The product is done. The engineers have signed off. QA has run its final passes. The code is sitting in a deployment queue, ready to go. And then, nothing happens for six weeks. This scenario plays out constantly across the tech industry, and the explanation almost never involves a bug, a bottleneck, or a panicked all-hands meeting. It involves a spreadsheet.

Tech companies already have the features you’re waiting for, and they’re simply not ready to give them to you yet. The deliberate delay of finished products is one of the least discussed and most consequential strategies in the business of technology.

The Inventory Problem Nobody Talks About

Software doesn’t sit in a warehouse. It doesn’t spoil. There are no carrying costs in the traditional manufacturing sense. So the logic of “holding back” a finished product feels counterintuitive until you understand what tech companies are actually selling.

What they’re selling isn’t the product. It’s the moment of the product. Timing is inventory in tech, and companies manage it with the same cold precision that a retailer manages shelf space during the holiday quarter.

Consider how hardware cycles work. When Apple or Samsung ships a new device, the launch isn’t just a consumer event. It’s a coordinated signal to investors, a lever on stock price, a headline that resets the competitive conversation. Releasing a camera upgrade in February, when the news cycle is quiet and no competing announcements are expected, is a fundamentally different asset than releasing that same upgrade at a major September event in front of a live audience and a global press corps. The product is identical. The value of the moment is not.

Manufactured Scarcity and the Launch Cycle

There is a more deliberate version of this strategy that borders on manipulation. Product managers at large tech companies routinely sequence feature releases not to match development timelines but to match subscription and upgrade cycles.

If your annual software subscription renews in October, there is a higher-than-coincidence probability that the most exciting new features will arrive in September. This isn’t conspiracy. It’s basic retention economics. A feature that lands three weeks before your renewal decision carries enormous commercial weight. That same feature, arriving in February, is just a changelog entry.

This logic extends to hardware. Planned release cadences at companies like Google and Samsung are often set 12 to 18 months in advance, long before a single line of production code is finalized. Engineers sometimes complete features that simply get filed for the next cycle, not because the features aren’t ready, but because releasing them now would cannibalize the upgrade narrative for a product that hasn’t shipped yet. This connects directly to a broader pattern in tech: companies regularly build features they never release at all, and the engineering logic behind that choice is more rigorous than it sounds.

The Competitor Clock

Holding a finished product also serves a competitive function that has nothing to do with your own upgrade cycle. It has to do with your competitor’s.

If a rival company is six weeks from a major launch, shipping your product into that window is a form of sacrifice. The press attention, the social conversation, the retail prominence, all of it gets absorbed by the larger event. Waiting until the competitor’s news cycle fades costs you six weeks of revenue but buys you the full oxygen of an uncontested launch. In categories with two or three dominant players, this calculation gets run constantly.

There is also the suppression play. When a competitor is rumored to be close to shipping something significant, a well-timed announcement (not necessarily a launch, just an announcement) can freeze the market. Potential buyers wait. Reviewers hold their coverage. The competitor’s momentum stalls. Microsoft has played this game against Google. Google has played it against OpenAI. The announcement becomes a strategic weapon even when the product behind it is months from completion.

Why “Ready” Is a Negotiated Term

There is a separate and more internally complex reason products get delayed after they’re technically finished. In most large tech organizations, “ready” means different things to different teams, and reconciling those definitions takes time that has nothing to do with engineering.

Legal needs to review for regulatory exposure. Marketing needs to align the launch with campaign budgets that were locked months ago. Sales needs training materials and pricing structures finalized before the product exists in a customer conversation. Support needs documentation, escalation paths, and staffing forecasts before the first complaint ticket arrives.

None of these are failures of process. They are the natural friction of a large organization attempting to coordinate a single moment across dozens of functions. But they produce a specific and recurring outcome: a finished product sitting idle while every non-engineering department catches up to the engineers. This is, in part, why the most productive engineering teams sometimes deliberately use outdated or constrained communication tools. Moving fast in isolation and moving fast as an organization are genuinely different problems.

The Economics of Anticipation

Finally, and perhaps most counterintuitively, waiting is sometimes the highest-value move simply because anticipation is itself a product.

The coverage generated by a leaked roadmap, a well-timed executive interview, or a controlled beta access program often exceeds the coverage generated by the actual launch. By the time a product ships, the most enthusiastic early adopters have already made their decision. The launch is a confirmation, not a conversion event.

This is why companies like Tesla have historically cultivated long pre-order queues for vehicles that won’t ship for years. The queue is the product. The wait generates social proof, signals market validation to investors, and keeps competitors uncertain about timing and demand signals.

It also explains a behavior that looks irrational from the outside: large tech companies launching products into categories where they know they will not win. The strategy behind launching products companies know will fail is coldly rational once you understand what those launches are actually for. Sometimes the goal of a launch is intelligence gathering, not revenue.

The through line across all of these strategies is that a finished product is a raw material, not a finished asset. The moment of release, the context of release, and the sequence of release transform that raw material into something with a specific and calculable value. Engineers optimize for completion. Executives optimize for the moment. And the gap between those two orientations is exactly where product delays live.

The next time a product you’ve been watching misses its expected ship date, it probably isn’t broken. It’s just waiting for the right moment to be worth the most.