PLM in Finance & Insurance Part 3: Why Platform Banking Demands a New Approach to Product Management

From owned channels to embedded finance

The way financial services reach customers has shifted dramatically. Instead of relying solely on their own branches, websites, and relationship managers, institutions now find their products embedded in journeys designed by others. Credit options appear at retail checkouts. Insurance offers surface during travel bookings. Banking capabilities are delivered through APIs to fintechs, platforms, and brands that sit closer to the customer. This move toward platform banking, embedded finance, and Banking-as-a-Service creates powerful new distribution opportunities, but it also introduces new layers of complexity in how products are defined and managed. 

In the traditional model, it was easier to think of a product as a single thing: one card product, one loan type, one policy form. In the platform model, that same product can be configured in different ways for different partners, channels, and regions. Each configuration may carry its own terms, pricing, risk profiles, and servicing arrangements. The result is that what looks like a simple product catalog from a distance is, up close, a dense web of variants. 

How platforms multiply product variants

To understand why this matters, consider a straightforward example. A bank designs a core lending product family. It might initially intend this to be the standard offering across its own channels and perhaps through a broker network. As it enters into platform partnerships, however, it begins tailoring that lending product to match specific partner propositions. One partner wants a streamlined application flow and particular repayment options. Another partner targets a different customer segment and requires different eligibility rules. A third partner operates in a jurisdiction with distinct regulatory constraints that necessitate altered disclosures and caps. 

Each of these adjustments creates a variant of the original product. Over time, as partners evolve, promotions come and go, and local rules change, further tweaks are made. Some of these variants are clearly documented, others emerge through small configuration differences in systems. If the institution does not deliberately model these variants, it can lose sight of how many truly exist and how they relate to one another. 

The same dynamic plays out in insurance. Where a core coverage construct may take on different characteristics depending on whether it is sold directly, through brokers, or as an add-on to other services. The pattern is consistent: platforms multiply variants. 

The risk of hidden variants

When product variants are not explicitly managed, they become hidden. They live in configuration tables, code branches, special-case procedures, and one-off agreements. In that state, they are difficult to oversee. Governance bodies see a high-level product family and perhaps a summary of channels and partners, but not the full richness of how the product behaves in each context. 

This hidden complexity can make the portfolio fragile in several ways. Onboarding new partners takes longer because teams must reinvent parts of the product each time, stitching together requirements and capabilities from multiple sources. Maintaining consistent standards across partners becomes challenging because there is no central view of which variants are in line with current policies, and which are not. Exiting a partnership or changing strategy for a channel becomes risky because it is not clear what will happen to customers on variants that were tailored for that partnership. 

Strategically, hidden variants obscure insight. It is hard to answer detailed questions such as which combination of product, partner, segment, and region delivers sustainable returns, or which configuration patterns tend to create more complaints or operational incidents. 

Applying PLM discipline to ecosystem products

PLM brings a way to surface and manage this complexity. Instead of treating each partner arrangement as a separate project, PLM encourages institutions to define product families and configurations systematically. A product family becomes a structured object that describes the core economic and risk characteristics, eligibility rules, and servicing models that apply broadly. Around that core are defined parameters that are allowed to vary, such as pricing ranges, certain fees, coverage extensions, service-level options, and branding elements.  

Partners and channels are then treated as dimensions in a configuration space. For a given product family, configurations are created to reflect specific combinations of partner, channel, region, and segment. Each configuration is explicitly linked to the family it belongs to, making it clear what is shared and what is unique. Lifecycle states are defined for these configurations, such as proposed, under review, approved, in pilot, active, and retired. Changes to configurations flow through governed workflows that ensure the institution remains within agreed boundaries and standards. 

In a PLM and digital thread platform, these product families and configurations are not static catalogs. They are living entities with histories and relationships. As partners request changes, as regulations shift, or as performance data suggests adjustments, those changes are captured in the platform. The institution can see how a given configuration has evolved and what decisions drove those changes. 

Using the digital thread to connect partners, controls, and performance

The digital thread complements this configuration discipline by tying the product space to other critical elements of the ecosystem. Partner contracts, for example, can be linked directly to the configurations that implement them, so that commercial terms and service-level commitments are understood in the context of specific product variants. Regulatory obligations and internal policies can be associated with product families and configurations, making it clear which rules apply to which variants across jurisdictions. 

Controls and monitoring activities can also be connected back into the product configuration space. For each configuration, the institution can specify which controls are in place, what tests are run, and what evidence exists that those controls are functioning. When a partner is added or a configuration is adjusted, the implications for controls are visible. When incidents or complaints arise, they can be linked to the relevant configuration, partner, and control environment, providing a basis for targeted remediation. 

Performance data completes this picture. Rather than viewing performance only at the level of broad product lines or channels, institutions can analyze results at the configuration level. They can see, for instance, how a particular partner configuration performs relative to similar configurations in other markets, or how changes in terms have affected outcomes within a configuration over time. 

Building strategic insight at the right level

One of the most powerful effects of this PLM and digital thread approach is that it moves strategic discussion to the level where meaningful variation actually occurs. Instead of debating abstractly whether a product or channel is working, leaders can examine specific configurations. They can identify which combinations of product family, partner, segment, and region are driving value and which are consuming resources without sufficient return. 

This granularity enables more thoughtful decisions about standardization and differentiation. Some configuration patterns may prove so consistently effective that they become reference designs for new partnerships, speeding up onboarding and reducing risk. Other patterns may be recognized as high-maintenance or prone to issues, prompting a decision to rationalize or retire them. Because the configurations and their histories are visible in the PLM and digital thread platform, these decisions are grounded in a clear understanding of what exists and how it behaves. 

A practical path toward a resilient portfolio

Moving to this level of discipline does not necessitate redesigning the entire portfolio at once. Many organizations start by focusing on one product family that is heavily used in ecosystem contexts, such as a particular lending or insurance product. They map the existing partner and channel landscape for that family, often discovering more variation than anyone realized. They then model the family and its variants in a PLM and digital thread platform, define lifecycle states, and begin managing new changes and partner requests through this structure. 

As they do this, they develop clear views of the relationships between product configurations, partners, obligations, controls, and performance. Onboarding new partners becomes faster because it builds on reusable configurations, rather than bespoke designs. Exiting or reshaping existing partnerships becomes safer because the institution understands precisely which customers and configurations are involved. Over time, the same patterns can be extended to other product families and lines of business. 

In a platform economy, resilience depends on being able to adjust your portfolio and partner mix without losing control. A PLM and digital thread platform provides you the ability to see your ecosystem products as they really are, to design and govern their evolution, and to base strategic decisions on a detailed, connected understanding of your offerings. That is the difference between being pulled along by the logic of platforms and actively shaping your role within them.