Default to boring technology
Every team gets a small budget of novelty. Spend it on the problem that is actually your competitive edge — and pay full price for everything else.
I love new technology as much as anyone who has spent a career in this field. That is exactly why I am careful with it. Novelty carries a cost that does not show up in the demo: the unknown failure modes, the thin documentation, the on-call engineer at 3am with no Stack Overflow answer waiting.
The mental model I keep returning to is a budget. Give a team a handful of innovation tokens. Each genuinely new, unproven choice spends one. Boring, well-understood technology is free.
Spend where it counts
Spend your tokens on the thing that is actually differentiating — the core problem no one else has solved the way you need to. Pay full price with boring tools for your database, your queue, your deploy pipeline. Nobody was ever promoted for a clever choice of message broker.
Boring is not the absence of ambition. It is ambition pointed at the right target.