Screening Singapore for Costco Market Entry.
A weighted country-screening model narrowed the comparison to Singapore, South Korea, and Japan. The important work came after the ranking: deciding whether Costco’s operating model could actually fit the market.
My role
I spearheaded the country-screening tool: model structure and formatting, input research, criteria development, data validation, and score reconciliation. I also contributed pricing/financial research and cross-module QA for the final recommendation.
Turning country comparison into an explicit decision model
The tool weighted Benefits at 50%, Costs at 30%, and Risks at 20%, using purchasing power, GDP growth, vehicle accessibility, corruption, infrastructure/business conditions, labor/property cost, inflation, property rights, and demographics.
Singapore ranked first—but the score was only the beginning
Change the actual criterion weights
The original tool used twelve criteria—not just three category totals. Move any criterion to see how its final 1–10 country ratings change the ranking. The initial settings reproduce the submitted 7.50 / 6.25 / 5.35 scores.
Original weights: income 20%, growth 20%, vehicle ownership 10%; corruption 5%, infrastructure 5%, legal costs 10%, labor 5%, property 5%; social/political unrest 5%, inflation 5%, property rights 5%, and aging population 5%. When edited values no longer sum to 100%, the explorer normalizes them proportionally before scoring.
Why the highest score still created an operating problem
Singapore’s purchasing power, growth, infrastructure, and business environment were attractive. But low vehicle ownership and higher property/labor costs cut directly against assumptions embedded in Costco’s traditional bulk-warehouse model. A score could tell us which country deserved deeper work; it could not tell us how the business should be localized.
From screening to recommendation
I also conducted pricing and financial research and contributed to membership-pricing, demand/revenue, and startup-capital analysis, plus substantial cross-module QA and final formatting. The project became a useful precursor to my later Monte Carlo work because it exposed the same conceptual boundary: a strategic factor can be important without automatically becoming a financial coefficient.
From weighted screening to stochastic simulation.
This project exposed the boundary between strategic importance and financial sensitivity. The later Monte Carlo research kept that boundary explicit, then added empirical uncertainty, dependence, financing capacity, rollout rules, and valuation over a 15-year horizon.