What Menu Personalisation Actually Means for a Restaurant

Menu personalisation is one of those phrases that gets used loosely enough to mean almost anything, from a server remembering a regular's usual order to a full recommendation engine borrowed from an online store. Strip away the marketing language and the actual idea is fairly simple: the same restaurant, the same dishes, the same prices, but the order a guest sees them in shifts based on what that particular guest tends to like.

What does not change

This is the part that gets lost in translation most often. Personalisation does not mean two guests eat from different menus, get different prices, or see dishes that don't actually exist on your kitchen's line. Every guest still sees the same categories: starters, mains, desserts, drinks. Every guest can still order anything on the menu, regardless of what the platform thinks they'll like. The dish list, the kitchen's actual capabilities, and the pricing stay exactly as they are for everyone.

What actually changes

• The order dishes appear within a category, based on a guest's stated or learned preferences
• Which items get surfaced as a quick recommendation when a guest seems undecided
• How a dish gets described, sometimes emphasising the angle most relevant to that guest, like spice level for someone who's flagged a preference for heat
• What a guest sees first when they reopen the menu on a return visit

None of this requires deception or hidden pricing tricks, which is a fair worry to raise and worth addressing directly. A personalised menu is closer to a good server who's waited on you before and has a sense of what you like, applied consistently and at a scale no single server could manage across hundreds of tables a week.

Where the preference signal comes from

Two sources feed a personalised menu. Declared preferences are things a guest tells the menu directly: a dietary restriction, a stated love of spicy food, an allergy. Learned preferences build up from behaviour over time: what a guest orders, what they linger on, what they skip past repeatedly. A new guest with no history yet still gets a reasonable starting point based on declared preferences and general patterns, and the menu gets sharper the more it learns, without ever requiring a guest to fill out a lengthy profile before they can order.

Why this is different from a loyalty programme

A loyalty programme rewards a guest after the fact, points, discounts, a free item after ten visits. Personalisation changes the experience during the visit itself, before an order is even placed. The two aren't competitors, a restaurant can run both, but they solve different problems: one is about retention through incentive, the other is about making a single visit, even a first one, feel more considered.

A concrete example of the same menu working two ways

Picture two guests scanning the same QR code at two different tables on the same night. One has flagged a preference for heavy spice and has ordered the same fiery curry on three previous visits. The other is a first-time guest who mentioned a mild palate when they set their preferences. Both see the exact same starters, the exact same mains, the exact same dessert list, at the exact same prices. What differs is which mains appear first: the fiery curry sits near the top for the regular, while a milder dish sits near the top for the newcomer. Neither guest is stopped from ordering whatever they want, and neither one even needs to notice that the order looks different from the table next to them.

Why a human still has the final say

Personalisation shapes what a guest sees first, but it doesn't make decisions on the restaurant's behalf without a check. An allergen conflict, for instance, gets flagged to the guest automatically before they can order it, but the kitchen still confirms the order before it goes out, the same way it always has. The system narrows and suggests, a person still signs off on anything that matters for a guest's safety. That division of labour is deliberate: automation is good at applying a rule consistently across thousands of orders a week, a kitchen staff member is still the right party to make the final call on something as serious as an allergy.

TasteGraph builds exactly this kind of personalisation: the same menu, the same categories and dishes every guest already expects, reordered around each guest's own declared and learned taste. It's live on the same QR code your guests already know, with nothing added for them to download or learn, and a restaurant is typically up and running with it in about 10 minutes.