The Frankenproduct Problem

There's a product failure mode I see in almost every growth-stage B2B SaaS company I audit. The product started coherent. The first version had a clear model: one job, one workflow, one mental map.

Then the roadmap happened.

A sales request here. A competitor feature there. A founder idea that got shipped without research. An enterprise deal that required a custom configuration panel. Two years later, the product has everything — and nothing works simply.

I call this the Frankenproduct.

What is a Frankenproduct?

A Frankenproduct is a product that has accumulated so many disconnected features, navigation patterns, and workflow branches that it no longer has a coherent experience model. Users can't form a reliable mental map of how it works. The product is technically capable but experientially incoherent.

The signs are:

• New users take weeks to understand what the product actually does
• Power users have developed workarounds for common tasks the product technically supports but makes difficult
• The same concept appears in multiple places under different names
• Navigation has layers that were added at different times and don't follow the same logic
• The settings panel has grown into a second product of its own
• Support tickets cluster around the same five workflows that have never been streamlined

Most teams recognise the symptoms. They usually misdiagnose the cause.

The real cause isn't design debt — it's decision debt

Frankenproducts are built by teams that make feature decisions faster than they make experience decisions. Every feature shipped without a clear answer to "how does this fit into the product's existing model?" accumulates as experience debt.

This is different from technical debt. Technical debt slows down engineers. Experience debt slows down users — which slows down retention, which slows down revenue.

The distinction matters because the fix is different. Technical debt gets paid down by refactoring. Experience debt gets paid down by establishing a product experience model — a clear set of principles that govern how features should fit together — and then applying it.

Why a redesign is usually the wrong answer

The instinctive response to a Frankenproduct is "we need a redesign." In my experience, this is almost always the wrong move, for two reasons.

First, redesigns don't address the underlying cause. If the decision-making process that created the Frankenproduct doesn't change, the redesign will become a Frankenproduct too — just with newer CSS.

Second, redesigns break what's working. Users have adapted to your current interface, however imperfect. A full redesign forces relearning across every workflow, generating a wave of churn and support volume that arrives precisely when the team is least equipped to handle it.

What works better is a targeted experience repair programme, guided by an information architecture review and a pattern consolidation audit.

How to diagnose the severity

Not every messy product needs the same intervention. Before deciding what to do, you need to understand where the incoherence is worst.

Navigation audit. Map every top-level navigation item and every sub-navigation state. Look for: duplicate destinations (the same page reachable from multiple unrelated paths), orphaned sections (areas that aren't linked from the main navigation), and navigation items that represent different levels of abstraction sitting alongside each other.

Vocabulary audit. List every term used in the product to describe core concepts. Look for: synonyms (two terms for the same thing), overloaded terms (one term meaning different things in different contexts), and terminology that doesn't match how users describe their own work.

Workflow audit. For the five most common user jobs, map the current workflow step by step. Count the number of clicks, the number of context switches, and the number of concepts the user needs to hold in mind simultaneously. Flag anything that requires more than seven steps or crosses more than two contexts.

What to do instead of a redesign

The alternative to a full redesign is a phased consolidation programme:

Phase 1: Consolidate patterns. Pick the five most common UI patterns in the product (form layout, list view, detail panel, confirmation dialog, empty state) and standardise them. This alone will make the product feel significantly more coherent without touching any functionality.

Phase 2: Simplify navigation. Reduce top-level navigation items. Move infrequently used features to secondary locations. Ensure the navigation model reflects user mental models, not the product team's internal taxonomy.

Phase 3: Rename and clarify. Align product terminology with user language. This is usually one of the highest-impact, lowest-effort changes available. It doesn't require engineering time — it requires editorial discipline.

Phase 4: Streamline high-traffic workflows. Once the structure is cleaner, focus engineering effort on the five workflows that account for 80% of daily usage. Make each one faster, clearer, and more forgiving.

This approach is slower than a big-bang redesign but significantly less risky. It also keeps the product usable throughout the process.

The governance change that prevents recurrence

The hardest part of fixing a Frankenproduct isn't the design work — it's the process change.

Every feature decision needs to answer the question: how does this fit the existing experience model? If it doesn't fit, the options are: adapt the feature to fit, adapt the model to accommodate the feature (a deliberate, documented change), or don't ship the feature.

This requires someone in the room with the authority to say "the experience isn't ready" — and the expertise to explain why. In a company without a full-time Head of UX, that person is usually missing. This is the gap fractional UX leadership is designed to fill.

---

If you suspect you're building or have built a Frankenproduct, a structured UX audit is the fastest way to get a severity-ranked view of where the incoherence is worst — and what to fix in what order.