Back to Irreducible: Why Product Work Is Quietly Returning to the Basics

Last week, I sat in on two very different conversations.

In one, a product manager was passionately explaining first-principle thinking—how to strip a problem down to its irreducible parts and build from there. In another, a team was scrambling through a UX audit after a spike in churn, trying to understand why their “polished” product felt harder to use than it did six months ago.

At first, these felt like separate threads. One philosophical. One reactive. But the more I listened—across discussions about synthetic users, data migration testing, PSD3 compliance, even a major payment processor failing to send emails to Google Workspace accounts—the more it became clear that they were orbiting the same center.

We’re being pulled back to the irreducible.

Not because it’s trendy. Because complexity has finally caught up with us.

When Complexity Outpaces Understanding

There’s a particular kind of tension that shows up when products mature. You see it in SaaS companies layering on growth engines. You see it in fintech teams navigating new regulatory constraints. You see it when AI-generated personas promise research at scale.

The tools are more powerful than ever. According to McKinsey’s 2024 State of AI report, over 65% of organizations now regularly use generative AI in at least one business function—nearly double from the previous year. That acceleration is real.

But power compounds complexity.

Every new integration, automation, regulation, or growth tactic adds surface area. And surface area increases the distance between what we’ve built and what we truly understand.

I’ve felt this in my own work.

A few years ago, I joined a product team that had grown quickly. The UI looked modern. The metrics were steady. But support tickets told a different story: users weren’t confused about individual features—they were confused about how everything fit together.

When we ran a proper UX audit, we discovered something subtle but telling: the product had evolved through optimization, not intention.

• Features had been added to close sales gaps.
• Micro-interactions had been refined in isolation.
• Edge cases had been patched rather than reconsidered.

Each decision made sense locally.

Collectively, the product had drifted.

This is what happens when complexity outpaces understanding. We keep building, but we stop revisiting the fundamentals that make the system coherent.

And lately, across product conversations, I see a collective discomfort with that drift.

First Principles Aren’t a Mindset. They’re a Reset.

The phrase first-principle thinking gets used casually now. But in practice, it’s less about bold innovation and more about disciplined subtraction.

It asks uncomfortable questions:

1. What problem are we actually solving?
2. What must be true for this product to work?
3. What assumptions have calcified into “facts”?

When teams perform a UX audit that actually improves the product—not just documents it—they’re engaging in first-principle work. They’re peeling back accumulated decisions and asking: If we were designing this today, knowing what we know now, would we structure it this way?

In my experience, the most valuable audits don’t focus on visual inconsistencies or missed accessibility labels (though those matter deeply). They focus on structural coherence:

• Does the information architecture reflect how users think, or how we’ve organized internally?
• Do states and feedback patterns behave predictably across contexts?
• Are we solving the same problem in five slightly different ways?

One study from the Nielsen Norman Group found that inconsistent interaction patterns can increase task completion time by up to 22%. That’s not just friction. That’s cognitive tax.

First principles reduce that tax.

They force us to decide what is essential—and what is residue.

And residue, left unexamined, becomes debt.

Speed Is Seductive. Infrastructure Is Patient.

The trend toward synthetic users is fascinating to me. The appeal is obvious: faster feedback loops, scalable persona testing, fewer scheduling headaches.

But when we simulate users before speaking to real ones, we risk optimizing for plausibility instead of reality.

Synthetic feedback is often internally consistent. Real humans are not.

Real humans contradict themselves. They hesitate. They click the wrong thing and then explain why it made sense in the moment. They bring context that no training dataset can fully anticipate.

As designers, those inconsistencies are gold. They reveal where our mental models diverge from lived experience.

This is why infrastructure conversations—data migration testing, PSD3 compliance, feature engineering as a product decision—feel so important right now.

They remind us that:

• A feature is only as good as the data integrity beneath it.
• A growth engine is only sustainable if the underlying experience holds up.
• A compliance update is not a backend task—it reshapes the user journey.

When a major European payment processor can’t send emails to Google Workspace users, it’s not a UI problem. It’s an ecosystem failure. But users experience it as product failure.

They don’t see the layers.

They just see whether it works.

In one fintech redesign I led, we underestimated the impact of a backend permissions change tied to new regulatory requirements. On paper, it was a minor update. In practice, it altered the visibility of key actions across roles.

We had to re-map the entire interaction model.

That project changed how I think about product design.

Infrastructure is not beneath the experience.

It is the experience.

The Growth Story Is Being Rewritten

There’s another thread running through these conversations: growth narratives are being scrutinized.

From case studies about account-driven engines in niche SaaS to stories of founders inflating numbers, we’re seeing the cost of building on unstable foundations.

A 2023 study by Paddle found that involuntary churn accounts for up to 20–40% of total churn in SaaS businesses—often due to payment failures, expired cards, or billing friction.

That’s not a marketing problem.

That’s systems design.

Yet growth strategies often focus on acquisition before resilience. On top-line metrics before experience durability.

In design critiques, I sometimes ask a simple question:

If we stopped acquiring new users tomorrow, would the product still feel coherent and valuable to the ones who remain?

It’s an uncomfortable lens. But it clarifies priorities.

We recently redesigned a SaaS pricing page for clarity. The original version had six tiers, multiple add-ons, and fine print that required real concentration. Conversion wasn’t terrible—but it wasn’t strong either.

When we stripped it back to three clearly differentiated plans, removed redundant micro-copy, and aligned feature groupings to actual usage patterns, two things happened:

• Time spent on the pricing page decreased.
• Plan selection confidence (measured through post-purchase surveys) increased by 18%.

Clarity doesn’t always increase time on page.

Sometimes it reduces it—because understanding arrives sooner.

Growth built on clarity feels different from growth built on persuasion.

One compounds trust.

The other compounds fragility.

Designing for Irreducible Clarity

So what does this mean for us, practically?

From a craft perspective, returning to the irreducible changes how we design.

It pushes us to:

1. Design Systems as Thinking Systems

A design system isn’t a component library. It’s a shared logic.

If components don’t encode decisions about hierarchy, feedback, accessibility, and state behavior, they’re just reusable decoration.

When we treat the system as infrastructure—governing how patterns behave across contexts—we reduce drift before it happens.

2. Audit Assumptions, Not Just Interfaces

During UX audits, add a layer:

• What assumption led to this flow existing?
• Is that assumption still true?
• If we removed this entirely, who would actually be harmed?

Often, the answer is fewer people than we think.

3. Treat Backend Changes as Experience Changes

Data models, permission structures, feature engineering choices—these are product decisions. Not just technical ones.

Bring design into those conversations early. Map user impact before code solidifies around it.

4. Protect Time for Subtraction

Most roadmaps are additive. Rarely do they allocate cycles for removal.

But subtraction requires intention:

• Decommissioning features with grace.
• Consolidating redundant flows.
• Simplifying states.

In one quarterly planning session, we dedicated 15% of engineering time to “structural simplification.” No new features. Just coherence work.

It didn’t generate headlines.

It reduced support tickets by 12% over two quarters.

That’s the quiet power of irreducible thinking.

The Human Cost of Avoiding the Basics

Underneath all of this is something more human.

When products drift from their fundamentals, users feel it first as friction—and eventually as fatigue.

They don’t articulate it as “architectural incoherence.”

They say:

• “It used to be simpler.”
• “I’m not sure where to find things anymore.”
• “I guess I’ll just do it manually.”

As designers, we can become enamored with what’s possible. But the people using our products are just trying to get through their day.

Clarity is a form of care.

So is restraint.

So is the willingness to revisit foundational questions, even when it feels less glamorous than launching something new.

Across these conversations—first principles, UX audits, synthetic users, compliance shifts, growth engines—I don’t see a field chasing novelty.

I see a field rediscovering its responsibility.

We are being asked to understand our systems deeply enough to simplify them honestly.

To design products where the infrastructure, the interface, and the intention align.

To build things that work not just in demos or dashboards, but in the quiet, repetitive, human moments where software either helps—or gets in the way.

The irreducible isn’t abstract philosophy.

It’s the point where complexity stops being impressive and starts being accountable.

And maybe that’s the work in front of us now.

Not to build faster.

But to understand better.

And from there, to design with clarity that lasts.