Published by Michael Schulz · Example article, please update before publishing
SD-WAN has long been the standard answer to connecting multiple company sites flexibly and cost-effectively. In consulting practice, however, it becomes clear that many projects fail not because of the technology itself, but due to unclear requirements defined upfront. Anyone planning an SD-WAN migration should therefore clarify a few points particularly carefully.
A second uplink alone does not make a network resilient. What matters is how the SD-WAN overlay responds to link failures: how quickly does failover occur, which applications are prioritized, and how does the solution behave during partial outages of individual providers? These questions should be tested with the vendor or integrator through concrete test scenarios – not for the first time during a live outage.
SD-WAN can be operated in-house, partially outsourced, or fully consumed as a managed service. For mid-sized companies, a hybrid model is often sensible: strategic control stays in-house, while day-to-day operations (monitoring, incident intake, configuration changes) are handed to a specialized partner. This split should be clearly defined contractually and organizationally from the start.
Because SD-WAN often enables direct internet breakouts at each site (local internet breakout), the security architecture shifts. Classic, centralized firewall concepts are no longer sufficient on their own. Integrated or cloud-based security functions (SASE approaches) should therefore be part of the planning from the outset, not added later.
Pure license costs are only part of the total cost. Migration costs, training effort for the internal team, and the cost of running legacy systems in parallel during the transition phase are frequently underestimated. A reliable calculation should cover the entire contract term, not just the first year.
SD-WAN projects succeed when resilience, operating model, security and cost are considered together from the start – not one after another. A structured requirements analysis before the tender typically saves more time than it costs.