How the Best Bars Create Their Seasonal Menus

By a named editor

Fredrik Filipsson — Co-founder & Editor in Chief · LinkedIn ↗
Last reviewed 2026-04-17 · How we pick bars

The process of how bars create seasonal menus is rarely as romantic as it looks from the guest side. The finished menu — twelve cocktails, elegant descriptions, a coherent concept — is typically the product of months of research, dozens of rejected ideas, and at least one catastrophic testing session that set the timeline back three weeks. We spoke with bar directors across New York, London, and Paris about how their seasonal menu development actually works.

The Seasonal Menu Creation Process

Every bar has its own version of the process. Some start with an ingredient, some with a narrative theme, some with a season's produce. What they share is a structured testing protocol that filters dozens of ideas down to the handful of cocktails that survive to print. These are the bars whose development processes are most worth understanding.

Lyaness — London

South Bank $$$ Ingredient-led

Lyaness starts seasonal menu development from a single ingredient brief — typically a flavour or botanical that the team feels is underexplored. Each new menu reworks that base ingredient through multiple techniques and spirits combinations. The testing phase runs approximately eight weeks, with team tastings three times per week. Only cocktails that pass a blind preference test against the previous menu make the cut.

Average number of ideas tested before final menu: Approximately 60–80 concepts per seasonal launch

Little Red Door — Paris

Le Marais $$$ Narrative-led

Little Red Door builds each menu around a published conceptual document — a multi-page brief covering the season's theme, its cultural references, its intended emotional register. The cocktails are then reverse-engineered from that concept, with each drink required to represent a specific chapter of the overall narrative. Menus that do not cohere as a story are sent back to R&D regardless of how good the individual drinks are.

Concept-to-menu timeline: Typically six months from initial brief to launch night

Death & Co — New York

East Village $$$ Collaborative / Democratic

Death and Co runs one of the most democratic menu development processes in the industry. Any bartender on the team can submit a cocktail concept for consideration. Submissions are reviewed at monthly development sessions, and drinks that make it past the initial round enter a structured testing schedule. This bottom-up approach has produced some of the bar's most celebrated serves — and gives junior bartenders real ownership of the menu.

Proportion of menu from junior bartender submissions: Typically 30–40 percent of any given seasonal launch

The Testing Phase Nobody Sees

Menu development testing is where most ideas die. A cocktail that tastes excellent on day one often fails when it needs to be produced 30 times per night under service pressure. Dilution changes, ice temperature matters, batch stability becomes critical. The testing protocols that the best bars use are designed to surface these problems before they reach the floor.

Existing Conditions — New York

West Village $$$ Technique-first

Dave Arnold's approach to menu testing runs through formal sensory analysis protocols borrowed from food science. New cocktail submissions are evaluated against a standardised set of parameters: aroma, entry, mid-palate, finish, and texture. Cocktails that score inconsistently across different tasters are revised until the flavour profile stabilises. The result is a menu where every drink delivers reliably regardless of who makes it.

Testing standard: Every cocktail must pass three consecutive tasting panels before making the menu

Dante — New York

West Village $$$ Seasonal / All-day

Dante's menu development accounts for the bar's unusual operating hours — cocktails need to work as a midday aperitivo as well as a late-evening serve. This pushes the testing team to evaluate drinks at different times of day and at different service temperatures. A cocktail that works at 7pm over fresh ice may read completely differently at 1pm in a warm room — and both versions need to pass.

Key testing variable: Every cocktail must be assessed at both peak service temperature and at 15 minutes of dilution

Nightjar — London

Shoreditch $$$ Theatrical / Experiential

Nightjar's menu development includes a full theatrical presentation audit — every cocktail is tested not just for flavour but for how it looks when delivered, how it smells on arrival, and what the serving moment communicates to the table. A technically perfect cocktail that arrives without ceremony is treated as an incomplete submission. The experience of receiving the drink is considered part of the recipe.

Unique testing criterion: Tabletop presentation assessed in darkness with only ambient candlelight

Weekly editorial

The bars worth going to, weekly.

One email per week. The bars our editors are recommending right now, across 60 cities worldwide.

Free. Unsubscribe anytime.

The Final Edit — What Gets Cut

The hardest part of seasonal menu development is not creating good cocktails — it is cutting good cocktails to arrive at a great menu. Most bars we spoke with had the same experience: the final edit session produces more disagreement than anything else in the process, because by that point every member of the team has a personal attachment to something that does not make the cut.

Himkok — Oslo

Storgata $$$ Nordic / Farm-to-glass

Himkok's seasonal menus are constrained by a strict local sourcing rule — if an ingredient cannot be sourced from within Norway or the Nordic region, it does not appear on the menu. This forces the menu development team to solve flavour problems creatively rather than reaching for imported spirits. The constraint has produced some of their most distinctive and consistently awarded menus.

Menu constraint rule: Minimum 70 percent of ingredients by volume must be Norwegian or Nordic origin

Salmon Guru — Madrid

Chueca $$ Irreverent / Flavour-driven

Diego Cabrera's Salmon Guru edits the final menu by asking one question: would a first-time customer order this again on their second visit? Cocktails that test well in development but feel too challenging for repeat ordering are held back for the experimental section rather than appearing on the main menu. The main menu is designed for maximum return visits, not maximum critical acclaim.

Final edit criterion: Predicted reorder rate — any cocktail estimated below 25% second-visit reorder is reconsidered

Connaught Bar — London

Mayfair $$$$ Classic / Elevated

Agostino Perrone's menu edit process at the Connaught is guided by the hotel's guest profile. New cocktails must satisfy both the knowledgeable drinker and the guest who is ordering their first craft cocktail — a dual brief that eliminates anything too esoteric or too simple. The Connaught Martini has been on the menu since the bar's relaunch because it has never failed that test.

Menu longevity test: Each new cocktail is evaluated against the question "would we be proud to serve this in five years?"

Our Verdict

The best seasonal menus share a quality that is difficult to articulate but immediately recognisable: they feel considered. Not designed, not constructed — considered. Every drink occupies a clear position within the menu's logic, and there is nothing on the list that exists only because somebody in the kitchen thought it was clever. The bars that consistently produce menus like this treat the process as seriously as the product.

When you next sit at one of the bars on this list and order from a menu that seems effortless, remember that the effortlessness is the result of months of early mornings, failed experiments, and arguments about flavour. That is what a truly seasonal menu actually costs.