A burgeoning divide between in-house and cloud finance

The cost of delay is rising as AI solutions and competitive advantages move to the cloud.

Finance and technology leaders are discovering that the chasm between traditional and cloud capabilities is becoming harder to ignore.

The discussion typically unfolds along these lines:

“Yes, we are on Dynamics GP. We know we need to modernise, but we are not in a rush. We will evaluate and migrate when the time is right.”

“When the time is right” is a decision in and of itself. One with strategic consequences. Remaining on the fence places organisations beyond the capabilities defined by modern finance.

Microsoft’s “feature comparison tables” affirm this as a technical reality. Generative guidance, Copilot, summary generation, and AI agents are cloud-native features. On-premises deployments lack these and more.

Emerging cloud finance capabilities

The Finance Agent, announced in Microsoft’s 2026 Release Wave 1, is a role-based AI experience that embeds financial intelligence into Excel, Outlook, Teams, and Copilot Chat. Finance professionals can query financial data and trigger workflows directly from their productivity tools. Controllers running self-hosted deployments cannot access these capabilities.

Copilot-assisted reconciliation automates the detection of mismatched transactions, proposes resolutions and drafts journal entries for approval. Finance teams using cloud-deployed Dynamics 365 Finance are already using these capabilities in production. Teams operating in-house continue to perform these activities manually.

Agent-enforced compliance controls, including segregation-of-duties enforcement, three-way matching, duplicate invoice detection, and Capex-to-RFQ policy validation, rely on AI agents. Locally hosted deployments are dependent on manual controls and periodic audit samples.

Work IQ, which reached general availability on 16 June 2026, is an intelligence layer that bridges Microsoft 365 signals with Dynamics 365 operational data. It is entirely cloud-based and is unavailable to on-premises clients.

The cost of remaining on-premises

Most in-house Dynamics customers understand their maintenance costs. The indirect costs are often less visible.

Upgrade cycles are pricey and disruptive. In-house ERP upgrades tend to involve six to eighteen months of planning, testing, and change management. On the other hand, cloud customers upgrade without the usual on-premises hassle when Microsoft launches new features.

Custom integrations also become technical debt. Many in-house deployments have accumulated years of custom code and integrations built before standard connectors existed. Every upgrade touches this code, and each AI capability that depends on clean, standardised data architecture is affected by it.

Compliance costs are rising. POPIA enforcement in South Africa has become a practical compliance concern. Organisations processing personal data, including every finance system that manages customer records, employee information, and supplier data, must demonstrate data governance and maintain audit trails, among other requirements. Cloud-based Dynamics 365 Finance includes these by design and is supported by Microsoft’s regional data centres and local residency. On-premises teams must build and maintain these controls themselves.

The capability gap may be the most notable “hidden” cost. A competitor employing cloud-based Dynamics 365 Finance, with AI agents automating reconciliation, strengthening controls, and accelerating close cycles, moves faster with the same team.

Remaining on-premises creates a maintenance burden and an ever-expanding gap in productivity and core competencies.

Security considerations in the cloud era

One of the most common objections raised by business leaders contemplating migration is security.

“We control our own infrastructure. We know where our data is. The cloud introduces risk.”

This argument carried greater weight a decade ago than it does today.

Self-hosted environments require dedicated teams to manage patch cycles, and every unpatched vulnerability creates an additional point of exposure. They require physical and network security, which many organisations do not meet enterprise standards for. In the event of failure, the organisation is responsible for backup and disaster recovery. Compliance certifications such as SOC 1 Type 1 and ISO 27001, which Microsoft maintains for its cloud ecosystem, must also be achieved and maintained by on-premises clients.

Microsoft’s cloud infrastructure operates at a level of security that few individual organisations can attain. Thousands of security engineers, AI-driven threat detection, continuous patching, and compliance certifications spanning numerous global standards support it. Microsoft’s local data centre region also allows South African organisations to align data residency requirements with POPIA without customisation.

Cloud security depends on proper identity governance, role-based access controls, and policies governing how AI agents interact with data. Dynamics 365 Finance and the Agent 365 framework incorporate these tools, but they require configuration and governance. Successful outcomes are contingent on selecting an experienced implementation partner.

Planning the path to modernisation

“We know we need to move to the cloud, but migration feels daunting.”

For many organisations, the scale of migration is the greatest source of hesitation.

The degree of disruption associated with ERP migration depends heavily on planning, phased execution and data readiness. Delaying migration, however, widens the capability gap over time.

Most effective migrations follow a structured sequence of activities.

1. Data readiness assessment. Before any migration discussion, organisations need to understand their data. Customer master duplication, vendor record omissions, chart-of-accounts inconsistencies and unresolved historical transaction errors can impede migration. Left unresolved, these same issues can also compromise AI accuracy after migration.
2. Process standardisation review. AI agents perform best on standardised processes. If an on-premises deployment contains years of customisation built around legacy workflows, this is the appropriate stage at which to rationalise them. Microsoft’s reference architecture for Dynamics 365 Finance should become the target. Every custom deviation introduces migration and long-term maintenance costs.
3. Migration planning. A phased approach is appropriate for most African enterprises. Financial management and reporting are often addressed first, followed by accounts payable and receivable, treasury and cash management, and finally the compliance and AI enablement layers. Each phase delivers independent value.
4. Cloud foundation and governance. Identity governance, role-based access controls, data residency configuration and agent policy setup are foundational activities that should begin at the outset of migration.
5. Agent enablement. Once organisations are operating on cloud-based Dynamics 365 Finance with clean data and governed access, agent capabilities can be introduced incrementally. Reconciliation automation and duplicate invoice detection are often among the quickest capabilities to demonstrate measurable value. Additional requirements can then be introduced in line with business priorities.
6. Continuous improvement. Cloud is a model of continuous delivery. Microsoft introduces new capabilities every six months through its release waves. Staying current becomes an ongoing operating discipline focused on evaluating and deploying capabilities that support business objectives.

Reframing the decision

CIOs and CFOs evaluating a modernisation strategy should consider one fundamental question: is the organisation choosing between on-premises and cloud, or between now and later?

If the decision is about timing, it is important to understand the cost of delay. Suspending transformation delays your access to cloud-native capabilities that are reshaping finance operations.

The case for remaining on-premises should account for the capability gap, the security and compliance burden, and the total cost of self-managed infrastructure.

A comprehensive assessment of these factors changes the economics of the decision.

The Braintree difference

Braintree works with organisations that operate legacy Dynamics systems, such as AX, GP, and NAV, as well as other similar on-premises ERP solutions, through a structured cloud readiness assessment.

Our process begins with a plain-language evaluation of the current estate and the complexity of migration. It assesses the potential outcomes of an initial phase of cloud deployment. Clients receive a decision brief that clarifies the choices ahead of them and the implications of each path.

The rising cost of delay reflects an undeniable reality. To build an intelligent finance function, you must first understand your existing infrastructure, and then make deliberate choices about the capabilities that will define its future.

Take an evidence-based approach to modernisation. Speak to Braintree about a cloud readiness assessment.

Specialists in Business Applications, Modern Workplace and Azure. Let’s grow.

Source: Microsoft Learn, “Comparison of cloud and on-premises features – Finance & Operations | Dynamics 365”. Feature availability is accurate as of July 2026 and may change as future release waves roll out.