Why teams still pick monoliths
A debate post asked why companies stick with monoliths if microservices scale better, sparking real-world trade-off examples: developer velocity vs. operational complexity and the cost of distributed consistency. The thread synthesizes why monoliths persist in early-stage and high-velocity contexts despite microservices' theoretical scaling benefits argued.
A Feb. 17, 2026 analysis argued modular monoliths can deliver roughly "90% of microservices’ organizational benefits with 10% of the operational cost," positioning them as a pragmatic compromise. (javacodegeeks.com) Shopify still runs a massive Rails-centered "modular monolith" worked on by "more than a thousand developers" and uses Packwerk to enforce boundaries inside a single codebase, according to Shopify’s engineering write‑ups from Feb. 21, 2019 and subsequent tooling posts. (shopify.engineering) Shopify’s scaling tactics are concrete: a recent case study described a pod/sharding strategy that helped the platform absorb an estimated 284 million edge requests per minute during Black Friday 2024 while keeping fault isolation tight. (sujeet.pro) Industry surveys and practitioner posts highlight that most orgs haven’t realized microservices’ promised autonomy—Signadot’s analysis titled "Why 90% of Microservices Still Ship Like Monoliths" and The New Stack reporting both document teams with microservice topologies but monolithic release processes. (signadot.com) Quantitative studies and vendor reports show measurable cost pressure from distributed stacks: Cast AI’s 2025 Kubernetes Cost Benchmark analyzed usage across 2,100+ organizations for 2024, while several engineering guides estimate microservices can raise total ownership by roughly 30–100% versus an equivalent monolith. (cast.ai) Consultancies and engineering blogs are now chronicling reversals and hybrids—Cogent’s recent piece and a March 10, 2026 roundup both document teams "rolling back" or choosing modular-monolith patterns to cut platform toil and cloud spend. (cogentinfo.com) Academic and tooling work offers concrete mitigations: an arXiv/Kubernetes cost‑optimization paper (SpotKube) and CNCF FinOps guidance report techniques that reduced cloud spending by identifiable percentages in case studies, and companies like Shopify point to Packwerk and the strangler pattern as incremental migration options. (arxiv.org)
