Alfredo is one assistant with three standing jobs. Each one answers a question you already ask yourself.
The weekly board ranks recommendations from your own data, and every one you try comes back with an honest verdict: worked, no effect, or backfired.
The menu lab lets you test changes safely, and cost intelligence connects what each dish costs to make with what it sells for.
Customer intelligence finds your confirmed guests and regulars from data you already have. Competitive intel watches the venues around you.
A chat panel lives inside every tab. Ask how the week is going, why a number moved, or what to try next, and Alfredo answers in plain language from your own data. Every chart and every recommendation has an ask button right next to it.
And when your data cannot answer something, Alfredo says so instead of guessing. Same rule everywhere.
Ask it cost questions too, like how profitable an item really is. It answers from your own data and shows the queries it ran to get there. If it cannot verify a number, it says so instead of estimating.
Nothing new to buy: Alfredo connects to your existing POS system. Pricing →
Dashboards show you charts and leave the thinking to you. Alfredo reads your POS, hands you a short ranked list of recommendations worth trying, and when you try one, it measures what actually happened and gets smarter about your business. Suggest, try, verdict, repeat.
No new hardware, nothing to install, no staff retraining. Alfredo connects to the POS system you already run.
Projected +$11.40/wk if it holds. Here is what the weeks actually said:
green: on target · red: below · gray: not enough data that week
Units fell 22% and net revenue came out $41/wk below baseline. The dollar jump was too big for this item.
Rolled back to $5.50. The engine logged the miss, and next week's pricing suggestions learned from it. A tool that cannot say backfired cannot be trusted when it says worked.
A trial in Alfredo is one item, one change, for a set number of weeks. Every guest sees the same menu; results are compared against your own history and a control item, never against your customers. If a change backfires, it rolls back in minutes, and the board says so. This is how you were already improving your menu, just measured honestly.
Which items can carry a price move, which cannot, and a built in way to prove it on one item before you commit the whole menu.
Bundles and pairings pulled from your real baskets, with the co order evidence attached. Not your best seller. The pairing you would not have guessed.
Where each item sits on your menu and in your market: highlight the hidden gems, dim the underperformers, watch the gaps your competitors leave open.
Table time, dayparts, happy hour economics, item availability. Where the floor and the data disagree, you find out early.
Run Alfredo on your own POS data and judge it by its verdicts.
Alfredo works out who your confirmed guests and regulars are from the data you already have: your loyalty program if you run one, and the transaction data itself if you do not. Either way, no new app to launch and no sign up sheet at the host stand. See who keeps the lights on, how regulars differ from occasional visitors, and which patterns deserve a nudge.
And in keeping with the numbers rule: only confirmed guests are counted. Alfredo never inflates the panel with estimated crowd sizes.
Alfredo groups your confirmed guests by how they actually behave: when they come, what they order, how they spend. You rename the segments in your own words, because you know your room better than any algorithm.
Weekend evenings, bigger tables, beverage forward orders.
Weekday middays, fast turns, high repeat frequency.
Long dwell, beverage only checks. Your best nudge candidates.
Group organizers. Fewer visits, the largest checks in the house.
Alfredo tracks every venue around you: their ratings, their hours, their prices, and what their customers keep saying in reviews. Then it puts the pattern on your weekly board. Walking past a competitor? Photograph their menu, upload it, and see it matched item by item against your own sales.
Upload a photo of a competitor menu. Alfredo reads it, matches items against your own, and shows its confidence on every pairing. You confirm or dismiss each one before anything else happens.
| Their item | Your item | Match | Gap | |
|---|---|---|---|---|
| Smash Burger · $13.50 | House Burger · $11.00 | 92% | You are $2.50 under | |
| Loaded Tots · $9.00 | Tater Tots · $8.00 | 87% | You are $1.00 under | ✓ confirmed |
| Chicken Sando · $12.00 | Crispy Chicken · $12.50 | 64% | Needs your review |
Also inside the Competitive tab:
Run Alfredo on your own data and see who keeps the lights on.