Parallel Business — AI Decision Intelligence & Digital Twin for SMBs
Target Users: Small and medium business (SMB) founders, retail store owners, cafe/restaurant chains, and e-commerce entrepreneurs. The Problem: Over 70% of SMBs fail within their first 12 months. The primary cause is not bad ideas, but catastrophic blind financial decisions: 1. Arbitrary Price Changes: Increasing prices without knowing Price Elasticity of Demand (PED) triggers massive customer churn to competitors. 2. Premature Branch Expansion: Opening a 2nd location drains cash reserves before reaching the 6-month ramp-up breakeven, leading to fatal liquidity insolvency. 3. Lack of CFO Guidance: Hiring a professional Chief Financial Officer costs $2,000–$4,000/month, which SMBs cannot afford. Consequently, founders rely on gut feeling and static spreadsheets, risking real capital and bankruptcy to test basic business hypotheses.
Preliminary score
- Value for the user17.0 / 25
A specific problem and a clear outcome from solving it.
- Prototype and use of AI15.5 / 30
A working core scenario, and what AI actually contributes to the result.
- Quality testing12.0 / 20
Tests, examples of failures, and comparison with the current approach.
- Feasibility7.5 / 15
Data requirements, running costs and a clear next step.
- Originality5.5 / 10
A meaningful difference from common solutions.
What the judges wrote
- Claude Opus 5.5Preliminary55/100
- Strengths
- Your problem framing is concrete: price changes without elasticity data, second-branch expansion that drains cash, and the lack of affordable CFO guidance are real SMB decisions with real money at stake. Your testing notes describe specific bugs and how you fixed them, such as the linear elasticity formula collapsing volume to zero above +35% and annual profit hiding insolvency in months 4–5. The Cash Runway Deficit alert is a useful product insight that came out of that testing. A live, no-login demo with industry presets, a Streamlit engine and a clear subscription table make the prototype easy to explore.
- To improve
- Your materials do not show what GPT-4o mini or Claude actually do. The core engine reads as deterministic formulas plus Monte Carlo, and a static Netlify site suggests the AI Board and the NLP universe may be templated. Show the LLM calls, example prompts and outputs, and cases where the NLP parser misread a strategy. Back the comparison with Excel with a measured case, for example one real cafe's numbers, your forecast versus what actually happened, or advisor agreement. Also state your elasticity data sources, the cost per simulation, and a concrete pilot plan, and check the '70% fail in 12 months' statistic, which overstates published failure rates.
- GPT-6 AstraPreliminary60/100
- Strengths
- Your side-by-side pricing, expansion and marketing scenarios address concrete SMB decisions rather than a generic chat workflow. Your reported tests identify specific modeling failures, especially the important distinction between annual profitability and cash insolvency. Combining liquidity alerts with stress scenarios gives the proposed workflow practical value.
- To improve
- Show a reproducible end-to-end example, including simulation outputs and how an LLM translates a free-text decision into model parameters; the supplied materials leave that integration unverified. Validate your forecasts against real business records and a matched spreadsheet baseline, reporting numerical errors and an adoption metric such as planning time saved. Document benchmark sources, per-simulation operating costs and a focused next pilot; subscription prices alone do not establish operating feasibility.
The team
Nazrin AsgarovaLead
data
Aysun Aghayeva
designer
Ceshmexanim Gasimova
developer