Why use SavvySipper?
Picture this: you're settling in for a lovely meal and get handed the wine list. With 5 sparkling, 23 whites, 4 rosés, and 20 reds, it's enough to overwhelm even a seasoned sommelier.
Then comes the real puzzle: you just fancy a glass. Now you're juggling measures... 125ml, 175ml, 250ml, or a 500ml carafe... across 3 sparkling options, 8 whites, 3 rosés, and 6 reds with up to four size choices each.
That’s a serious matrix of pricing data. How do you know if a 500ml carafe of Gavi actually gives you a fairer pour than the 175ml? Or if the venue's mark-up is reasonably fair vs the wine's retail price?
That’s why we built SavvySipper. In just a few moments, our AI Wine List Wizard:
- Scans & analyses the entire menu.
- Calculates the "Glass Tax" to find the absolute best-value pour.
- Cross-references Vivino scores and surfaces hidden sommelier gems.
Scroll down to see the magic in action.
The Mistley Thorn
SavvySipper's Verdict
Score83/100
Markup+166%
Vivino3.9
GLASS TAX+10%
SavvySipper analysed 52 bottles: 23 Whites•20 Reds•4 Rosés•5 Sparkling. Plus 20 glass options: 8 Whites•6 Reds•3 Rosés•3 Sparkling.
What would you like to see?
The Mistley Thorn Wine List Scores
Score Breakdown:
Raw Averages Extracted:
- Menu Average Mark-up: 166% (Lose 1 pt for every 10% above 100% across the menu)
- Menu Average Crowd Score: 3.9 (Lose 1 pt for every 0.1 decrease below 4.5 across the menu)
- Menu Average Glass Tax: 10%(Drops 1 pt for every 2% of glass tax applied. Benchmarked exclusively on the standard 175ml pour.)
- Menu Discovery: 25 / 25 (AI extracted: 52 bottles (5 pts), 42 grapes (10 pts), and 29 regions (10 pts))
SavvySipper scores are 100% algorithmically generated. Claim your menu to identify hidden margin leaks, highlight competitive pours, and see exactly what your customers see.
Why did The Mistley Thorn's wine list scan score 82%?
▼
Our internal OCR and Extraction engine evaluates menu legibility during the scan. A score of 82% indicates that some data on the physical menu (such as specific vintage years or exact producer designations) was missing or difficult to read.
When this happens, it forces our AI to extrapolate and find the closest match across global pricing APIs. We dock the confidence score to be fully transparent, indicating that some prices or ratings on this list represent a highly educated "best-guess" scenario based on the available text.