Is Microsoft still a sensible choice for a new server-based project?

There are also practical business reasons to choose Microsoft: predictable vendor support, enterprise tooling, and a clear party to hold accountable when something goes wrong. If clients require that ecosystem or your staff already knows it well, the licensing premium may be justified.

Cloud hosting doesn't automatically make Microsoft cheaper. Managed SQL Server and Windows services can become very expensive at scale, while Linux, PostgreSQL, and MySQL options are often less costly. Cloud is still useful if you need someone else to handle hardware, backups, patching, redundancy, networking, and operations.

Microsoft can still make sense when you need compatibility with existing Windows applications, .NET Framework software, an ERP system, or an organization that already has strong Microsoft expertise. Replacing an established SQL Server and Windows stack can be a long, expensive migration, even if PostgreSQL is attractive for new development.

For a genuinely new project, Linux and PostgreSQL are usually worth considering first. They avoid major licensing costs and are well-supported, scalable technologies. That said, the best choice depends on your team's experience, existing infrastructure, and application requirements rather than ideology alone.

The most useful question is what problem you are solving. If this is a greenfield service with no Microsoft-specific dependencies, compare total operating cost and team productivity against a Linux/PostgreSQL design. If the project must integrate deeply with existing Microsoft systems, staying with Windows and SQL Server may reduce risk even when it costs more.