Est. MMXXVIEN · es

menu engineering

GUERIDON

The science of the menu, served properly.

Open the menu ↓

Aperitif

The only advertisement every guest reads

Most restaurants agonise over the food and photocopy the menu. The menu is where the money is decided.

Every party that sits down reads one piece of marketing with total attention: the menu. It is read at the exact moment of decision, wallet already open, and yet it is usually the least designed object in the room. Menu engineering is the discipline that treats the carta as what it is — a portfolio of products with margins, demand curves and real estate — and edits it accordingly.

The discipline has a birth certificate: in 1982, Michael Kasavana and Donald Smith of Michigan State University published Menu Engineering, crossing each dish's popularity with its contribution margin — the euros a dish actually leaves behind, not the percentage. Everything since — the psychology of prices, the eye-tracking, the copywriting studies — hangs off that simple two-axis frame.1

A warning before the tasting begins: this field is full of folklore dressed as science. Where the evidence is solid we will say so; where it is a trade rule of thumb, we will call it doctrine; where a famous study has caveats, you will get the caveats. A menu built on verified findings beats one built on conference slides.

First course

Stars, plowhorses, puzzles, dogs

Two questions per dish: do they order it, and does it earn? Four answers, four very different jobs.

Kasavana and Smith's matrix classifies every dish inside its own category — mains compete with mains, not with coffee. A dish counts as popular when it captures at least 70% of an equal share of its category's sales: in a section of ten dishes, anything above 7% of orders. It counts as profitable when its contribution margin beats the category's demand-weighted average. Two cutoffs, four quadrants.1

Stars are popular and profitable: protect them, box them, never discount them. Plowhorses are loved but thin — the crowd-pullers whose price you nudge quietly or whose plate cost you engineer. Puzzles earn well but sit ignored: rename them, move them, have the server say their name out loud. Dogs do neither job; they occupy card and kitchen. Retire them or rebuild them — sentiment is not a line item.1

One refinement worth knowing: David Pavesic pointed out that averaging margins can flatter expensive, slow-selling dishes, and proposed weighting the contribution margin by what actually sells. The analyzer below uses the classic thresholds and Pavesic's demand-weighted margin — the version that keeps a €38 showpiece from grading the rest of the card on a curve.2

A menu is a portfolio. Every dish must justify its square centimetres.

Classify a dish
Star

Protect it. Feature it. Never discount it.

Category of 8 dishes → popularity cutoff is 70% of an equal share (8.75%). Margin cutoff here: €7.00.

Second course

The architecture of a price

Prices are read, not computed. What the eye reads first, and what the number feels like, are design decisions.

Start with the symbol. In a field experiment at the Culinary Institute of America's restaurant, guests who received a menu with bare numerals — no dollar sign, no decimals — spent about 8% more per cover than those shown symbolic prices. The currency sign is a small invoice delivered before the food arrives. Fine dining dropped it long ago; the finding says everyone else may quietly follow.3

Numeral-only prices, spend per check3+8%

Then the left digit. Thomas and Morwitz showed the one-cent drop only works when it changes the first digit read: 3.00 to 2.99 registers, 3.60 to 3.59 does not. Charm endings are not magic dust — they are left-digit engineering. And they carry a register: Wadhwa and Zhang found rounded prices feel right for hedonic, feeling-based purchases, while precise prices suit calculated ones. A tasting menu at 85 reads as confidence; at 84.90 it reads as arithmetic.19 20

Anchors and decoys do the quiet lifting. Huber, Payne and Puto showed that adding an option clearly worse than one alternative shifts choices toward the dominating option — the asymmetric-dominance effect. On a carta, the €38 chuletón nobody orders is not a failure; it is the reason the €26 entrecôte looks reasonable. Every category deserves one deliberate ceiling.4

Finally, the skeleton. The trade attributes to 'Omnes' four rules of price structure; the attribution is murky — no primary paper exists — but the doctrine has kept French menus honest for decades. The two that survive scrutiny: the dearest item in a category should not exceed roughly three times the cheapest, and the middle third of the price range should hold at least as many dishes as both extremes together. Break the first and your card reads schizophrenic; break the second and you are pushing guests to the edges you did not choose for them.6

Dispersion ceiling, dearest over cheapest63×
The cent that matters193.00 → 2.99
Snap a price to its psychology
10.209.95

€10.20 sits just past a left digit — money left on the table of perception. Charm endings keep the left digit low — the digit people actually read.

Audit a category — Omnes' rules, live
Dispersion (dearest ÷ cheapest)3.37× · broken
Middle band holds the majority0/6 · broken

One price per line. This runs the exact engine the analyzer uses.

Third course

How a menu is actually read

The famous 'golden triangle' has a problem: when researchers tracked real eyes, it wasn't there.

Menu folklore says the gaze lands mid-page, leaps to the top right corner, and that dishes placed in this 'sweet spot' sell themselves. When Sybil Yang put infrared eye-tracking on diners reading mock menus, she found something duller and more useful: people read menus sequentially, like a book, left page to right, top to bottom, slowing where the content interests them. No triangle, no magic corner. The practical lesson survives the debunking — if everything is read, everything is placement.7

Position does matter — just one level down. Dayan and Bar-Hillel manipulated where dishes sat within their own category list and found items at the top or bottom of the list chosen up to twice as often as the same dishes buried in the middle. The ends of a section are shelf space at eye level; the middle is where dogs go to hide. Put the dishes you need sold at the edges of their block, not the edges of the page.23

First or last in the section vs. buried23×2

And how long should the list be? The jam study made 'less is more' a slogan: 30% bought at a six-jam table against 3% at twenty-four. But when Scheibehenne pooled fifty experiments, the average choice-overload effect was essentially zero — it appears under specific conditions: unfamiliar options, no favourites, time pressure. A tourist reading their first Catalan carta is exactly that condition; a regular at their neighbourhood bar is not. Trim for the guest you actually serve, not for the slogan.21 22

Jam stand, 6 options vs 242130% / 3%
Mean overload effect across 50 experiments22≈ 0

Fourth course

Words you can taste

The description is part of the recipe. Diners eat what the label tells them they are eating.

The most quoted number in menu copy comes from Brian Wansink's cafeteria experiment: descriptive labels — 'Grandma's zucchini cookies', 'succulent Italian seafood filet' — lifted sales of the labelled dishes by 27%. Handle the number with tongs: it was a university cafeteria, 140 diners, and Wansink later left Cornell amid retractions of other work. The direction, though, is corroborated: in a companion study, the same dishes under evocative names were rated better tasting and more appealing, with favourable comments doubling.5 25

Descriptive labels, sales of labelled dishes5+27%

The cleaner experiment is older. Levin and Gaeth gave people identical ground beef labelled either '75% lean' or '25% fat': the lean framing won every rating, even after tasting — the frame literally seasoned the meat. On a menu this is free money: 'slow-braised eight hours' and 'braised since morning' describe the same pot, and only one of them feeds the imagination before the kitchen fires a single pan.24

House rules for copy that sells without lying: name the technique, the origin or the time — never the adjective alone. 'Delicious' is the word a menu uses when it has nothing to say. Sensory and specific beats superlative and generic; and every claim must survive the plate arriving. A description is a promise with a delivery window of eleven minutes.

Fifth course

The room is part of the menu

Tips, tempo and the weight of things: the psychology continues after the order is taken.

Tipping runs on reciprocity, and the field experiments are almost embarrassingly practical. A sweet with the bill lifts the tip — most when it feels personal and spontaneous. 'Thank you' handwritten on the check adds about two points. A light touch on the shoulder, a server who gives her name at a Sunday brunch — 23.4% against 15% — and, most striking, a Dutch waitress who simply repeated each order back word for word raised her tips by more than two thirds. None of this is about service quality; it is about being seen.8 9 10 11 12

Repeating the order back, average tip12+68%
Server introduces herself by name1123.4% vs 15%
Candy with the check, best condition8≈ +21%

Before the toolbox goes to your head, Michael Lynn's meta-analysis across 2,547 dining parties found the link between rated service quality and tip size positive but weak — tips are a social ritual more than a performance review. His survey work suggests servers who stack several of these small techniques earn meaningfully more. The house takeaway: train the gestures, not just the carrying of plates.13 14

The soundtrack sets the spend. Milliman alternated slow and fast background music in a restaurant: slow-music tables lingered eleven minutes longer and drank roughly 40% more at the bar, with food sales flat. North and colleagues played French or German music by a wine display: French days sold French bottles five to one; German days flipped the ratio — and almost nobody credited the music. Ambience is a pricing instrument that never appears on the bill.15 16

Slow music, bar spend15≈ +40%
French music, French vs German bottles165 : 1

Even the object in the guest's hands is talking. Identical yogurt from a heavier bowl was rated denser, more expensive-seeming and better liked; heavier menus raised the perceived class of the whole establishment before a single dish was read. This is why this page's namesake matters: the guéridon, the trolley wheeled to the table, is theatre by weight and presence — the physical proof that presentation is pricing.17 18

Nobody thinks the music chose the wine. The music chose the wine.

Digestif

The fine print is also design

Three rules with fines attached — and each one can be plated as hospitality instead of bureaucracy.

In the EU, fourteen allergens must be declared for unpackaged food — gluten to molluscs, sulphites to lupin — under Regulation 1169/2011. In Catalonia, article 15.2 of Llei 20/1985 bans promoting alcohol through offers 'announced as barra libre, 2x1, 3x1 or similar'; the happy-hour chalkboard is a fine waiting for an inspector. And since Spain's Ley 7/2022, article 18.3, every hospitality establishment must offer free non-bottled water alongside its own offer.26 27 28

Allergens of mandatory declaration2614

The engineering view: compliance lines are copy too. 'Ask us about allergens — we cook around them daily' reads as care, not as legal armour. 'Tap water, always, on the house' turns an obligation into the cheapest generosity on the card. The analyzer below checks all three automatically; turning the findings into charm is the house's job.

II

Put your menu on the pass

Upload your menu. Get the matrix, the price moves and the fixes — with the reasoning shown.

The analyzer →

Free · no sign-up · try it on a real pub's carta first

III

The receipts

Every claim above, with its source. Confidence noted where the literature is thin.

  1. 1Michael L. Kasavana & Donald I. Smith (1982)→
    Menu Engineering: A Practical Guide to Menu Analysis, Hospitality Publications (Lansing, MI); book, ISBN 9780932235015. Introduced the menu-engineering matrix: each item is plotted by sales volume (popularity) vs contribution margin, yielding four classes — Stars (high/high), Plowhorses (high popularity/low margin), Puzzles (low popularity/high margin), Dogs (low/low) — each with a prescribed action (promote, reprice, reposition, remove).
  2. 2David V. Pavesic (1983)→
    Cost/margin analysis: a third approach to menu pricing and design, International Journal of Hospitality Management, 2(3); DOI 10.1016/0278-4319(83)90033-6. Refined Kasavana & Smith by combining item food-cost percentage with the WEIGHTED contribution margin (margin weighted by units sold), so popularity enters the margin itself. Items classify as Primes, Standards, Sleepers and Problems; the goal is a sales mix that lowers overall food-cost % while maximizing total gross profit.
  3. 3Sybil S. Yang, Sheryl E. Kimes & Mauro M. Sessarego (2009)→
    $ or Dollars: Effects of Menu-price Formats on Restaurant Checks, Cornell Hospitality Report, Vol. 9 No. 8, Cornell Center for Hospitality Research. Field experiment at the Culinary Institute of America's St. Andrew's Cafe comparing three price formats: '$20.00', '20.' (numeral only), and 'twenty dollars' written out. Guests given the numeral-only menu spent significantly more than either the dollar-sign or written-out groups; a monetary cue ($ or the word) appears to trigger the 'pain of paying'. — Numeral-only prices (no $ symbol, no decimals): parties spent on average $5.55 more, about 8% more per check
  4. 4Joel Huber, John W. Payne & Christopher Puto (1982)→
    Adding Asymmetrically Dominated Alternatives: Violations of Regularity and the Similarity Hypothesis, Journal of Consumer Research, 9(1), 90-98; DOI 10.1086/208899. The original decoy / asymmetric-dominance paper: adding a third option that is clearly worse than one target (but not worse than the competitor) INCREASES the choice share of the target it is dominated by, violating the regularity axiom of rational choice. Demonstrated across six product categories (beer, cars, restaurants, film, lotteries, TV sets).
  5. 5Brian Wansink, James Painter & Koert van Ittersum (2001)→
    Descriptive Menu Labels' Effect on Sales, Cornell Hotel and Restaurant Administration Quarterly, 42(6); DOI 10.1177/0010880401426008. Six-week field experiment (university cafeteria, 140 diners): adding descriptive labels (geographic 'Traditional Cajun Red Beans', nostalgic 'Grandma's Zucchini Cookies', sensory 'Satin Chocolate Pudding') raised sales of the labeled items and improved attitudes toward food and restaurant plus repatronage intent. Caveat: it did NOT increase willingness to pay more for the item. — Descriptive menu labels increased sales of the labeled items by 27%
  6. 6Attributed to 'Omnès' (attribution disputed: many trade sources say André Omnès; L'Hôtellerie-Restauration credits consultant Jean Toulemonde and derives 'omnes' from Latin 'all') ()trade doctrine→
    Les principes d'Omnès (menu price-structure rules), French hospitality trade doctrine; no locatable primary academic publication. Best secondary source: L'Hôtellerie-Restauration fiche pratique. Four price-structure rules for a menu category: (1) range span — most expensive item no more than 2.5-3x the cheapest; (2) dispersion — split the range into three price bands; the middle band should hold at least as many items as the low and high bands combined; (3) supply/demand — average price actually paid divided by average price offered should sit around 0.9-1.0; (4) promotion — highlighted/promoted items should sit near the category's average price. — Highest/lowest price ratio within a menu category should not exceed 2.5-3
  7. 7Sybil S. Yang (2012)→
    Eye movements on restaurant menus: A revisitation on gaze motion and consumer scanpaths, International Journal of Hospitality Management, 31(3), 1021-1029. Infrared retinal eye-tracking on mock two-page menus: diners read menus sequentially like a book (left page top-to-bottom, then right), slowly and for information, rather than jumping to a 'sweet spot'. Found no support for the classic 'golden triangle' / upper-right sweet-spot doctrine; gaze did not concentrate above the midline of the right page. — No evidence of a menu 'sweet spot'; diners scanned sequentially like reading a book
  8. 8David B. Strohmetz, Bruce Rind, Reed Fisher & Michael Lynn (2002)→
    Sweetening the Till: The Use of Candy to Increase Restaurant Tipping, Journal of Applied Social Psychology, 32(2), 300-309. Two field experiments: giving candy with the check raised tip percentages via reciprocity, and the effect scaled with amount and especially with HOW it was offered — the biggest lift came when the server gave one candy per guest, turned away, then spontaneously came back and offered a second ('personalized' gift). — Relative tip increases: +3.3% (one candy), +14.1% (two candies), largest (~+21-23%) in the '1+1 spontaneous extra' condition (secondary sources report ~19.0% control vs ~23.0% tip)
  9. 9Bruce Rind & Prashant Bordia (1995)→
    Effect of Server's "Thank You" and Personalization on Restaurant Tipping, Journal of Applied Social Psychology, 25(9), 745-751. Field experiment in an upscale US restaurant: the server wrote nothing, 'Thank you', or 'Thank you' + her first name on the back of the check. 'Thank you' raised tip percentages; adding her name gave no extra lift. NOTE: this is the 1995 paper — the same authors' 1996 paper is the smiley-face-on-check study. — Writing 'Thank you' on the check raised tips ~11% in relative terms (about 2 percentage points, ~16.3% to ~18.1% per secondary sources)
  10. 10April H. Crusco & Christopher G. Wetzel (1984)→
    The Midas Touch: The Effects of Interpersonal Touch on Restaurant Tipping, Personality and Social Psychology Bulletin, 10(4), 512-517; DOI 10.1177/0146167284104003. Waitresses briefly touched customers on the hand or shoulder while returning change, or did not touch. Both touch conditions produced significantly higher tip percentages than no touch, with no difference by customer gender and no change in ratings of waitress or restaurant — the effect operated without awareness. — Touch raised tips from ~12% (no touch) to ~14% (shoulder) and ~17% (hand touch) — rounded figures from secondary sources
  11. 11Kimberly Garrity & Douglas Degelman (1990)→
    Effect of Server Introduction on Restaurant Tipping, Journal of Applied Social Psychology, 20(2), 168-172. Field experiment with 42 two-person Sunday-brunch parties: when the server introduced herself by first name, tip rate was dramatically higher than without introduction (p < .001). Small single-server study, so treat the size of the gap as indicative rather than universal. — Tips of 23.4% with a name introduction vs 15.0% without
  12. 12Rick B. van Baaren, Rob W. Holland, Bregje Steenaert & Ad van Knippenberg (2003)→
    Mimicry for money: Behavioral consequences of imitation, Journal of Experimental Social Psychology, 39(4), 393-398. Two Dutch restaurant experiments: a waitress who literally repeated customers' orders back (verbal mimicry) received larger and more frequent tips than when she paraphrased or just acknowledged. Replicated with a hypothesis-blind waitress in Experiment 2 (78% vs 52% of groups tipped; average tip 2.73 vs 1.36 guilders). — Repeating the order back increased the average tip by more than 68% (Exp. 1: 2.97 vs 1.76 guilders; verified in primary PDF)
  13. 13Michael Lynn & Michael McCall (2000)→
    Gratitude and gratuity: a meta-analysis of research on the service-tipping relationship, Journal of Socio-Economics, 29(2), 203-214. Meta-analysis of 13 studies (2,547 dining parties, 20 restaurants): the correlation between service evaluations and tip size is positive and statistically significant but MUCH weaker than commonly assumed — tips are a poor incentive/measure of service quality; situational and social factors dominate. — Positive but weak service-tip relationship across 2,547 dining parties in 20 restaurants
  14. 14Michael Lynn & Michael McCall (2009)→
    Techniques for Increasing Servers' Tips: How Generalizable Are They?, Cornell Hospitality Quarterly, 50(2); DOI 10.1177/1938965509334221. Internet survey of 1,066 US restaurant servers: those who more frequently used experimentally-tested tip-enhancing behaviors (introducing by name, squatting at the table, writing on checks, small gifts, etc.) reported larger tips relative to coworkers, across many server and restaurant types — supporting the generalizability of the field-experiment toolkit Lynn compiled in 'Mega Tips'. — Lynn's companion 'Mega Tips' manual estimates using several techniques together can raise a server's tips by 10-30%
  15. 15Ronald E. Milliman (1986)→
    The Influence of Background Music on the Behavior of Restaurant Patrons, Journal of Consumer Research, 13(2), 286-289; DOI 10.1086/209068. Field experiment in a mid-Atlantic restaurant alternating slow- vs fast-tempo background music on weekend nights: slow tempo lengthened the meal and markedly raised bar/beverage revenue, while food revenue was essentially unchanged. Estimated gross margin was higher on slow-music nights. — Slow music: 56 min average dinner vs 45 min with fast music; bar/drink spend up ~40% (~3 more drinks per table); food sales not significantly different
  16. 16Adrian C. North, David J. Hargreaves & Jennifer McKendrick (1999)→
    The influence of in-store music on wine selections (full report; brief report 'In-store music affects product choice', Nature 390:132, 1997), Journal of Applied Psychology, 84(2), 271-276; Nature 390, 132 (1997), DOI 10.1038/36484. Supermarket field study: French and German wines displayed side by side while French accordion or German Bierkeller music alternated daily for two weeks. Music nationality steered purchases toward the matching country's wine, and shoppers were largely unaware of (or denied) the influence. — French-music days: 40 French vs 8 German bottles sold (5:1); German-music days: 22 German vs 12 French (2:1); only 6 of 44 questioned shoppers acknowledged the music's influence
  17. 17Betina Piqueras-Fiszman, Vanessa Harrar, Jorge Alcaide & Charles Spence (2011)→
    Does the weight of the dish influence our perception of food?, Food Quality and Preference, 22(8). Participants tasted identical yogurt from visually identical bowls differing only in weight (held in hand): yogurt from the heaviest bowl was rated denser, expected to be more expensive, and was liked more; perceived flavour intensity did not change. The tactile weight of what the diner holds (bowls, and by extension menus, bottles, cutlery in the authors' related studies) transfers to perceived quality and price. — Same yogurt from a heavier bowl was rated denser, pricier-seeming and better liked than from a lighter bowl
  18. 18Vincent P. Magnini & Seontaik Kim (2016)→
    The influences of restaurant menu font style, background color, and physical weight on consumers' perceptions, International Journal of Hospitality Management, 53, 42-48. 2x2x2 experiment (italic vs plain font, gold vs white background, heavy vs light menu): a physically HEAVIER menu increased perceptions that the restaurant is upscale and will deliver better service; italic font had a similar upscale effect; background color (gold vs white) showed no significant effect. This is the study to cite for 'menu weight signals quality' (the Piqueras-Fiszman/Spence line covers plateware, bottles and cutlery). — Heavier menus significantly raised perceived upscaleness and anticipated service quality; gold vs white background made no difference
  19. 19Manoj Thomas & Vicki Morwitz (2005)→
    Penny Wise and Pound Foolish: The Left-Digit Effect in Price Cognition, Journal of Consumer Research, 32(1), 54-64. Five experiments showing WHY charm prices work: 9-ending prices are perceived as smaller than a price one cent higher ONLY when the left-most digit changes ($2.99 vs $3.00 — yes; $3.59 vs $3.60 — no), because magnitude encoding starts with the left digit. The effect strengthens when compared prices are numerically close. — The 1-cent drop matters only when it changes the left digit ($3.00 -> $2.99), not otherwise ($3.60 -> $3.59)
  20. 20Monica Wadhwa & Kuangjie Zhang (2015)→
    This Number Just Feels Right: The Impact of Roundedness of Price Numbers on Product Evaluations, Journal of Consumer Research, 41(5), 1172-1185. Five studies: rounded prices (200.00) are processed fluently and 'feel right' when the purchase is feeling-driven (hedonic, celebratory — e.g. champagne for a holiday), increasing purchase likelihood; precise, non-rounded prices (198.76) fit cognition-driven, utilitarian evaluations better. Rationale for round prices on fine-dining/experience menus and charm prices in value contexts. — Rounded prices boost purchase intent for feeling-based (hedonic) purchases; non-rounded prices for cognition-based ones — a fluency ('feels right') mediation
  21. 21Sheena S. Iyengar & Mark R. Lepper (2000)→
    When Choice is Demotivating: Can One Desire Too Much of a Good Thing?, Journal of Personality and Social Psychology, 79(6), 995-1006. The 'jam study' and two follow-ups: an extensive display (24 jams) attracted more browsers than a limited one (6 jams), but conversion collapsed with more options. IMPORTANT NUANCE: subsequent meta-analysis (Scheibehenne et al. 2010) found the average choice-overload effect across 50 experiments is near zero — the effect is real only under moderating conditions (unfamiliar options, no dominant option, time pressure, no clear preferences). — ~30% of tasters bought jam at the 6-option booth vs ~3% at the 24-option booth (verified in the paper: 31 vs 4 buyers)
  22. 22Benjamin Scheibehenne, Rainer Greifeneder & Peter M. Todd (2010)→
    Can There Ever Be Too Many Options? A Meta-Analytic Review of Choice Overload, Journal of Consumer Research, 37(3), 409-425. Meta-analysis of 63 conditions from 50 experiments (N = 5,036): the mean effect size of 'choice overload' is virtually zero, with large between-study variance. Choice overload is not a law — it appears only under specific conditions (option complexity, lack of expertise/preferences, no dominant option). The honest counterweight to the jam study. — Mean choice-overload effect size across 50 experiments (N=5,036): virtually zero
  23. 23Eran Dayan & Maya Bar-Hillel (2011)→
    Nudge to Nobesity II: Menu Positions Influence Food Orders, Judgment and Decision Making, 6(4), 333-342. One lab and one real-restaurant study manipulating where items sit WITHIN their menu category list: items at the beginning or end of their category were substantially more chosen than the same items placed mid-list. Empirical support for placing high-margin items at category edges (and a within-category effect — they did not test across categories). — Items at the top or bottom of their category list were up to twice as popular as the same items in the middle (~55% vs 45% of selections at category level)
  24. 24Irwin P. Levin & Gary J. Gaeth (1988)→
    How Consumers Are Affected by the Framing of Attribute Information Before and After Consuming the Product, Journal of Consumer Research, 15(3), 374-378. Classic attribute-framing study: ground beef labeled '75% lean' was rated better tasting and less greasy than identical beef labeled '25% fat'. The framing effect shrank (but persisted) after actually tasting the product. Foundation for positive-attribute framing in menu copy. — '75% lean' beat '25% fat' on all evaluation dimensions for the identical product; effect diluted but not erased by tasting
  25. 25Brian Wansink, Koert van Ittersum & James E. Painter (2005)→
    How descriptive food names bias sensory perceptions in restaurants, Food Quality and Preference, 16(5), 393-400. Companion to the 2001 sales study: evocative names ('Succulent Italian Seafood Filet' vs 'Seafood Filet') made the identical dish taste better, look more appealing and even seem more caloric, and doubled positive post-meal comments. Descriptive labels shift the sensory experience itself, not just the click/order decision. — Descriptive names raised taste ratings (7.31 vs 6.83) and appeal (6.66 vs 5.87) for identical dishes; positive comments doubled
  26. 26European Parliament and Council (2011)→
    Reglamento (UE) n.º 1169/2011 sobre la información alimentaria facilitada al consumidor — Anexo II, Diario Oficial de la Unión Europea L 304, 22.11.2011 (Anexo II, p. L304/43). Annex II lists the 14 substances causing allergies/intolerances whose declaration is mandatory (including in non-prepacked food served by restaurants, per Art. 44; in Spain implemented by RD 126/2015): 1 cereales con gluten (trigo, centeno, cebada, avena, espelta, kamut), 2 crustáceos, 3 huevos, 4 pescado, 5 cacahuetes, 6 soja, 7 leche (incl. lactosa), 8 frutos de cáscara (almendras, avellanas, nueces, anacardos, pacanas, nueces de Brasil, pistachos, macadamias), 9 apio, 10 mostaza, 11 granos de sésamo, 12 dióxido de azufre y sulfitos (>10 mg/kg o mg/l SO2), 13 altramuces, 14 moluscos. — 14 mandatory-declaration allergens (Anexo II), verified against the Official Journal text
  27. 27Parlament de Catalunya (1985)→
    Ley 20/1985, de 25 de julio, de prevención y asistencia en materia de sustancias que pueden generar dependencia — artículo 15.2, DOGC / BOE (texto consolidado BOE-A-1985-18553); apartado 15.2 añadido por la Ley 18/2009, de 22 de octubre, de salud pública de Cataluña (d.f. 1.1). Article 15.2 (consolidated text) prohibits in Catalonia the promotion of alcoholic drinks in bars/restaurants/venues via offers announced as 'barra libre', '2x1', '3x1' or similar that incite abusive or unlimited consumption, and bans advertising such promotions by any means. Happy-hour-style alcohol deals fall under this prohibition; the wording was introduced in 2009, not in the original 1985 text. — Art. 15.2: 'Se prohíbe la promoción de bebidas alcohólicas... mediante ofertas que se anuncian con nombres como "barra libre", "2x1", "3x1", o similares, que inciten al consumo abusivo o ilimitado' (verified in BOE consolidated text)
  28. 28Cortes Generales (España) (2022)→
    Ley 7/2022, de 8 de abril, de residuos y suelos contaminados para una economía circular — artículo 18.3, BOE-A-2022-5809, en vigor desde el 10 de abril de 2022. Article 18.3 (waste-prevention measures) obliges hospitality and restaurant establishments to ALWAYS offer customers the possibility of consuming non-bottled (tap) water free of charge, complementary to the establishment's own offer. Selling bottled water remains legal, but the free option must exist and no direct or indirect charge (glass, ice, service) may be applied. — Art. 18.3: '...en los establecimientos del sector de la hostelería y restauración se tendrá que ofrecer siempre a los consumidores... la posibilidad de consumo de agua no envasada de manera gratuita y complementaria a la oferta del mismo establecimiento'