Blog The Paradox at the Heart of Improving Operations Efficiency

The Paradox at the Heart of Improving Operations Efficiency

October 20, 2025

Last Updated: October 17, 2025

Read Time: 6 minutes

Authors: Eashwar Viswanathan

Innovation & Tech

Inst'l Asset Managers

When one of the world’s largest banks sets out to cut $2.5 billion a year from IT operations, every other financial services firm should reconsider its own technology strategy. Barclays, in this example, has engaged McKinsey ⁱ to streamline duplication, automation, and know-your-customer procedures.

Barclays is not alone. CTOs across the financial sector are being asked to demonstrate how technology decisions translate into measurable operational results. But there is a paradox in the way many financial firms — and especially those that are complex or specialized — approach improving operational efficiency and economy. While conventional wisdom maintains that complexity demands building bespoke, in-house solutions, the reality is that complexity strengthens the case for partnering, allowing firms to improve efficiency without sacrificing customization.

What is partnering and why is it different?

Since building involves developing in-house from the ground up and buying means directly licensing solutions and installing them independently, many firms have opted for either a build or buy solution. That can mean buying a foundation to build on, or buying point solutions and stitching them together with custom integrations.

Partnering offers a best-of-both-worlds solution: continuous development, managed services, and collaborative innovation. Partnering offers collaborative innovation and continuous future-proofing driven by the partner’s ongoing investment in R&D. This ensures the foundation evolves with changing regulations, data standards, and technologies without the firm absorbing perpetual rebuild costs.

Equally important, partnering delivers results in weeks rather than years. For firms under pressure to demonstrate operational impact, speed to value is as critical as flexibility.

Why a rethink of ‘build’ is necessary

Let’s go back to our paradox. Traditionally, financial institutions invested heavily in their own systems. Given the growing complexity of service offerings, it would seem logical to continue with a “build” approach. But today, infrastructure sprawl is reaching a breaking point. Years of patching legacy systems and layering homegrown tools have made change slow and control fragmented. Internal complexity is a liability in an already complex world.

Many firms are now reckoning with the unintended consequences of building too much themselves. Proprietary APIs, brittle data, and workflows that no one remembers how to maintain have become impediments to scalability. At Barclays, for instance, bespoke solutions also put added strain on firms to keep up with fast-moving technology developments, especially the rapid advances in artificial intelligence, which are already reshaping core operations.

Financial firms should develop their own ‘secret sauce’ to meet client needs, but there is no reason to build and maintain the raw ingredients. Foundational layers such as data ingestion, normalization, reconciliation, and lineage don’t create differentiation for most firms. They create fragility when managed in-house. These are best delivered through a partner, while firms focus on building value in the analytics and workflow layer.

Partnering with the right provider, able to adapt to a firm’s specialized needs, is the more future-proof and economical choice.

Rather than devoting resources and time to functions outside their core areas of expertise, financial firms that partner with specialists providing a platform for bespoke data ingestion, normalization, reconciliation and audit trails, for example, can reallocate their efforts to the truly complex solutions that provide demonstrable value for their clients.

The limits of ‘buy’

Buying standalone solutions presents another set of shortcomings. They may offer a faster time to market, but rarely align with a firm’s long-term needs.

Off-the-shelf products leave firms with little control over the product roadmap and grow in step with the firm’s changing needs. Firms tend to underestimate the integration cost of standardized solutions and overestimate fit. Worse still, the gaps they uncover can take months to remedy while the firm searches for new fixes.

While a “buy” decision can be appear to offer tempting time and cost savings, it’s likely to leave a firm with complex needs locked into solutions that are difficult to adapt to changing needs and technology.

Data: where partners excel

The database layer is a natural starting point for efficiency gains, where optimized indexing and querying improve performance and scalability. It is also the spot where partnering can deliver significant value most rapidly by providing scalable ingestion, normalization, reconciliation, lineage, and data ROI relative to partner economics. Just as important, embedded audit trails and transparency features strengthen regulatory resilience, helping firms meet inspection, reporting, and compliance obligations without custom builds.

BizTech Magazine notes that 66% of investment banks struggle with data quality and integrity, and 83% do not have access to real-time data or data analytics.ⁱⁱ Firms that treat data as an enterprise asset often gain an edge in speed and confidence across operational teams.

In a world gone digital, trust is more important than ever. Progressive organizations are prioritizing the development of integrated, privacy-centric data ecosystems that foster trust as a core value—not merely operational efficiency. These systems are engineered to support clear and reliable consent management, ensure uniform data practices across departments, and maintain traceable data flows that meet the rigorous demands of both regulatory bodies and discerning customers.

The emergence of AI and AI-driven analytics enables data to be used to identify compliance violations like fraud and money laundering. Yet this works only if firms have organized and managed data in the appropriate ways. This functionality requires a modern data platform that can accommodate multiple input sources, effective data governance to ensure that information is accessed and used properly, an organizational awareness around the complexity of data, and an informed approach to deploying it.

Working with a trusted partner allows firms to benefit from this functionality without having to build it in-house or buy an ill-fitting off-the-shelf solution.

Unused software licensure cost, for example, is a growing source of waste. Last year, cloud provider Zylo published a report showing that the average company spends $18 million per year on software that it never touches.ⁱⁱⁱ Infosys research, meanwhile, reports more than $300 billion in untapped corporate cloud commitments ^iv — a staggering pool of contracted spend that illustrates how much capital enterprises tie up in cloud services.

To put all this in perspective, estimates suggest that up to half of software licenses go unused.^v Here, too, partnering brings economy and efficiency because up-to-date specialized partners are strongly incentivized to keep their systems up to date and to develop the best solutions for their clients.

Choosing a partner

Selecting the right partner is central to a successful partnership experience. Validating if the partner option is truly best for your firm involves calculating the relative payback periods of building in-house versus partnering to attain the required scope. Payback periods will vary due to available baseline functionality and flexibility to support scale. Yet having a well-defined set of priorities and acceptance criteria is essential before initiating potential partnership discussions.

Beyond defined scope, the next most important selection criteria is the level of integration of a partner’s ecosystem, meaning its existing connectivity with third-party market-leading data and software providers, counterparties, and fund administrators. Other key partner selection criteria include:

Business ROI. During the implementation phase, the cost-to-value ratio should be stable, with improvements over the lifetime of the relationship.
Vendor reputation and validation. Firms should assess the reputation and reliability of potential vendors, inclusive of their solution’s flexibility and the viability of their business overall. Validate partner claims by requesting client testimonials and independently assessing their track record.
Support and SLAs. Assess the maturity of incident-response workflows, review coverage of all in-scope processes, confirm flexibility for ad-hoc support, and evaluate reliability on uptime and resolution.
Data security and compliance. Assess the data security and compliance aspects of solution and ensure alignment with industry regulations and internal security standards.

Seven factors in the build/buy/partner decision

Coming up with the right mix in making build/buy/partner choices can be fraught. Here are seven factors that may be helpful to consider when coming to a decision:

1. Strategic importance: The extent to which the solution aligns with the firm’s long-term goals.

2. Time sensitivity: The urgency with which a solution must be delivered. Is there time for a build solution even if that may be the preference?

3. Cost-benefit: Comparing all costs of development, implementation, and maintenance against all savings, offsets, costs avoided, and potential revenue enhancements — viewed through a discounted cashflow lens.

4. Expertise: The availability of human resources to plan, develop, and maintain whichever solution is chosen.

5. Scalability: Ability to accommodate future volume growth in a manner that permits customization and integration with dependency-mapped systems.

6. Risks: These include the possibility for delays, cost overruns, failure to complete a “build” project successfully or failure of a “buy” or “partner” solution to perform as expected.

7. Vendor evaluation: A thorough consideration of the reputation, security certification, and likelihood that the vendor will be able to support its products over their useful life.

Finally, those making data-related decisions should be honest about their own biases and assumptions. Those burned by vendors who promised too much and delivered too little may be determined never to repeat their mistake. Veterans of failed in-house efforts should recognize that many projects can work out well. And those who have never tried a partnering solution should view partnering as a proven approach to efficiency, adaptability, and resilience.