Navigating the Balance: Cloud vs. Workload Repatriation

By Dirk Alshuth, Cloud Evangelist at emma, cloud management platform.

The cloud revolution has transformed the IT industry, providing organisations worldwide with scalability, agility, and innovation. According to a Deloitte study, 90% of firms view cloud technology as essential for growth, digital transformation and competitiveness in the marketplace. Despite this, embracing the cloud can present various challenges for organisations. The high cost of cloud services is a major factor driving organisations to consider cloud repatriation. GEICO recently revealed that moving to a fully cloud-based infrastructure ended up costing the automotive insurance company $300 million in public cloud bills and they were still unable to innovate at the pace they needed to. This paired with security concerns, vendor lock-in and a rapidly evolving regulatory landscape has prompted organisations to reassess workload distribution, with research from IDC finding 83% of enterprise CIOs are considering repatriation for at least some of their workload on public cloud.

However, this does not signify a decline for cloud in 2025. Instead, it highlights a shift as organisations strive to maximise the value of their cloud infrastructure by embracing the advantages of a true multi-cloud environment. Rather than viewing repatriation as an exit from the cloud, it’s better understood as part of a balanced, strategic approach. By integrating on-premises workloads with private and public cloud environments, organisations can achieve long-term benefits and greater returns on their investments.

Mitigating risks and staying compliant

The cloud provides scalability and opportunities for innovation, but concerns about restrictive agreements, data security and regulatory scrutiny are prompting many organisations to reassess how they utilise cloud.

Industries such as healthcare and finance are subject to strict data privacy laws which require sensitive identifiable information to remain in house. With legislation, such as the Digital Operational Resilience Act (DORA) which came into effect on 17 January 2025, and a general increased scrutiny on these sectors for their data security, many organisations find themselves looking to repatriate their data and applications to ensure compliance.

As a result of this reassessment, many businesses are now moving towards a “cloud-if” strategy rather than a “cloud-first” approach, recognising that not all workloads need to reside on cloud infrastructure.

Selectively repatriating workloads and potentially migrating them back into the cloud when needed means that organisations can leverage the strengths of both on-premises and cloud environments. This allows for more efficient resource scaling, cost savings, and enhanced data sovereignty and security.

Overseeing cloud repatriation

Whilst cloud repatriation promises various advantages like cost optimisation and increased control, it can also present complex challenges. Fundamental hurdles faced by organisations during cloud repatriation include the lack of cloud-native services, limited access to provider-managed applications and potentially high capital investments needed for building or upgrading on-premises infrastructure.

Migrating workloads back on-premises often leads to the creation of hybrid environments, or when multiple public cloud providers are involved, multi-cloud environments. This change increases the complexity of infrastructure management, with many organisations needing to take on the management in-house, which requires more coordination and specialised expertise, such as system administrators, network engineers, and data administrators.

Public cloud providers typically offer an extensive selection of managed services and automated management tools that simplify operations and alleviates the pressure on IT teams. When repatriating workloads, organisations need to find alternatives or establish in-house solutions to replicate these functionalities. This can be costly, time-consuming, and might lead to capabilities inferior to those of cloud-native offerings.

Consequently, organisations must meticulously balance the trade-offs between the sophisticated functionalities of cloud-native solutions and the control afforded by on-premises environments. Successful navigation of the repatriation process requires careful planning, ample resources, and strong expertise in cloud management.

Making informed decisions with data

Workload repatriation presents another key challenge which is the high cost of data egress. Companies looking to transfer their data from the cloud back to on-premises systems may encounter unexpectedly high fees and charges. These fees, known as egress fees, are charged by public cloud providers for moving data out of their networks.

When substantial amounts of data need to be repatriated, egress fees can accumulate quickly as most public cloud providers calculate these charges based on the data volume being transferred, as well as factors like distance and time. A survey by Global Market Intelligence found that 34% of enterprises reported that egress fees have influenced their cloud storage strategies, leading them to either repatriate data to on-premises systems or switch to providers that do not charge for egress. These financial implications can cause a major challenge for companies contemplating the migration of their workloads away from the cloud.

To ensure repatriation is financially viable, organisations need to implement strategic measures to streamline and standardise data movement. By doing so, they can simplify data movement, reducing both costs and complexity for current and future repatriations. Leveraging automation, employing multi-cloud networking, and utilising direct inter-cloud connectivity enables organisations to gain more control, minimise egress costs, and enhance transfer efficiency.

Repatriation as a strategic advantage

As increasing costs, operational complexities, and regulatory pressure challenges the benefits organisations gain from cloud investments, workload repatriation is becoming a viable strategy to restore positive cloud value. However, workload repatriation should not be seen as a

standalone tactic. The true advantages of the cloud go beyond cost efficiency, including its capacity to encourage innovation, increase business resilience, and establish new growth opportunities.

Integrating on-premises environments as part of a “true” multi-cloud strategy means that organisations can strategically deploy workloads where they achieve optimal performance. This maximises flexibility, ensures compliance with regulations, and streamlines operational costs to align more effectively with business goals.

By Jonathan Dedman, Director at Cloudhouse.
By Isaac Douglas, CRO, servers.com.
By Martin Hosken, Field CTO, Cloud Providers, Broadcom.
By Jake Madders, Co-founder and Director at Hyve Managed Hosting.
By Terry Storrar, Managing Director at Leaseweb UK.
By Dave Errington, Cloud Specialist, CSI Ltd.