This isn’t a hypothetical use of AI in a professional kitchen. Dawn McClung, culinary innovation manager at Piada Italian Street Food, has described using ChatGPT as a trusted sounding board for new product research — including pulling competitor menus to research beverage trends before developing the chain’s seasonal Italian sodas. Sodexo reports cutting recipe development time from three days to half a day using AI for seasonal planning and trend analysis, generating dishes with higher success probability before the culinary team focuses on actual testing.

Independent restaurants don’t have a Sodexo-sized R&D budget. But the underlying method — competitor research plus seasonal produce plus structured AI ideation — works exactly the same with a free ChatGPT account and an hour on a quiet afternoon. This guide builds that exact process as a repeatable weekly ritual.

Step 1 — Pick 5 Competitor Restaurants to Track

Choose the restaurants you genuinely compete with for the same customers — similar price point, similar cuisine style, similar location radius. Not every restaurant in town, and not aspirational restaurants three price tiers above you. Five is the practical number: enough to spot a genuine trend in what’s appearing locally, without the weekly task becoming unmanageable.

Step 2 — Save Their Current Menus

Copy the menu text directly from each competitor’s website or delivery platform listing (Uber Eats, Deliveroo, Just Eat all display full menus). This takes a few minutes per restaurant and needs no paid tool — just a document where you paste each one, updated weekly or whenever you notice a change.

Step 3 — List What’s in Season Locally

Note what’s currently available and at its best from your suppliers or local markets. This is the ingredient that turns generic AI suggestions into something genuinely usable — a dish idea built around produce you can actually source well and affordably right now, not a generic “seasonal vegetables” placeholder.

Step 4 — Run the AI Dish Ideation Prompt

With competitor menus and your seasonal produce list ready, paste both into ChatGPT or Claude with this prompt:

You are a chef and menu development consultant. Here are the current menus of my 5 closest competitor restaurants:

[PASTE COMPETITOR MENUS]

Here's what's in season locally right now:
[LIST YOUR SEASONAL PRODUCE]

My restaurant style is: [describe your cuisine and positioning, e.g. "modern British bistro, mid-range pricing, locally sourced ingredients"]
My target food cost percentage is: [X]%

Suggest 3 new dishes that:
1. Use the seasonal local produce listed above
2. Would genuinely stand out from what these 5 competitors are already offering
3. Fit my target food cost percentage

For each dish, give me:
- A dish name
- A full ingredient list with approximate quantities
- A brief recipe outline (method, not just ingredients)
- A suggested menu price
- An estimated food cost percentage based on typical ingredient prices

Be specific — no generic "seasonal vegetable medley" type suggestions.

Step 5 — Generate a Dish Photo for Each Idea

Once you have the dish concepts, generate a visual reference using an AI image tool — this helps you and your team picture the dish before committing kitchen time to testing it. Describe the dish in detail (key ingredients, plating style, colours) for the most useful result.

Step 6 — Confirm Price Point and Profitability

Frequently Asked Questions