Blindness and Insight

Jane Jacobs looked at an ordinary city street and saw a self-organizing social order.

The urban planners she challenged saw land use, density, traffic, housing and infrastructure. What their maps missed was the life between those categories: shopkeepers watching the sidewalk, children moving between stoops and stores, strangers becoming familiar through repetition, the small exchanges that help a neighborhood regulate itself.

Jacobs was an urbanist and activist whose 1961 book, The Death and Life of Great American Cities, upended modernist planning orthodoxy. She did more than disagree with the establishment. She changed what counted as evidence that a city was working.

Her work taught me that powerful frames always bring something consequential into view. They also make other things harder to see.

How to see

The title of this essay comes from Blindness and Insight, a book by the Belgian literary critic Paul de Man. I borrow the phrase because it captures something fundamental about frames: what helps us see one part of reality more clearly can make another part harder to grasp.

A frame can't reproduce reality in its entirety. It selects. It draws boundaries. It makes certain relationships legible while pushing others toward the edge.

Successful frames don't remain ideas for long. They become products, processes, metrics, incentives and eventually institutions.

That's a feature of framing, not a bug. In fact, it's what makes framing possible. Without some way to distinguish signal from noise, the world is just an undifferentiated flood of information.

But the same mechanism that produces insight also produces blindness. A dashboard can expose one important pattern while hiding another. A category can clarify a market while obscuring adjacent forms of demand. An explanatory model can become so successful that contradictory evidence starts to look like noise.

The task is not to find a frame without blind spots. There is no such thing. The harder question is whether what a frame reveals matters more than what it causes us, for a time, not to see.

Frames harden

Business makes this dynamic unusually visible because successful frames don't remain ideas for long. They become products, processes, metrics, incentives and eventually institutions.

Southwest Airlines didn't simply build a cheaper airline. Its original model rested on a different way of seeing the market. For many short trips, the relevant alternative was not another airline. It was ground transportation.

Yesterday’s insight becomes today’s assumption. Today’s assumption can become tomorrow’s blind spot.

Once you see the market that way, different things matter. Frequency matters more. Fast turns matter more. Simplicity matters more. Low fares matter more. Many conventions of full-service aviation begin to look like unnecessary complexity.

The frame created coherence because it told Southwest what to ignore. That was a strength for decades. But frames do not stay propositions. They harden into operating models, habits and identity.

Eventually the question is no longer whether the frame helped the organization see something its competitors missed. It is whether the frame now prevents the organization from seeing what has changed.

Southwest has recently moved away from some practices that once defined it. Assigned seating replaced open seating in 2026, and checked-bag fees now apply to many fares. The point is not that the original frame was wrong. It is that no frame is timeless.

Yesterday’s insight becomes today’s assumption. Today’s assumption can become tomorrow’s blind spot.

SaaSpocalypse now

Enterprise software may be living through that cycle now. A generation ago, software was primarily something an organization bought, installed, maintained and periodically upgraded.

Salesforce helped make another way of seeing software legible when it launched cloud-delivered CRM in 1999. Software could be delivered continuously over the internet as a service.

Salesforce made the argument explicit in its early “No Software” campaign. The crossed-out software logo was deliberately cheeky, but the provocation attacked assumptions so familiar they had become nearly invisible.

Why should customers install the application themselves? Why should upgrades arrive as disruptive projects? Why should software be understood primarily as a product transferred from vendor to buyer?

Large language models and AI agents may now be exposing assumptions inside the SaaS frame that we had stopped recognizing as assumptions.

Other companies extended the same shift. Workday brought cloud delivery into finance and human capital management. My former employer ServiceNow was founded around a cloud-based platform for routing work through the enterprise. It evolved from IT service management into a much broader workflow platform.

Together, companies like these made the burdens of the old frame newly visible. Once the frame changed, whole systems changed with it: pricing, deployment, product development, vendor economics and customer relationships.

Then SaaS won. And in winning, the frame largely disappeared. What had once been a provocative way of seeing software became the industry's default assumption. That is the point when frames are easiest to mistake for reality, and their blind spots become hardest to see.

Large language models and AI agents may now be exposing assumptions inside the SaaS frame that we had stopped recognizing as assumptions. For the past quarter-century, enterprise work has increasingly been organized around discrete applications: CRM, finance, HR, IT service management. Users navigate interfaces to activate the capabilities beneath them.

But what if an agent can increasingly invoke those capabilities on the user’s behalf? What if the user asks for an outcome rather than navigating the software required to produce it?

ServiceNow currently sells AI agents that act across enterprise systems and an AI assistant through which users can make a request and have work carried through to completion. Salesforce is likewise positioning itself around agentic CRM.

Does the application remain the natural unit through which people experience software? Maybe. The application layer may prove extremely durable. Agents may simply become another interface sitting on top of existing systems.

Or perhaps the application is less fundamental than SaaS made it appear. We don't know yet, and that uncertainty is useful. A genuine reframing doesn't not announce itself by declaring a paradigm shift. It begins by revealing something the prevailing frame has trained us not to notice.

Authority > power

There is another, cruder theory of changing minds. A Vietnam-era maxim of uncertain authorship held that once you have people by the balls, their hearts and minds will follow. It’s a theory of power through leverage: control behavior first, and belief can accommodate itself later.

Frame-making works differently. Jane Jacobs couldn’t order planners to rethink the city. Southwest couldn’t compel travelers to see a short-haul flight as an alternative to driving. SaaS companies couldn’t force customers to rethink how software should be delivered.

Their influence came from making a different interpretation of reality useful enough that other people began acting on it.

Power can compel action inside an existing frame. Frame-making changes what the action appears to be about. That distinction matters because organizations have plenty of tools for exercising power. They include capital, managerial authority, pricing, distribution, contracts and technology.

Those mechanisms can change behavior, but authority requires something else. It emerges when other people begin using your explanation of reality to interpret their own experience.

Beyond strategy

There is a danger in treating frame-making as just another name for good business strategy.

Strategy asks what an organization should do given its understanding of the market. Positioning asks how customers should understand a company relative to alternatives.

Frame-making goes one level deeper by asking whether the market is being understood correctly in the first place. What counts as evidence? What belongs inside the category? What has been defined too narrowly? What assumptions have become invisible because everyone shares them?

That leads to a more useful question than the one most companies ask.

Not: What do we want people to think about us?

But: What does our market not yet know how to see?

Answering that question gives an organization a frame: a different way of explaining what is happening in the market and why it matters.

Frames become consequential when they can be defended. That requires evidence: proprietary data, research, customer outcomes, product behavior, operating experience, benchmarks or repeated observations that others can test against their own reality.

As that evidence accumulates, the frame becomes harder to dismiss. And when other people begin using it to interpret the market for themselves, the organization starts to earn authority.

The goal is not to make a claim louder. It is to make the old way of seeing increasingly inadequate.

What organizations should do

Look for inherited assumptions. Ask which parts of your market feel so obvious that nobody bothers to defend them anymore. That is often where blindness begins.

• Study anomalies. Pay attention to customer behavior, data or operating experience that doesn't fit the prevailing explanation. What looks like noise may be evidence that the frame is wrong.

• Build proof, not slogans. A new frame needs evidence that outsiders can examine, test and use. Otherwise it's mere positioning.

• Watch your own blind spots. Once your frame starts working, actively look for what it excludes. Success makes assumptions harder to see.

• Help other see. Distribution, scale and market leverage can force attention. Authority begins when others adopt your explanation because it helps them understand the world.

Markets are full of companies trying to provide better answers. The bigger opportunity is sometimes to discover that everyone is looking at the wrong question.

The moment your answer becomes common sense, the responsibility changes. You have to ask what your own way of seeing has made invisible.

Every frame creates insight. Every frame creates blindness. Authority comes from knowing the difference.

Sources

Jane Jacobs and urban planning. Jane Jacobs, The Death and Life of Great American Cities (1961); Jane Jacobs Walk, “Who was Jane Jacobs?”; Project for Public Spaces, “Jane Jacobs,” on her use of observation, mixed uses, sidewalks and community-scale planning.

Paul de Man. Paul de Man, Blindness and Insight: Essays in the Rhetoric of Contemporary Criticism (1971; revised 2nd ed., University of Minnesota Press, 1983)

Southwest Airlines. Southwest Airlines annual reports and company filings describing its historical short-haul, high-frequency, point-to-point, low-fare model; current Southwest materials on assigned seating and checked-bag fees.

Salesforce and SaaS. Salesforce, “Our Story,” on its 1999 launch and role in pioneering cloud-delivered CRM and software as a service; Salesforce materials on the early “No Software” campaign and crossed-out software logo.

ServiceNow and enterprise workflows. ServiceNow histories of Fred Luddy’s original cloud-based platform vision and the evolution of the ServiceNow platform; current ServiceNow materials on AI agents and work executed across enterprise systems.