Enterprises keep buying software upgrades that IT thinks will make things worse

We have all been there. You are finally getting your workflow dialed in, the integrations are stable, and the team is actually shipping. Then, a notification pops up: a mandatory upgrade is coming. Not because you asked for it, and not because it solves a problem, but because a vendor decided it was time to move the needle on their own quarterly earnings.

A recent survey of over 2,000 IT leaders at major corporations confirms what most builders already know: enterprise software is becoming a tax on productivity rather than a tool for it. According to the data, roughly 28 percent of software budgets are now being burned on unwanted upgrades. The kicker? The people actually running the systems believe these forced changes often make things worse.

The Illusion of Progress

In the world of SaaS, stagnation is seen as death. If a company isn't shipping new features, the market thinks they are irrelevant. This creates a perverse incentive where vendors push out UI overhauls and "AI-powered" sidebars that nobody requested. For the builder on the ground, these aren't features. They are breaking changes disguised as progress.

The survey highlights a massive disconnect between the executives signing the checks and the IT heads who have to live with the consequences. Decision-makers are often sold on a vision of "modernization" that ignores the technical debt and training costs associated with every unnecessary version jump. When 28 percent of a budget is spent just to stay in the same place—or worse, to take a step back in usability—we aren't looking at innovation. We are looking at a protection racket.

The Vendor Leverage Trap

Why does this keep happening? It comes down to leverage. Most enterprise firms are so deeply entrenched in their legacy stacks that the cost of switching exceeds the cost of a bad upgrade. Vendors know this. They use the "end-of-life" threat to force migrations onto new tiers that usually cost more and deliver less.

For those of us in the crypto and AI space, there is a lesson here. We often talk about decentralization as a philosophical goal, but this data proves it is a practical necessity. When you rely on a centralized provider for your core infrastructure, you aren't a customer; you are a tenant. And the landlord just raised the rent while painting over the outlets.

Why IT Teams Are Skeptical

The survey notes that IT heads aren't just being grumpy or resistant to change. Their skepticism is rooted in historical failure. Every time a major platform does a forced migration, it breaks custom scripts, disrupts APIs, and requires hundreds of hours of retraining for non-technical staff. The "upgrades" frequently prioritize data collection for the vendor over functionality for the user.

In the AI space, we see this with the rush to integrate LLMs into every single toolbar. A builder needs a clean IDE and a fast compiler, not a chatbox that takes up 15 percent of the screen and hallucinated code suggestions. When these features are forced through, they don't add value—they add noise.

What Builders Can Do

If you are a founder or a lead dev, you have to start thinking about your stack as a liability. The more "locked-in" a feature is, the more likely you are to be part of that 28 percent budget waste in two years. Here is how to fight back:

• Prioritize Portability: Use open-source standards whenever possible. If your entire backend relies on a specific vendor's proprietary hook, you have no leverage when they force an upgrade.
• Audit the Hype: When a vendor pushes an upgrade based on "AI enhancements," ask for the raw performance metrics. If they can't show how it saves time, it is probably just a UI layer designed to justify a price hike.
• Demand Long-Term Support (LTS): In the enterprise world, stability is a feature. If a vendor won't commit to a five-year stability map, they are planning to force your hand sooner than you think.

The Opportunity for New Players

This massive waste of capital is actually a massive opportunity for new builders. There is a growing hunger for "boring" software—tools that do one thing well, respect the user's workflow, and don't change the UI every six months. The next generation of successful enterprise tools won't be the ones with the most features; they will be the ones that promise the most stability.

We are seeing a shift where "feature-rich" is becoming a red flag for IT departments. They want tools that stay out of the way. If you are building in AI, the goal shouldn't be to add more steps to the user's day, but to remove them without requiring a total system overhaul.

The most expensive software you buy is the stuff that forces your team to stop working so they can learn how to work again.

The 28 percent waste mentioned in the survey isn't just a financial metric; it is a measure of lost momentum. When a firm spends nearly a third of its budget on things that make the job harder, they are essentially subsidizing their own decline. As builders, we need to stop chasing the upgrade cycle and start demanding tools that actually serve the mission.

Final Takeaway

Enterprise software is currently a seller's market, but the frustration shown in this survey suggests a breaking point is near. If you are a founder, look at your own budget. If you are spending more time managing your tools than using them, it is time to pivot. The future belongs to the lean, the portable, and the stable. Everything else is just expensive noise.

Read the original at Ars Technica →