Switching-Cost Wedge Assessment
Build an initial advantage, then make its accumulated value hard to replace.
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 95%
This assessment evaluates an AI business on two linked dimensions: the wedge that earns initial adoption and the switching cost that sustains the relationship. First, identify the specific workflow, audience, or capability that makes the product preferable today. Next, map the value accumulated through legitimate continued use, such as configured integrations, approved company context, trained workflows, or collaborative history. Then estimate what a customer would lose in time, performance, or continuity by moving elsewhere. The framework treats first-mover advantage as conditional rather than automatic: arriving early matters only if usage compounds into durable customer value. Teams should improve interoperability and portability while ensuring their product remains preferable because it works better, not because customers are unfairly locked in.
Origin
Extracted from Marketing Against the Grain during Kipp Bodnar and Kieran Flanagan's analysis of defensibility in enterprise AI businesses.
Core principles
- 01A wedge creates the initial reason to adopt.
- 02Accumulated customer value creates the reason to stay.
- 03Customer-specific data can strengthen a product over time.
- 04First-mover advantage matters only when adoption produces durable switching costs.
How to run it
- 1
Define the wedge
State the narrow advantage that persuades a specific customer to adopt the product before competing alternatives.
Pro tip Describe the wedge as a concrete use case rather than a broad claim about better AI.
Watch out A vertical label alone is not necessarily meaningful differentiation.
- 2
Map accumulated value
List the integrations, approved context, workflows, and collaborative knowledge that make the product more useful over time.
Pro tip Focus on value the customer can observe and measure.
Watch out Do not confuse preventable data lock-in with genuine accumulated value.
- 3
Run the switching test
Estimate the effort, disruption, and temporary performance loss involved in moving to an alternative.
Pro tip Include retraining, reconfiguration, validation, and employee adoption costs.
Watch out Nominal inconvenience is unlikely to protect a business from a substantially better competitor.
- 4
Stress-test differentiation
Ask what price reduction or functional improvement would persuade customers to switch despite the accumulated value.
Pro tip Test the answer against open-source and multi-model alternatives.
Watch out Assume competitors will eventually copy generic features.
- 5
Compound customer outcomes
Invest in mechanisms that improve results with responsible continued use while preserving reasonable export and exit paths.
Pro tip Track retention alongside measurable customer outcomes.
Watch out Artificial friction may increase short-term retention while damaging trust.
In the wild
An enterprise connects approved internal applications and data to an AI assistant. Over several months, teams configure workflows and improve retrieval quality. A rival model may be technically comparable, but migration requires reconnecting sources, validating permissions, and rebuilding established workflows.
→ Accumulated configuration and validated company context create meaningful switching costs.
The hosts compare enterprise AI to a successful advertisement whose platform algorithm has learned from prior performance. Replacing the ad resets part of that learned advantage, so a new creative starts behind the proven one.
→ The analogy shows how accumulated learning can make an incumbent asset difficult to replace.
Common mistakes
Treating a feature as a moat
A capability that a foundation-model provider can absorb is not durable differentiation by itself.
Ignoring the customer's wedge
High switching costs do not help if the product lacks a compelling initial reason for adoption.
Manufacturing lock-in
Blocking exports or creating unnecessary friction substitutes coercion for customer value and can erode trust.
Is it for you?
Best for
It is best for AI startups and software teams evaluating product differentiation and retention.
Not ideal for
It is not ideal for commodity products where customers accumulate little product-specific value.
From the transcript
“what your what's your wedge and what's the thing that's going to increase switching costs”
“this is the first mover Advantage truly matters in AI in this space if you can make that switching costs real”
“how do I make my switching costs like really high how do I add a lot of friction from a customer wanting to like go…”
From the episode
OpenAI’s GPT 4 Pro Will Run Your Business Better Than You… (#152)