ShelfGap Buyer Lens

The description states that ShelfGap Buyer Lens is a buyer-choice simulator inside ChatGPT. It allows merchants to describe their offer, competitors' offers, and buyer profiles. It then shows which offer each buyer picks, why they lost others, and what the smallest change worth testing would be.

It uses two components:

• A GPT-5.6 model for conversation and explanation
• A deterministic TypeScript engine for scoring, ranking, and simulation

The tool is described as read-only, meaning it does not crawl or modify anything, and it operates within a ChatGPT app interface.

Evidence The author states that the product simulates buyer choices using buyer profiles with hard requirements, budget, delivery limits, and weights. It compares fit, cost, delivery, trust, and convenience. It also allows for "what-if" simulations of small changes like showing verified features or lowering shipping costs.

Inference The system is built to be deterministic — same input, same output — but the GPT handles conversation while the engine handles calculations.

The description states that ShelfGap sits between two questions:

• Buyer: "Which offer fits me?"
• Merchant: "Why did they choose the other one?"

It positions itself as a tool for merchants who don't want another dashboard or account, but who still need to understand buyer behavior.

The author claims it helps merchants find the smallest decision worth testing next — not just predictions or forecasts, but actionable insights based on simulated buyer choices.

Evidence The product is positioned as a ChatGPT-native solution that avoids traditional dashboards and requires no setup. It emphasizes that recommendations are based on real buyer profiles and evidence, not assumptions.

Inference The positioning reflects an attempt to simplify complex buyer behavior into digestible, testable actions — though the description does not indicate whether this has been validated with actual merchants or customers.

The description states that ShelfGap is for e-commerce merchants who do not write code and do not want another dashboard. It targets those who ask "why did the buyer pick my rival?" and who are looking for actionable insights without technical overhead.

It also mentions that the interface is a conversation, making it accessible to users who have never opened a terminal.

Evidence The target customer is described as online store owners or merchandisers who want to understand buyer behavior but do not want to use complex tools or accounts.

Inference The ICP appears to be small-to-medium-sized e-commerce sellers who are looking for smarter merchandising decisions, though no specific segment (e.g., food, electronics) is named.

The description does not provide any information about pricing, monetization, or business model. It only states that the tool is read-only and requires no API keys or accounts to use.

Evidence Not evidenced.

Inference The product appears to be in a demo or beta phase, with no indication of how it will be monetized or whether it has any revenue streams.

The project was built using:

• GPT-5.6 (for conversation)
• TypeScript engine (for deterministic logic)
• React, Next.js 15
• OpenAI SDKs and MCP tools
• Docker + Caddy + Dokploy for deployment

It is described as a pnpm monorepo with shared Zod contracts, unit tests, end-to-end tests, and a public HTTPS endpoint.

Evidence The author describes the architecture as a ChatGPT app with a read-only engine, deterministic logic, and no crawling or modification of data.

Inference The technical stack suggests a modern, scalable approach to building AI-powered tools, but there is no evidence of production use or scalability beyond the demo.

The description states that this project was submitted to the OpenAI 2026 hackathon and includes a live demo at shelfgap.onurkeskin.com/demo. It also mentions that it has a public architecture brief.

However, there is no evidence of:

• Revenue
• Customers
• Adoption
• Product usage metrics
• Any form of traction beyond the demo

Evidence Not evidenced.

Inference The product appears to be in an early-stage prototype or beta phase, with no indication of real-world usage or customer validation.

The description does not mention any competitors. It focuses on the unique positioning of ShelfGap as a ChatGPT-native tool that avoids traditional dashboards and provides deterministic buyer simulations.

Evidence Not evidenced.

Inference The competitive landscape is unknown, but it likely competes with tools that help e-commerce merchants analyze buyer behavior or optimize product listings — though no specific names or products are mentioned.

• No traction or revenue evidence: The project appears to be in a demo or prototype stage with no indication of real-world adoption.
• Unverified claims: The tool is described as deterministic and read-only, but there’s no independent verification that these features work as claimed.
• Limited scope: The product is built for ChatGPT only, which may limit its reach or scalability.
• No monetization strategy: No information on how the product will be sold or funded.

Evidence Not evidenced.

Inference The lack of any real-world data, customer feedback, or revenue indicates a high risk that this is an unproven concept with no clear path to commercial viability.

1. What specific buyer behavior patterns have you observed in your simulations? Are there any real-world validations?
2. How do you plan to scale beyond the current demo and into actual merchant use cases?
3. Have you tested this with real merchants or only fictional scenarios?
4. Is there a roadmap for monetization or pricing models?
5. What are the technical limitations of running this at scale, especially in terms of GPT-5.6 usage and deterministic engine performance?
6. How do you plan to validate that the recommendations are actionable and effective?