Which 2012 predictions already look naive, and which are dangerously underestimated

At the end of every year dozens of pieces come out with predictions for the next one. I read most of them with the same feeling: the authors describe what has already happened with great precision, and then boldly extrapolate exactly the same thing a little further forward.

I want to do something different. Not a prediction, but a review: which 2012 forecasts already look overblown today, and which have quietly been left without the attention they deserve.

Overestimated: interfaces and the "experience revolution"

A large share of technology commentary this year was about how user interfaces are changing. Tablets. Touchscreens. New apps. "User experience as a competitive advantage."

This is real. But overestimated - because the interface is a surface. It changes how a person interacts with a system. It does not change what is underneath: data, processes, integrations, logic.

Companies that invest in a polished front end while the back end is broken do not gain a durable advantage. They get a well-dressed problem.

Underestimated: the accumulation of technical debt

While everyone discussed interfaces, most companies were quietly accumulating technical debt. Systems written "temporarily" ten years ago are still running and still serving as the source of truth. Integrations built through file exchanges in 2003 still run every night.

This is not an abstract problem. It is operational risk. When such a system fails, or becomes impossible to update for new requirements, the cost of fixing it turns out to be several times higher than timely maintenance would have been.

The 2012 forecasts paid almost no attention to this. It is a quiet threat that does not make covers.

Overestimated: "big data" as a ready-made answer

Big data in 2012 is roughly what "the cloud" was in 2010. A word that explains everything and explains nothing. Every other project is now called big data, regardless of whether there is a single characteristic of genuine large-scale data work involved.

The problem is not the technologies - they are real and useful when applied correctly. The problem is that "big data" has become a purchasing argument rather than a description of a need. Companies take on Hadoop-class technology for volumes a normal relational server would handle comfortably.

This is expensive, hard to maintain, and most of the time unnecessary.

Underestimated: data quality as an operational task

While the market talked about storing large volumes, companies continued to work with data where the same counterparty is recorded three different ways and "revenue" is calculated differently depending on who prepared the report.

Data quality is not a technology task. It is an organizational one. And the 2012 forecasts almost completely ignored it - because it is not presentable, does not require new hardware, and does not fit on a slide with a beautiful growth chart.

This is most often the real constraint for any analytics or AI project. Not the absence of tools - the absence of data discipline. This prerequisite holds just as much where there is already a lot of data and little value.

Overestimated: mobility as strategy

"Mobile strategy" in 2012 is nearly a mandatory item in any IT plan. But in most cases it means "build a mobile app," not "rethink how the processes work."

An app without rethinking the processes is the automation of inconvenience. Instead of filling out a form at a computer, the employee fills out the same form on a phone. That is not a strategy.

One question that rarely gets asked

Forecasts usually answer the question "what will appear." Rarely the question "what does it take for this to actually work inside a real company."

Before believing any forecast, I suggest asking the second question. Not "does this technology exist." But "what organizational, process, and data preconditions are needed for it to create value." If there is no answer - what you are reading is sales material, not strategy.