The Art of Crafting Investment Premises: A Practical Guide

The quality of your investment premise directly determines the quality of insights you'll receive. After analyzing thousands of queries from our users, we've identified clear patterns that separate great premises from mediocre ones.

This guide will help you craft premises that surface truly differentiated opportunities.

The Anatomy of a Good Premise

Effective premises share several characteristics:

• Specificity: Narrow enough to filter out noise
• Thesis: Contains an implicit view about the world
• Actionability: Points toward investable securities
• Differentiation: Not already consensus

Let's break down each element.

Specificity: The Goldilocks Zone

There's a sweet spot between too broad and too narrow.

Too Broad (Useless):

• "Good tech stocks"
• "Companies that will go up"
• "Undervalued stocks"

These return generic results because they don't express a differentiated view.

Too Narrow (Limited Results):

• "Semiconductor companies in Phoenix with female CEOs"
• "Retailers in Maine with negative working capital"

These might match zero or one company—not enough for analysis.

Just Right (Actionable):

• "Enterprise software companies with 40%+ gross margins expanding into federal contracts"
• "Regional banks with deposit costs below 2% and commercial real estate exposure under 15%"
• "Consumer brands with pricing power demonstrated by positive unit growth despite 2024 price increases"

These return 5-20 companies—enough for a portfolio, few enough to research deeply.

Second-Order Thinking: Where the Edge Lives

First-order thinking identifies obvious beneficiaries. Second-order thinking finds the non-obvious ones.

First-Order (Crowded):

"AI companies will benefit from AI growth" → NVDA, MSFT, GOOGL

Everyone knows this. It's priced in.

Second-Order (Less Crowded):

"Data center REITs with expansion capacity in low-energy-cost regions will benefit from AI infrastructure demands"

This identifies the enabling infrastructure, not the headline technology. The beneficiaries are less obvious and potentially mispriced.

Third-Order (Contrarian):

"Natural gas producers with pipelines to data center clusters will benefit from AI power demands that can't be met by renewables alone"

This is genuinely differentiated—many ESG-focused investors won't touch it, creating potential mispricing.

Combining Trends: 1 + 1 = 3

Some of our most successful user premises combine multiple trends into a single thesis:

• "Healthcare companies using AI to reduce clinical trial costs" — combines AI + healthcare cost pressure
• "Retailers with strong e-commerce platforms targeting aging demographics" — combines digital commerce + demographic shifts
• "Industrial companies with robotics exposure benefiting from reshoring" — combines automation + deglobalization

The intersection of two trends is less crowded than either trend alone. If Trend A is well-known and Trend B is well-known, the companies benefiting from both A and B might be overlooked.

Questioning Consensus: The Contrarian Approach

If everyone believes something, the opportunity may already be priced in. Contrarian premises ask: what happens if consensus is wrong?

Examples:

• "Homebuilders that will benefit if mortgage rates stay higher for longer than expected" — contrarian to the "rates will fall" consensus
• "Oil services companies positioned for sustained capex if the energy transition takes longer than planned" — contrarian to aggressive decarbonization timelines
• "Companies with significant China revenue that will benefit if US-China relations improve" — contrarian to decoupling consensus

Contrarian premises require more conviction because you're betting against the crowd. But when they work, the payoff can be substantial.

Time Horizon: When Matters

Specify whether you're looking at near-term catalysts or long-term structural changes. The results will differ.

Near-Term (6-12 months):

"Companies with positive earnings revisions going into Q4 earnings season"

This captures tactical opportunities around specific events.

Medium-Term (1-3 years):

"Companies investing in automation that will benefit from labor cost inflation"

This captures multi-quarter trends playing out over business cycles.

Long-Term (3-10 years):

"Companies developing solid-state batteries for next-generation EVs"

This captures secular shifts that may take years to materialize.

Your time horizon should match your holding period. Don't use a 10-year premise for a 3-month trade.

Common Mistakes to Avoid

1. Describing What Already Happened

Bad: "Companies that benefited from COVID"

Better: "Companies with durable work-from-home tailwinds that persist post-pandemic"

2. Using Jargon Without Substance

Bad: "Disruptive innovation leaders"

Better: "Companies taking market share from incumbents in industries with legacy technology debt"

3. Ignoring Valuation

Bad: "The best AI companies"

Better: "AI companies trading below their historical P/E multiples despite revenue acceleration"

4. Being Too Clever

Sometimes simple premises work best. "High-quality companies with recent price weakness" is straightforward but effective.

Putting It Together: A Template

Here's a framework for constructing premises:

[Sector/Industry] companies with [specific characteristic] that will benefit from [catalyst/trend]

Examples:

• "Semiconductor companies with foundry relationships outside Taiwan that will benefit from supply chain diversification"
• "Healthcare services companies with high Medicare Advantage exposure that will benefit from demographic aging"
• "Industrial distributors with inventory management technology that will benefit from reshoring demand"

Iteration: The Key to Better Results

Your first premise is rarely your best. The iteration process:

1. Start with an initial premise
2. Review the results
3. Ask: Are these the companies I expected? Why or why not?
4. Refine the premise to be more specific or adjust the focus
5. Repeat until results match your thesis

The goal isn't to find premises that return the stocks you already like. It's to discover stocks you didn't know fit your thesis.

That's where the real alpha lives.