AI Is Eroding the Two Moats SaaS Companies Depend On Most

In a panel discussion clipped to 70 seconds, a speaker lays out a concise argument for why AI will reshape the competitive landscape of SaaS. The claim is specific: two of the seven classical business moats -- switching costs and process power -- are losing their strength because of what large language models can now do. The framing draws directly from Hamilton Helmer's 7 Powers, the strategy book that gained mainstream attention through the Acquired podcast.

The 7 Powers Framework, Briefly

Hamilton Helmer's 7 Powers identifies seven durable sources of competitive advantage: scale economies, network effects, counter-positioning, switching costs, branding, cornered resources, and process power. The speaker, filmed at what appears to be a tech industry event in front of a dark stage backdrop with an audience of a few dozen, argues that AI is not weakening all seven equally. Some moats get stronger. Two specific ones get weaker.

Switching Costs Are Shrinking

The first moat under pressure is switching costs. SaaS companies have historically benefited from the pain of migration -- data locked in proprietary formats, workflows built around specific UIs, integrations that took months to configure. The speaker's point is that AI models can now act as a translation layer. If a model can read your data, understand your workflows, and port them to a competing product, the cost of switching drops. The lock-in that kept customers paying year after year becomes less reliable as a defensive strategy.

Process Power Is Getting Automated

The second moat is process power, which Helmer defines as advantage built through superior internal processes that competitors struggle to replicate. The speaker references Claude 4.7 specifically, noting that it can "hill climb anything" -- meaning you give it a target, tell it to iterate, and it will optimize toward that goal without manual intervention. Companies whose competitive edge rested on proprietary workflows and operational know-how face a problem: AI can reverse-engineer and replicate process advantages faster than humans can build them. The operational complexity that once took years to accumulate can now be approximated in far less time.

What This Means for SaaS

The speaker frames this as part of a broader "SaaS apocalypse," though the argument is more nuanced than the headline suggests. The claim is not that all SaaS dies. The claim is that the specific moats many SaaS businesses rely on -- the ones tied to stickiness and workflow complexity -- are the exact moats that AI degrades first. Companies built on network effects or cornered resources may be fine. Companies whose primary defense was "it's too painful to leave" or "our internal processes are too complex to copy" have a problem.

Key Takeaways

• Hamilton Helmer's 7 Powers framework is a useful lens for evaluating which SaaS moats AI threatens and which it does not.
• Switching costs weaken when AI models can translate data and workflows between competing products.
• Process power weakens when AI can autonomously iterate toward operational targets, replicating what previously took years of institutional learning.
• The speaker singles out Claude 4.7's ability to "hill climb" -- iterating autonomously toward a goal -- as a concrete example of process power erosion.
• SaaS companies whose moats rest on lock-in or workflow complexity are more exposed than those built on network effects, branding, or cornered resources.