2023 Corporate In Memoriam: The Year Software Sobered Up

Every year buries a few assumptions. 2023 buried the idea that growth would pay for any amount of infrastructure and that you could always sort out the bill later. It was the year the industry sobered up. Budgets tightened, the free-money era ended, and teams across every sector had to look at their platforms and ask a question they had been avoiding: what does this actually cost, and is it worth it?

At RIP Corp we spend our days inside other companies' platforms, so a year in review for us is a list of the themes that quietly reshaped how good software gets built. Here is what 2023 laid to rest and what it left behind.

The year cloud cost became a first-class concern

For a decade, cloud spend was treated as a rounding error against growth. In 2023 it became a boardroom topic. Finance started asking engineering to explain the bill, and engineering discovered how little visibility it actually had into where the money went.

The healthy response was not a panic migration off the cloud. It was cost discipline as an engineering practice: attributing spend to teams and features, killing idle capacity, right-sizing instances, and treating efficiency as a design goal rather than an afterthought. The best teams learned that a platform you cannot afford is not a platform, it is a countdown.

The most valuable dashboard of 2023 was not about users or revenue. It was the one that finally showed where the cloud money went.

Data platforms consolidated

The years before had scattered data across a dozen tools, each a good idea in isolation and a governance nightmare in aggregate. In 2023 the pendulum swung toward consolidation. Companies pulled their analytics into fewer, better-governed platforms, invested in a shared warehouse or lakehouse, and finally started treating data as a product with owners, contracts, and quality standards.

The motivation was partly cost and partly the dawning realization that you cannot do anything serious with AI on top of a data mess. A clean, well-modeled data platform stopped being a nice-to-have and became the prerequisite for everything that came next.

Generative AI arrived, loudly

The defining shock of the year was generative AI going mainstream. Suddenly every product roadmap had an AI line item, every executive had a mandate, and every engineering team had to separate what was genuinely useful from what was demo theater.

The mature teams treated it as exactly what it was: a powerful new capability that still had to live inside real software with real constraints. That meant thinking hard about accuracy, cost per call, latency, data privacy, and how to keep a probabilistic model from making confident nonsense a customer would act on. The magic was real. The engineering discipline to ship it responsibly was the harder and more valuable part.

The themes worth remembering

Stripped of the noise, 2023 came down to a short list that touched nearly every industry we work in.

• Efficiency over growth-at-any-cost: platforms judged on unit economics, not just scale.
• Consolidation over sprawl: fewer tools, clearer ownership, better governance.
• Data as a foundation: the unglamorous work that makes AI and analytics possible.
• GenAI as capability, not magic: real value where it fit, healthy skepticism where it did not.
• Reliability regained respect: after years of feature sprints, uptime and cost became features too.

What survived is what mattered

The obituary for 2023 is really an obituary for the assumption that infrastructure would always be cheap and growth would always cover it. What survived is a more grown-up way of building: platforms that justify their cost, data foundations that can carry the next wave, and a healthy respect for the difference between a demo and a durable system.

That is the through-line of everything we build. The flashy trend of any given year fades. The software that outlives the cycle is the one whose owners understood what it cost, what it stood on, and why it would still be worth running the year after.