Nobody Can Predict the Future. Don't Waste Your Time Trying
Key Takeaways
- Confident tech predictions from the 1980s to today have consistently missed the mark.
- Rigid five-year roadmaps leave companies fragile when the guessed future doesn't arrive on schedule.
- Kentico built a dedicated AI enablement role in R&D instead of betting on a specific AI timeline.
- AI now speeds up parts of Kentico's platform development and legal contract execution.
- Adaptability compounds over time. Being right is just luck spread across enough guesses.
Every few years, someone in our industry makes a confident prediction about where technology is headed, and a few years later that prediction is quietly retired. Expert systems were going to run every business in the 1980s. Enterprise software was supposed to move entirely on premise, then entirely to the cloud, then entirely to no-code, then entirely to AI, each time framed as the last shift anyone would need to plan for. None of those calls were stupid. They were just wrong in the specific way that confident predictions about technology are usually wrong.
It happens every time something moves this fast. Nobody standing in front of the first electrical grid could tell you what it would do to daily life twenty years later. Nobody sitting in front of the first PC could describe the networked world that would follow. Nobody online in 1995 could have mapped out what the internet would become. When you are inside a shift like that, you do not get to see the other side of it in advance. You only find out once you are there, and by then it is not a prediction anymore, it is just what happened.
I get asked some version of the same question constantly right now. Where is AI heading. What should we be planning for in 2027. What does the org chart need to look like in three years. I understand why people ask it. It would be reassuring to have an answer.
I don't have one. And I have stopped pretending that not having one is a gap in my strategy.
Prediction was Never the Skill
Here is the reframe I keep coming back to. Companies that struggle in moments like this are not the ones that guessed wrong about the future. They are the ones that built an organization that only works if the guess was right. A five-year roadmap, a fixed structure, a team designed around last year's assumptions, all of that is a bet on a specific version of what comes next. When that version does not show up on schedule, and it usually does not, the organization does not have a plan B. It has a plan A that is now late.
The alternative is not a better crystal ball. It is building something that does not depend on one.
What that Looks like Inside Kentico
This is not an abstract point for us. A few weeks back, we restructured part of R&D, and the honest reason had very little to do with any specific technology bet. It had to do with the fact that our previous structure assumed a pace of change we could no longer assume. We created a dedicated AI enablement role inside R&D, not because we knew exactly what AI would do to software development, but because we knew we needed a standing capability to keep re-answering that question every week instead of once per quarter.
The results showed up faster than I expected. Development on parts of our platform is running significantly faster with AI embedded into the workflow, not because we predicted a number, but because we built a structure that could absorb the tools as they got good enough to matter. The same logic is playing out in legal, where contract execution is meaningfully faster because we treated AI adoption as an ongoing capability to build, not a project to finish.
None of that came from a roadmap that called AI in 2023. It came from deciding, in general, that the org needed to be able to move faster than our own forecasts.
The Uncomfortable Part for Leaders
The uncomfortable implication is that if your organization's ability to perform depends on you correctly predicting a fast moving market, you have already made yourself fragile, regardless of how good your last few predictions were. Nobody bats a thousand on this. Not analysts, not vendors, not CEOs who write confident LinkedIn posts about 2030.
The question worth asking your leadership team is not "what do we think happens next." It's "how quickly can we react to whatever happens next." For most companies I talk to, that answer is uncomfortably slow, and it has nothing to do with the quality of their forecasting.
I would rather run a company that is quick to adjust than one that is occasionally right. Being right is luck spread over enough predictions. Being able to adapt is a decision you make about how you build the organization, and it's the only one of the two that compounds.
Our mantra at Kentico is not to be right about the future. It's to be quick to adjust to whatever the future brings.
Frequently Asked Questions
Get ready for your upgrade
Upgrading from Kentico Xperience 13 to Xperience by Kentico gives your marketing and development teams a unified and future-proof platform with native AI that reduces complexity, and helps you create more impactful customer experiences.
