Why the banking CMS has become a governance decision, not a marketing one.

A customer researching a mortgage may begin on a bank’s website, continue the journey through a mobile application, speak with a relationship manager, and later interact with an AI-powered assistant. While these touchpoints appear seamless, ensuring that every channel delivers accurate and compliant information remains one of the biggest operational challenges facing financial institutions today.

The customer sees one institution. Behind the interface, the bank is often running four disconnected sources of truth.

Executive Thesis

When banking leaders discuss digital transformation, the conversation usually centres on mobile apps, AI assistants, cybersecurity and open banking. Each of those receives a budget line, an executive owner and a place on the board agenda.

Yet every customer interaction depends on one asset that rarely appears there: content.

Product information, loan rates, compliance disclosures, branch details, onboarding journeys, FAQs, calculators, offers and financial education material all need to be consistent across websites, mobile apps, portals, contact centres and, increasingly, AI-powered interfaces. Everything else — the app, the portal, the assistant — is a delivery mechanism for it. Content is the only asset in the digital estate that every channel touches, and it is typically the only one without a single accountable owner.

The challenge is no longer creating this content. Banks produce it at volume, across every product line and every jurisdiction they operate in. The challenge is governing it and delivering it everywhere, accurately, at the same moment.

That distinction has changed the strategic weight of the banking CMS. A content platform was once judged on how easily marketing teams could publish and restyle pages. Those criteria now sit well below the ones that matter: whether the institution can prove what was published, where it appeared, who approved it, when it changed, and how quickly it can be corrected everywhere at once. A CMS selected against the first set of criteria and asked to satisfy the second will fail — usually quietly, and usually at the point of a regulatory review or an AI deployment.

Why This Matters to Banking Leaders

A retail bank today may operate:

• Hundreds of branches
• Multiple product lines
• Thousands of web pages
• Mobile banking applications
• Customer service portals
• Wealth management portals
• Partner and broker portals
• AI chat experiences

Every one of those channels requires the same underlying information. Yet in many banks, that information still lives across disconnected systems — some owned by marketing, some by product, some by operations, some inside a legacy platform nobody has decommissioned because nobody is certain what depends on it.

Four costs, four different owners

The costs of fragmentation are well understood individually. The reason the problem persists is that no single leader sees all of them.

Cost of fragmentation	Executive who carries it	How it shows up
Inconsistent messaging	CX and marketing	Customer receives one figure online, another from the contact centre
Regulatory exposure	Risk and compliance	An outdated disclosure on a low-traffic channel is still a published disclosure
Slow campaign execution	Revenue owner	Rate changes need coordinated manual updates across a dozen systems
Duplicated operational effort	COO	The same content authored, reviewed and approved multiple times

 

The operational reality is more granular than the summary suggests. When a product term changes, the update has to reach the website, the app, the PDF factsheet, the contact centre knowledge base, the branch collateral, the broker portal and any partner syndication feed. In most institutions, several of those are updated by a person copying text from an email. The failure mode is not dramatic. It is a single channel quietly retaining last quarter’s figure because one manual step was missed.

A multiplication problem, not a list

Take the number of product lines, multiply by channels, multiply again by jurisdictions or regulatory regimes. That figure is the number of places a single rate change must land correctly, on the same day, with the correct disclosure attached. For a mid-sized retail bank it runs into the thousands, and it grows every time the institution launches a product or enters a market.

At that point content stops being a marketing concern. It becomes a boardroom issue.

From Channels to Journeys

Historically, banks managed channels. Each had its own team, its own budget and, frequently, its own content store. That structure made sense when a customer used one channel per interaction, and the systems built to support it faithfully reflect the org chart of the decade in which they were commissioned.

Today, banks manage journeys. Customers do not think in terms of website, app, branch and call centre. They simply expect continuity — and they attribute any break in it to the institution rather than to a system boundary they cannot see.

The institution that delivers consistent experiences earns trust. The institution that delivers conflicting information creates friction, and in financial services friction compounds. A retailer showing a stale price loses a sale. A bank showing a stale rate creates a compliance event, a complaint, a contact centre call and a remediation exercise — then spends the following quarter rebuilding confidence it had taken years to accumulate.

Most banks have already restructured their teams around journeys. Fewer have restructured the content layer those journeys run on, which is why journey mapping exercises so often terminate in a slide deck rather than a change customers can perceive. The content platform is no longer a supporting system. It is the layer on which journey continuity actually depends.

The Compliance Challenge

This is where banking diverges sharply from retail or hospitality.

Every content update in a bank may involve product disclosures, jurisdiction-specific regulatory language, mandated wording that cannot be paraphrased or shortened for a mobile layout, and multi-stage approval by legal and compliance. A mortgage rate update cannot appear on one channel and remain outdated on another.

The requirement extends past publication. Institutions are expected to reconstruct what a customer would have seen at a given point in time — which version of a disclosure was live, on which channel, under whose approval. Where content is fragmented, that reconstruction becomes a manual investigation across systems, conducted under time pressure, usually by the people least equipped to do it quickly.

Most institutions treat speed and control as a trade-off. They accept slow publishing as the price of compliance, or they accept compliance risk as the price of responsiveness. Both positions are rational responses to a broken content layer, and both are avoidable. Modern CMS platforms remove the trade-off by making approval workflow, versioning and audit history properties of the platform rather than of a manual process wrapped around it. Review becomes both faster and more defensible: the approval is recorded because it is a system event, not because someone remembered to file the email.

CASE IN POINT — REGULATED INSURER, UK

In one implementation for a UK private medical insurer serving close to a million customers, the constraint was never creative capacity — it was control. Product information, claims guidance and policy documentation had spread across a growing estate of sites, all requiring approval before publication. The resolution was structural rather than editorial: a single knowledge base, a formal content approval workflow, and multi-site delivery from one governed source.

Banking’s version of that problem is larger by an order of magnitude. Every additional jurisdiction adds a disclosure variant, and every additional channel adds another place for that variant to go stale.

Centralised content governance is not a marketing efficiency argument. It is a risk control, and it belongs in the same conversation as every other control of the institution funds and audits.

Personalisation at Scale

Banks are investing heavily in personalisation — customer data platforms, analytics, segmentation models, next-best-action engines. Many of those same banks report that the results have underdelivered against the business case.

The assumed diagnosis is data quality. More often, the constraint is content structure.

A first-time account holder, a business owner, a high-net-worth investor and a student should not receive identical experiences. Delivering different experiences requires content that can be assembled differently for each — modular components with clear metadata describing what each one is, which product it belongs to, which audience it serves and which jurisdiction it is approved for.

When content exists only as complete web pages, there is nothing for the data layer to assemble. The segmentation model resolves correctly, identifies the right customer, and then has one generic page to serve them. The investment in data cannot return, because the content it was meant to orchestrate was never built to be orchestrated.

This is why personalisation programmes so often stall after a successful pilot. The pilot works because a small team hand-built a handful of variants for a single campaign. Scaling it would require hand-building thousands, so the programme quietly plateaus at the level of a homepage banner. The bottleneck was set years earlier, by a content modelling decision made when the only consuming channel was a website.

The real challenge, in other words, is not customer data. It is having content that can adapt dynamically across audiences and channels — which is precisely what headless CMS architectures and digital experience platforms were designed to enable.

Where Sitecore Fits

To address these challenges, many financial institutions are moving away from traditional website-centric content management systems toward modern digital experience platforms and headless architectures. Platforms such as Sitecore enable organisations to centralise content governance while delivering personalised experiences across websites, mobile applications, customer portals, and emerging AI-driven interfaces.

The architectural principle is straightforward: content is authored and governed once, held in a structured form independent of any particular presentation, and delivered through APIs to whatever consumes it. A rate change is made in one place and propagates to the website, the app, the portal and the branch system without any of them being edited.

The emergence of AI interfaces has made this shift urgent rather than aspirational. An AI assistant does not consume web pages; it consumes structured, retrievable content with clear provenance. An institution whose product information exists only as rendered HTML across dozens of templates cannot reliably ground an assistant in it — and an assistant that cannot be grounded in governed content is a compliance liability wearing a friendly interface.

CASE IN POINT — ENTERPRISE PLATFORM MODERNISATION

The common objection is that core systems make this impractical. It is an integration problem rather than a blocker. In one enterprise engagement outside financial services, a major platform upgrade completed in four months — alongside single sign-on across legacy and modernised systems and integration with an enterprise ERP — delivering a 40% performance improvement and a 30% reduction in defects.

The pattern transfers directly: the content platform sits above the systems of record. It does not replace them, and it does not require them to be replaced first. That distinction matters for sequencing — content governance can be addressed as a discrete programme, channel by channel, without waiting for a core modernisation roadmap that may run for years.

Who Owns Content Governance

Platform capability is a precondition, not a solution. The institutions that succeed treat governance as an operating model with a named owner, in the same way data governance acquired one over the past decade.

That model needs three things the technology alone will not supply:

• A defined content taxonomy — an agreed structure describing what a product term, a disclosure or a rate actually is as a content object, so it can be reused rather than retyped.
• Clear editorial rights — who may author, who must approve, and which changes require compliance sign-off versus which are routine.
• Measurement — time-to-publish across all channels, the count of systems holding customer-facing product information, and the volume of manual re-keying.

All three are measurable, and all three are currently unmeasured in most institutions. Without them, a modern platform simply becomes a faster route to the same fragmentation.

Five Questions for Banking Leaders

The state of an institution’s content infrastructure can be assessed quickly:

1. How many separate systems hold customer-facing product information today?
2. How long does a rate or disclosure change take to appear correctly on every channel?
3. Can compliance produce an audit record of what was published, where, and when?
4. Is content structured enough for an AI assistant to consume and cite accurately?
5. Who owns content governance — marketing, IT, compliance, or no one?

The fifth question is usually the most revealing.

The Infrastructure Test

Return to the mortgage customer. Their experience of the institution is assembled, in real time, from whatever content each channel happens to hold. If those sources disagree, no amount of interface design conceals it.

Content in banking meets every practical test of infrastructure. Every channel depends on it, its failure carries regulatory consequences, and that failure is invisible until it is public. Infrastructure gets funded, governed and measured accordingly.

Most banks are still treating it as marketing.

 

Author: 

Devendra Rupani

Founder of Sourceved, a Sahana System Group company and a Sitecore Solution Partner. With more than 20 years of experience in enterprise technology, digital transformation, and customer experience platforms, he has led over 100 enterprise projects and contributed to more than $30 million in digital transformation initiatives across industries including banking, healthcare, education, manufacturing, and telecommunications.