# Owner "Margin Manager" — concept prototype

**Built by Corey Brown · July 2026**
**For:** Owner.com, Principal Product Manager, AI Restaurant Experience (RestX) application
**Live prototype:** https://coreywbrown.com/lab/owner-margin-manager/
**Status:** Unlisted concept. Not affiliated with or endorsed by Owner.com. Menu, prices, competitor figures, commodity trends, and review counts are representative and rounded for the demo.

---

## Why this exists

The role builds RestX, Owner's operator command center: an "always-on co-owner that surfaces what matters, recommends the next best move, and increasingly does the work." So rather than describe how I would approach that, I picked the clearest gap in Owner's own promise and built a co-owner surface for it. Show, don't tell.

## The problem

Owner's promise to operators is **sales and profit growth**. But the operator app today is strong on sales and thin on profit. The dashboard reports sales, orders, top items, reviews, and SEO, and it flags where to focus, but it stops at the flag. Nothing looks forward at food cost, the single biggest lever on restaurant margin, and nothing acts. An owner can grow sales all quarter while a quiet run-up in ingredient prices erodes the profit underneath, and find out only when the month closes.

## The insight

A co-owner is not a better dashboard. It is the judgment a good GM brings: see the cost coming, weigh the move against the business, and do the work, not just surface a number. So the unit of the product is not a metric, it is a **move**: what is happening, why it matters in dollars, the recommended action, and one tap to do it. And the judgment has to include reputation, because the margin-optimal price is sometimes the wrong price.

## The solution

**Margin Manager**, an operator-app surface with three modules:

- **Forward-buy save.** Watches commodity and ingredient price trends and warns before a spike lands ("coffee is up, your supplier rises next Tuesday"), then offers to lock today's cost on the affected items. Forward-looking, not a rear-view report.
- **Price swaps.** When an ingredient cost pushes a dish under the margin floor, or a dish is priced below the local market, it recommends a price move with a **reaction-risk score** that tempers it. The score reads review signal, whether the item is a signature, and the size of the change. On a signature dish praised for its value, it recommends a smaller raise than the math alone would, and flags the aggressive move as "not recommended" while still letting the owner override.
- **On Shelf.** New or swapped-in items get an AI-generated photo so they are sellable the moment they exist (items without a photo convert worse), with the owner approving the image before it goes live.

Every decision on the surface, approve, schedule, dismiss, or override, collapses into a one-line record with the decision kept and a way to reopen and adjust, so the operator stays in control.

## Why it maps to the role

This is the RestX thesis, built: a co-owner that surfaces what matters, recommends the next move with a why and a one-tap how, and does the work. It lives on the operator surface the role owns, it attacks the profit half of Owner's promise that the product under-serves today, and every module maps to a playbook the job description names (profit-aware demand capture, self-optimizing menus, reputation-aware pricing, closed-loop actions).

## How it was built

- A single self-contained HTML/CSS/JS file, no framework, no backend, styled to read as Owner's operator app.
- Grounded in the real product: I worked through Owner's live operator surfaces (the app, reporting and analytics, reviews, and how-it-works) to model what the owner actually sees today and where the loop stops at a flag instead of an action.
- The reaction-risk score is the deliberate part. It is what turns a margin calculator into a co-owner, weighing profit against the brand equity a good operator would never spend carelessly.

## What I would do next with real data

- Wire the forward-buy and margin math to real supplier and commodity feeds and each restaurant's actual recipe costs, rather than representative values.
- Ground the reaction-risk score in the restaurant's own review corpus and order history: which items are truly signature, which price changes historically moved volume.
- A/B test the "moves" surface against contribution margin and profit retention, and measure how often owners accept, edit, or override, so the co-owner learns each operator's risk tolerance over time.