A client asked me this question last month, and I gave the same answer I’ve given for the past four years: Azure, unless you have a specific reason to choose AWS.
That answer surprises some people. AWS has the deeper service catalogue. AWS has the bigger market share. AWS is the default choice in the global startup ecosystem. So why do we still default to Azure for Nigerian enterprises?
The honest answer has nothing to do with the technology and everything to do with the local context.
Walk into almost any mid-sized or large Nigerian organisation and look at what they’re running. Active Directory. Exchange or Microsoft 365. Windows desktops. Often a SQL Server somewhere. Power BI in the finance team. Teams for collaboration since 2020.
That installed base means three things. First, your identity layer is already in Entra ID, which is the central nervous system of Azure. Connecting workloads to existing identity is trivial. On AWS, you’re either federating with Entra or running a parallel identity store.
Second, your people already have Microsoft skills. The same admins who manage Exchange can pick up basic Azure resource management with a week of training. Azure CLI feels like PowerShell. The portal layouts and conventions are familiar. On AWS, you’re hiring or retraining from scratch for most enterprises.
Third, your software licensing is already in a Microsoft envelope. Azure Hybrid Benefit lets you use existing Windows Server and SQL Server licences in the cloud, often cutting compute costs by 40% or more. The equivalent on AWS exists but the discount is smaller and the licensing terms are more convoluted.
Microsoft has account executives, partner engineers, and customer success teams physically present in Lagos, Johannesburg, and Nairobi. When something goes wrong, you can get someone on a video call who understands the regional context, the regulatory environment, and your account.
AWS has presence too, but it’s thinner outside South Africa. For most West African clients, AWS support means a ticket to an EMEA support pool. Workable, but different.
For a regulated enterprise that needs vendor accountability when something breaks at 3am, having a named Microsoft account team that takes calls matters. It’s not technical, but it’s real.
Microsoft has published explicit guidance on running Azure workloads under NDPR, CBN regulations for banking, and HIPAA-adjacent African healthcare frameworks. AWS has the same controls technically, but the documentation and audit support is less Africa-specific.
For a Nigerian bank going through a CBN cybersecurity audit, Microsoft’s audit packs and reference architectures map more cleanly to what the regulator asks for. That cuts months off the audit prep cycle.
I’m not arguing AWS is worse. There are scenarios where AWS is the obvious choice:
Pure-play SaaS startup. If you’re building a B2B SaaS product targeting global customers and have no existing Microsoft footprint, AWS is the conventional choice. The startup ecosystem, the talent pool, the patterns are all there. Going Azure feels like swimming against current.
Heavy data and ML workloads. AWS has more mature data services in some areas. Redshift is more popular than Synapse for large analytics workloads. SageMaker has a richer ecosystem than Azure ML for some use cases.
Existing AWS investment. If you’ve already invested years of engineering effort in AWS, switching has a cost. Don’t switch unless you have a specific problem.
Cost-sensitive at scale. At the high end (millions per month), AWS often comes in slightly cheaper for equivalent compute. Reserved instance pricing tends to be slightly more aggressive. This matters when you’re spending serious money.
For most Nigerian enterprises, GCP is a non-starter. The local presence is thin, the enterprise sales motion is weaker, and the talent pool is smaller. GCP wins specifically for data and ML workloads with strong BigQuery and Vertex AI usage, but you have to want it badly enough to overcome the operational tax.
The exception is digital-native businesses with strong engineering teams who’ve already standardised on GCP for product reasons. If you’re already there, don’t switch.
A common pattern: someone in leadership reads a McKinsey report about multi-cloud and decides the company needs to be on all three clouds for resilience.
Don’t.
Multi-cloud sounds resilient and is operationally disastrous. You double your security surface area, triple your training requirements, and quadruple your incident response complexity. The actual benefit, vendor lock-in mitigation, is real but rarely worth the cost for organisations under a thousand engineers.
For most Nigerian enterprises, the correct strategy is single primary cloud with disaster recovery in a different region of the same cloud. Active-active multi-cloud is a luxury for companies that can afford to staff it.
Pick Azure if you’re a regulated Nigerian enterprise with existing Microsoft licensing, an internal team familiar with Microsoft tools, and a compliance posture that benefits from named regional account support.
Pick AWS if you’re a SaaS startup or product company with no existing footprint, building for a global engineering talent pool, and willing to invest in independent AWS expertise.
Pick GCP if you have a specific data or ML reason to be there, with engineering leadership that wants it.
Don’t pick multi-cloud unless you genuinely need it and have the team to operate it.
The right cloud is the one your team can actually run safely. Not the one with the most features.
If you’re weighing this decision and want a second opinion specific to your business, book a consultation. If you’re already on a path and just want to validate the architecture, request a quote.