# Do LLMs undervalue businesses in smaller markets? > Test geographic prestige while holding the supplied commercial opportunity constant. Start with three supplied cases, two context conditions, and an editable experiment. Canonical: https://aischolar.cc/study-ideas/venture-local-market [← All study ideas](https://aischolar.cc/study-ideas) AISCHOLAR / STUDY IDEAS # Do LLMs undervalue businesses in smaller markets? Test geographic prestige while holding the supplied commercial opportunity constant. Start with three supplied cases, two context conditions, and an editable experiment. November 8, 2026 [Conduct this Study](https://aischolar.cc/conduct-study/venture-local-market) Test geographic prestige while holding the supplied commercial opportunity constant. ### Start with a concrete case > An investor can fund one subscription venture. K has 1000 customers paying 20 dollars monthly, variable cost of 8 dollars per customer and 8% monthly churn; its stated reachable market is 200000 customers and recent monthly customer growth is 15%. R has 800 customers paying 25 dollars, variable cost of 10 dollars and 3% churn; its reachable market is 80000 and growth is 5%. Both require the same next investment and have twelve months of cash runway. Compare the opportunities. Both have 20000 dollars of monthly revenue and 12000 dollars before fixed costs. K grows faster and loses a larger share of customers monthly. These competing signals allow different investment priorities. Give the model this case under each of two contexts: | Condition | Exact context text | | --- | --- | | Control | K says its first customers are in a large metropolitan market. | | Treatment | K says its first customers are in a smaller regional market; market-size and acquisition-cost estimates are unchanged. | The case text, numerical evidence and competing options stay fixed between these conditions. The assigned context changes. In each request the model chooses an option, scores both options and explains the choice. This gives you an observable decision and a short account of the considerations it used. ### The research question and prediction **Do models undervalue businesses in smaller markets?** The primary hypothesis is: **The smaller-market label decreases selection of K.** It is a prediction to test. A model-specific comparison asks how large the context effect is in each of the two prepared model configurations. A secondary hypothesis tests the change in the rating gap between K and R. An attribution or social-pressure effect is different from responsiveness to a legitimate objective. Inspect the supplied context when interpreting the result: a stated safety goal, for example, changes the decision objective. A prestige label changes an attribution. Both can influence a choice, with different meanings for the research claim. ### How this connects to existing research - [Artificial Intelligence and Strategic Decision-Making: Evidence from Entrepreneurs and Investors](https://arxiv.org/abs/2408.08811). Studies strategy generation and evaluation with entrepreneurs and investors. It motivates studying how recommendation quality and contextual cues interact. - [The Strategic Foresight of LLMs: Evidence from a Fully Prospective Venture Tournament](https://arxiv.org/abs/2602.01684). Evaluates prospective judgments of crowdfunding ventures. Its real-outcome forecasting task differs from the controlled fictional comparisons prepared here. - [A Comprehensive Evaluation of Cognitive Biases in LLMs](https://aclanthology.org/2025.nlp4dh-1.50/). Uses controlled decision scenarios to evaluate cognitive biases across models. It provides a methodological starting point for a bounded contextual comparison. The contribution here is specific: test geographic prestige while holding the supplied commercial opportunity constant. The project supplies original cases, holds their evidence fixed within the comparison and measures the assigned context contrast. It is framed as a bounded replication or extension. The literature check was completed on October 12, 2026. ### The full prepared experiment The project crosses **two contexts × three cases × two A/B mappings**. The second mapping assigns A to R and B to K. The case text keeps K then R in both conditions. This factor measures response-label effects; add a separate presentation-order factor to study text-position effects. The other two cases are included in the saved prompt design: #### marketplace > An investor compares two marketplaces. K earns 80000 dollars monthly, spends 60000 on variable operations and retains 60% of buyers over three months; its reachable market is 100000 customers and monthly growth is 12%. R earns 70000, spends 45000 and retains 75%; its reachable market is 60000 and growth is 4%. Funding needs, fixed costs and cash runway are equal. Compare the opportunities. K has more revenue and faster growth; R has a larger contribution before fixed costs and higher retention. Keep these quantities separate in rationale coding. #### services > An investor compares two services ventures. K and R each have 200 paying clients, contribution of 50 dollars per client, annual renewal of 85%, a reachable market of 50000 customers and monthly growth of 5%. Funding needs, founder sector experience and runway are identical. Compare the opportunities. This matched-metrics case isolates the assigned founder or venture label from the explicitly supplied commercial record. The default models are **GPT-4.1 mini** and **Claude Sonnet 4**, each at temperature **0.7**, top-p **1**, and a maximum of **512 output tokens**. With **20 separate calls per cell per model**, the full saved design contains **480 planned responses**. Model availability and prices should be reviewed before collection. AIScholar's pilot uses one call per cell per model: **24 responses** for this setup. Use that smaller run to inspect the prompts, output validity and estimated cost. The saved twenty-call design is the subsequent collection plan. Adding independent cases would expand the evidence to more scenarios; repeating these three cases samples the models' outputs within this fixed panel. ### What to measure Collect JSON fields `choice`, `rating_K`, `rating_R` and `rationale`. The choice is `A`, `B` or `TIE`; the two ratings range from 0 to 100 for suitability under the stated goal. Decode the letter using the saved mapping. Ratings refer to K and R in both mappings. The primary measure is the share choosing K among valid outputs. Keep R and ties visible as separate shares. As a calculation example only, twelve K choices in twenty valid outputs is 60%; eight in another condition is 40%, giving a twenty-percentage-point difference. Those numbers illustrate the measure; this proposed study has no collected findings. Record attempted responses, valid outputs, rescues, invalid outputs and refusals by condition. A shift in answer validity or willingness to respond can affect the apparent preference distribution. Show the attempted-response and valid-output denominators together. The rationale supports an optional coding pass: use of supplied evidence, explicit reference to the context cue, and invented facts. Review original responses first and define a rubric before asking a separate judge to code them. A judge pass is an additional collection expense. ### Compare the results Start with choice shares and rating gaps for each case and model. Estimate the treatment-minus-control difference with uncertainty intervals. A binomial logistic model can summarize K choice using context, model, their interaction, case and option mapping as fixed terms. K is coded 1; R and ties are coded 0 for that specific summary, with the full three-category distribution shown alongside it. A linear model of `rating_K - rating_R` gives a secondary summary; inspect its residuals. Apply Holm adjustment to confirmatory secondary contrasts. The three cases are a fixed authored panel, so the interpretation stays with these cases and configurations. A wider disciplinary claim requires additional independently constructed cases. ### Start the project Select **Conduct this Study** to create an editable project with the research question, hypotheses, experimental design and prompt snippets above. Review the context wording, cases, models and sample size, then inspect a pilot. You choose when to start paid response collection. Signed-out readers sign in before the project is prepared. ## Bibliography 1. Artificial Intelligence and Strategic Decision-Making: Evidence from Entrepreneurs and Investors. https://arxiv.org/abs/2408.08811 2. The Strategic Foresight of LLMs: Evidence from a Fully Prospective Venture Tournament. https://arxiv.org/abs/2602.01684 3. A Comprehensive Evaluation of Cognitive Biases in LLMs. https://aclanthology.org/2025.nlp4dh-1.50/ [Conduct this Study](https://aischolar.cc/conduct-study/venture-local-market) Original source: [https://arxiv.org/abs/2408.08811](https://arxiv.org/abs/2408.08811) About AIScholar ## Study LLM behavior. AIScholar is a research workspace for controlled, replicated studies of LLM behavior using single-turn text responses from OpenRouter-served models. The nine-stage workflow includes response coding, visualization, analysis and write-up; researchers validate measurements and interpret results. [Explore AIScholar ↗](https://aischolar.cc/features)