Methodology

How We Build BeverageIndex

BeverageIndex aggregates publicly reported beverage prices into a single, searchable terminal. Here’s exactly what’s behind the numbers — what we track, where it comes from, how we normalize it, and what it does and doesn’t claim.

Data sources

• Numbeo — community-submitted cost-of-living entries for cappuccino, beer, wine, and soft drinks across 100+ cities.
• Kaggle public datasets — historical alcohol and beverage price tables released under open licences.
• Editor curation — we manually review outliers and cross-check against menu boards and supermarket aggregators where possible.

Coverage

As of the current snapshot, we track approximately 30 drinks across 98 cities (~2,940 city × drink price points). Coverage is uneven: espresso, beer, and bottled water are the densest. Cocktails, specialty teas, and regional spirits often appear in only a handful of cities — those pages are explicitly labelled as a limited sample.

How prices are calculated

• Statistic shown — we report a trimmed mean (top and bottom 10% removed) of available submissions per city × drink. Where only one or two data points exist, we show the raw value and label the sample size.
• Venue type — most prices reflect a mid-range restaurant or café for prepared drinks, and a typical supermarket for bottled/packaged drinks. Bar and hotel prices are excluded unless specifically noted.
• Tipping & service — base price only; service charges and tips are not included.
• Taxes — VAT/sales tax included where standard for the venue type in that country.

Currency conversion

Local prices are stored in the source currency. The USD value is computed using the per-city currency_rate_usd snapshot taken on the data refresh date. We do not back-fill old prices with new exchange rates.

Refresh cadence

Source datasets are pulled on a quarterly cadence. Individual price points show their updated_at field; anything older than 12 months is flagged as stale and excluded from rankings.

Rankings & comparisons

Global and continental rankings are computed across cities that have a price point for that exact drink. A "global average" built from only 2–3 cities is labelled as a limited sample rather than treated as a reliable global benchmark.

What we don’t claim

• We are not a real-time pricing API. Numbers reflect aggregated public data, not what you’ll pay tonight.
• We do not represent any single venue, brand, or supplier.
• Cocktail recipes and cultural notes are editorial colour, not professional bartending or hospitality advice.

Spot a problem?

If a price looks wrong, a city or drink is missing, or you’d like us to add a source, please contact us.