.20 to 4.94 on a 7-point scale) and purchase intention for a backpack, but not for a calendar or a self-chosen product.
Weak evidence: Intentions, not purchases; one of three products; baseline-liking explanation untested.
Source: Spangenberg 1996, Journal of Marketing Link to this fact
The industry / Other evidence
The authors measured mood with a standard three-factor scale and report no effect of scent on mood, but they show no mood data (available only on request).
Weak evidence: The null is asserted in one sentence with data withheld, so it cannot be checked; ratings of the store as 'cheerful' and 'pleasant' did rise.
Source: Spangenberg 1996, Journal of Marketing Link to this fact
The industry / Other evidence
A 2001 review arguing nasal sprays should avoid benzalkonium chloride was written by the lead investigator of the key positive trials and supported by an AstraZeneca educational grant; AstraZeneca also sold a BKC-free steroid spray (budesonide) at the time (inference from another review's table).
Weak evidence: Funding does not by itself invalidate the findings; the commercial interest is inferred, not stated in the paper.
Source: Graf 2001, Respiratory Medicine Industry-funded Link to this fact
The industry / Other evidence
In a 2001 field experiment in one gift shop, shoppers who smelled lavender with slow music or grapefruit with fast music reported higher approach intentions and satisfaction than shoppers given mismatched scent-and-music pairings.
Weak evidence.
Caveats
One small store, about 27 respondents per cell, self-reported 7-point ratings, young mostly female sample; interaction explains 6-7% of variance; scent-arousal labels come from aromatherapy books, not measurement.
Source: Mattila & Wirtz 2001, Journal of Retailing, Journal of Retailing Link to this fact
The industry / Other evidence
The authors tell retailers that pairing a relaxing scent with slow music, or a stimulating scent with fast music, can encourage shoppers to engage in impulse buying.
Weak evidence: Impulse buying was one self-reported item, not sales data, and the paper's Table 5 impulse means disagree with its Fig. 1.
Source: Mattila & Wirtz 2001, Journal of Retailing, Journal of Retailing Link to this fact
The industry / Other evidence
The authors argue price cues may help explain why costly drugs are popular over cheap equivalents and why patients switching from brand to generic may report weaker effects, but they tested neither real drugs nor generics.
Weak evidence: Authors' interpretation of a placebo-only experiment, offered as a possibility; no brand or generic drug was tested.
Source: Waber 2008, JAMA Link to this fact
The industry / Other evidence
Scent marketing research is thinner than its reputation: a 2011 review found only about nine refereed studies of scent in shopping settings, and none had measured shoppers before exposure to check that scented and unscented groups started out alike.
Weak evidence: A narrative review by two authors with no stated search protocol; the count is theirs and may miss studies.
Source: Teller 2012, Journal of Marketing Management Link to this fact
The industry / Other evidence
In a shopping-mall field experiment (136 unscented vs 176 scented shoppers), a citrus-and-spice ambient scent had no significant effect on mall ratings, mood, spending, or time spent once each shopper's own starting attitudes were controlled.
Weak evidence.
Caveats
One mall, one scent, one scent week and one control week (not randomized), no scent concentration reported; the smell rating itself did not differ between groups, so the dose may have been low. Shows no effect here, not that scent never works. A later counterbalanced field test with register sales (de Groot 2022, one clothing store) found a very small positive spend effect for a product-congruent scent and none for citrus.
Source: Teller 2012, Journal of Marketing Management Link to this fact
The industry / Other evidence
The same mall data seemed to show scent benefits (higher ratings of the mall and more arousal) when analyzed with the simpler designs common in earlier studies; the apparent benefit disappeared once baseline differences between groups were accounted for.
Weak evidence: One dataset analyzed three ways by the authors; shows how the design can create false support, not that earlier studies were wrong.
Source: Teller 2012, Journal of Marketing Management Link to this fact
The industry / Other evidence
In the mall experiment, shoppers who said ambient scent was unimportant to them (109 of 312) spent less and visited fewer stores in the scented week, while for the other shoppers the only significant effect was a small positive one on rating of restaurants and services.
Weak evidence: Post hoc subgroup split, p < .05 without multiple-comparison correction, exploratory only.
Source: Teller 2012, Journal of Marketing Management Link to this fact
The industry / Other evidence
Two members of the original 1988 Nature high-dilution team were funded by the French homeopathic maker Boiron, according to this critique.
Weak evidence: Stated without citation in a secondary account; the Davenas paper itself does not state funding.
Source: Grimes 2012, Focus Altern Complement Ther, Focus on Alternative and Complementary Therapies Link to this fact
The industry / Other evidence
The store-scent boost to mood and ratings appeared only in shoppers who report intense emotional reactions; there was no effect in low-affect-intensity shoppers.
Weak evidence: Interaction terms were not significant (Table 4); effect rests on simple-slope tests at one SD with one-sided p-values.
Source: Doucé & Janssens 2013, Environment and Behavior, Environment and Behavior Link to this fact
Information, not medical advice. See also: myths we won’t tell you.