Building the company
7 Powers
Hamilton Helmer's taxonomy of durable advantage — benefit plus barrier — and the order startups usually unlock them.
Hamilton Helmer's 7 Powers is the cleanest taxonomy of durable advantages. A Power is a condition that creates both a benefit (better cash flows for you) and a barrier (something that stops rivals from matching you). All seven matter. Startups usually unlock them in a predictable order.
Concept
A moat is any structural defense. Helmer's Powers are a specific list: each one must have a benefit and a barrier. "We have a great team" fails the test — a rival can also hire.
The seven
| Power | Benefit | Barrier | Startup example |
|---|---|---|---|
| Scale economies | Unit costs fall as volume rises | Rival must match volume to match cost | Cloud infra; ad platforms |
| Network economies | Product value rises with users | Rival's empty network is less useful | Marketplaces, social, messaging |
| Counter-positioning | New model incumbents cannot copy without self-harm | Incumbent's old P&L blocks the response | Netflix vs Blockbuster; cloud vs on-prem |
| Switching costs | Customers face cost/risk to leave | Rival must pay that cost for the customer | ERP, design systems, data warehouses |
| Branding | Willingness to pay / trust premium | Hard to buy overnight with ads | Consumer brands; developer trust |
| Cornered resource | Preferential access to something scarce | Resource is not freely available | Unique dataset, talent, IP, license |
| Process power | Embedded organization that executes better | Culture and routines are slow to copy | Toyota production; elite ops cultures |
Scale economies are the classic software story: serving the millionth customer costs almost nothing compared with the first. That is why SaaS can run ~70–80% gross margins. AI products often weaken this power because inference is real COGS — scale still helps, but the slope is flatter.
Network economies are different: the product gets better as more people use it. Types live on network effects.
Counter-positioning is how startups beat incumbents without matching their cash. Salesforce's multi-tenant SaaS undercut on-prem CRM vendors who lived on large license deals. If the incumbent copies you, they destroy their own book; if they ignore you, you take the future.
Switching costs accumulate after product-market fit: integrations, trained users, historical data, compliance, workflows. Salesforce and Snowflake are sticky because ripping them out is a multi-year program.
Branding in B2B is often "trusted default": Stripe for payments, Figma for design. Cornered resource must be hard for a rival to acquire at similar cost — most "proprietary data" claims fail. Process power shows up late and is the hardest for a young startup to claim honestly.
Common mistakes
- Claiming all seven. Early companies usually have a seed of one.
- "We have proprietary data" that a rival can scrape or buy at similar cost.
- Calling process power a year-one story. Routines are not yet a barrier.
Check yourself
What two things must a Power have?
A benefit (better cash flows for you) and a barrier (rivals cannot match you cheaply).
Why is counter-positioning powerful against incumbents?
Copying you would cannibalize their existing revenue. Ignoring you cedes the future.
Key takeaways
- A Power is benefit plus barrier. Features fail the test.
- Startups usually grow switching costs and networks first, then scale, brand and process.
- AI weakens classic scale economies because inference is real COGS.
- Counter-positioning beats cash when the incumbent's P&L blocks the response.
Further learning
7 Powers (Hamilton Helmer)7 Powers – by Hamilton Helmer7powers.comNext, types of network effects.