Every webinar and datasheet makes it sound like you just flip a switch. Replace your MPLS with Magic WAN, save a fortune, done.
But have they ever tried to unwind a global, multi-year MPLS contract? The early termination fees alone can obliterate their projected savings for the first 24 months. And good luck getting your local telco in some regions to even cooperate with the migration. They're incentivized to drag their feet.
It feels like they're selling to a greenfield startup with no legacy infrastructure, not an enterprise that's been around for 20 years. The real cost isn't in their monthly bill—it's in the transition.
Your vendor is not your friend.
You're absolutely right about the transition costs being the real barrier. I see this all the time when we try to roll out new collaboration tools. The software itself might be free for a trial, but the cost of migrating our training materials and getting everyone off the old, familiar system is massive. People don't factor in the sheer time it takes for teams to unlearn old workflows.
Your point about regional telco cooperation is something I hadn't even considered, but it makes perfect sense. Their sales teams are probably measured on retention, not on helping you leave. It adds a whole other layer of project risk.
So is the play to just wait until your MPLS contracts are naturally up for renewal, and only then evaluate a switch? That seems like a multi-year planning headache in itself.
That renewal window is the only realistic trigger. You're right, it's a headache. The trick is to start the evaluation 12-18 months before the contract ends, because you'll need that time.
We did it. Had to run Magic WAN tunnels over the existing MPLS for a parallel proof-of-concept for six months before we felt confident to cut over at renewal. Even then, we kept the MPLS live as a backup circuit for another quarter. The sales pitch never mentions needing to pay for both during transition.
YAML all the things.