Source: DR Journal By Erik Salo, SVP, VDURA , July 2, 2026 – No matter how much diligence you bring to a storage decision, it’s hard to be fully confident even after the contract is signed. You sit through presentations, study the benchmarks, model your growth, stress-test the cost assumptions. But the real evaluation only begins once the platform is in production and carrying real workloads.
That isn’t a flaw; it’s the nature of the work. The catch is that storage decisions are expected to hold for years, and things change in ways nobody could have priced in on the day they signed.
Right now, the clearest example is flash. Plenty of organizations committed to all-flash architectures when NAND (flash memory) was cheap and abundant, and that assumption is coming apart underneath them. Enterprise solid-state drives (SSD) pricing has moved by multiples over the past year, the gap between flash and hard disk drive (HDD) capacity costs has widened to more than 20x at the extreme, and the cause is structural rather than a passing cycle: AI demand is pulling supply one way while manufacturers shift capacity toward higher-margin memory. When the single assumption that anchored your architecture stops holding, the decision that looked obvious at procurement starts to feel very different in production.
What follows tends to be a familiar sequence. Five stages, really, and they look a lot like grief. Recognizing them early is the difference between managing the situation and being managed by it.
Denial. You’ve made the investment and everyone wants the project to succeed, so early issues get read as teething problems. Performance isn’t quite there, complexity is creeping in, but the team is still learning the environment, and the workloads are still being tuned. Give it time. The risk is that real warning signs get rationalized away, because the deeper problems don’t fully surface in the early days. By the time they’re impossible to ignore, you’re further into the project and your data volumes are larger.
Anger. Eventually something forces a reassessment. A serious performance event, or an expansion quote that lands far above plan because the media you standardized on has repriced. The platform isn’t necessarily broken, but it’s no longer delivering the confidence that justified the original spend. Attention turns to accountability, and you start revisiting the assumptions behind the purchase, asking whether they ever reflected how the system would actually be used.
Bargaining. Now the focus shifts to compensating for the shortcomings without reopening the architectural decision. Sometimes that’s the right call and a targeted addition solves it. More often, the fixes treat symptoms rather than the cause. Each one buys an incremental improvement and leaves the underlying limitation in place. Complexity starts to grow faster than value, budget creep sets in, and the real question quietly changes from “can we improve this” to “is continuing to invest in it still the best option.”
Depression. This is where it gets uncomfortable. Staying and starting over both look unattractive. You feel over-committed to the work already done, and the temporary fixes are now permanent fixtures. The economics make change hard to justify, because migration costs are immediate and visible while the cost of staying is spread quietly across budgets and operations. It’s an expensive, stressful catch-22. But somewhere in it the question flips: instead of “how do we make this work,” you start asking “what would we do differently if we were deciding today.”
Acceptance. This isn’t admitting defeat. It’s the point where frustration turns into clarity. The original decision wasn’t necessarily wrong; it was made with the information available at the time. What changes is the questions you now know to ask. How will this behave over the full life of the investment, not just at deployment? How easily does it scale? How resilient are the economics if media prices move again, the way flash just did? How much complexity does it add as the environment grows?
The most valuable outcome of going through all five stages is recognizing the board. You come out the other side knowing which questions actually matter and which trade-offs you can live with in your own environment. That’s hard-won, and in storage terms it’s worth a great deal.