Guijiang Value Comparison Research Framework

Problem

When a company lacks a clear peer set, research can easily turn into a pile of company-specific facts: reading only its own reports, or lining up a few competitors without a clear rationale. This skill turns “finding mirrors” into a structured workflow. It helps analysts build a comparable-company pool from business similarity, financial structure, market-cap or capital allocation patterns, and mature overseas benchmarks, then compares 3-5 companies across business model, competitive advantage, financial quality, and valuation. The focus is not just whether a stock looks cheap, but whether the target company’s business essence, competitive edge, and margin of safety can be defended.

How it works

The workflow is anchored by the Guijiang value-investment principles: buy low PB, understandable business, and survival. First, it builds the peer pool by scanning direct competitors, substitutes or alternative solutions, upstream and downstream players, financial-profile screening, and overseas “century benchmarks.” Second, it compares companies on three layers: the core business model and ecosystem position, financial quality and cycle position, and valuation with market sentiment. It uses metrics such as ROE, ROIC, net margin, asset turnover, leverage, and free cash flow / net income to test earnings quality. It also compares PB, PE, dividend yield, historical percentile, and PB-ROE positioning, and checks ROE / PB as a lens for implied shareholder return. Third, it flags anomalies: whether a premium reflects durable growth or narrative risk, and whether a discount reflects temporary pain, hidden value, or structural weakness. It then produces a comparison card with peers, business-model differences, key data, insights, investment ratings, and next research steps.

Boundaries

This is best used for value-oriented equity research and peer-set analysis, especially when the question is “why this company rather than its competitors?” It is not a short-term trading signal and does not guarantee data accuracy. Analysts should mark data cutoff dates, adjust valuation metrics by industry such as P/EV for insurers and PB for banks, and keep qualitative judgment ahead of metric fitting. If long-term financials or cycle context are thin, the conclusion should remain more tentative.