Michael Shmulevich, CTO of First Touch specializing in startup technology strategy and innovation.
Every startup founder hits the same dilemma. Do you wait for technology to mature, or build on what’s emerging now?
Wait too long, and competitors take the market. Jump too early, and you’re building on tools that can’t actually do the job. Both approaches are problematic: Arriving too soon means constant rebuilding that drains engineering time and creates technical debt. Late entry means competing against established players with deeper pockets, which is even more expensive.
Solving this dilemma is not gambling; it is a strategic timing based on observable patterns.
Reading The Signals
Technology evolution follows predictable patterns.
Moore’s Law stated that computing power doubles roughly every two years while costs drop. Cloud computing followed a similar pattern. When AWS launched in 2006, it offered basic storage and compute that was expensive and limited. Early adopters built on it anyway. Over the next 15 years, prices dropped a reported 107 times and capabilities exploded.
The pattern keeps repeating. A breakthrough technology emerges. Early versions cost too much and work imperfectly. Demand drives investment. Competition accelerates development. Costs drop while quality improves.
Recognizing this cycle lets you time your entry strategically.
The Six-Month Horizon
When we started building First Touch, generative AI was promising but problematic. GPT-4 supported only an 8,000-token context window. Hallucinations were common. At $30 to $60 per million tokens, the economics didn’t work for our use case.
We needed AI that could handle complex research and generate accurate outreach content at scale. The existing technology couldn’t support it profitably.
But we saw the trajectory: Huge capital was flowing into AI development. GenAI vendors were competing aggressively. Research papers showed rapid progress in model capabilities. The demand was clear, and the investment was substantial.
We made a calculated projection: Within six months or so, models would reach the quality and cost thresholds we needed. So, we built the architecture around that future capability, not the current one.
When GPT-4.1 launched in early 2025, it supported a 1 million token context window at $2 per million input tokens. The quality with proper prompting was exceptional. We swapped in the new model and went live.
Our timing worked because we bet on observable trends, not speculation.
Four Indicators That Technology Is Ready to Use
Not every emerging technology follows the same predictable path. Look for these four indicators when deciding whether to build on early-stage tech:
1. Capital Influx: Follow the money. When investors and corporations pour billions into a technology category, development accelerates. Large investments show confidence in near-term breakthroughs and naturally create the resources to achieve them.
2. Competitive Pressure: Multiple technology vendors competing in the same space drive rapid iteration. Each wants to be the market leader, forcing faster innovation cycles and price reduction pressure.
3. Fundamental Capability: The core technology must demonstrate the basic functionality you need, even if it’s expensive or inconsistent. If the foundation isn’t there, no amount of investment will create it quickly.
4. Clear Demand Signal: Market pull matters more than technology push. When customers are hungry for solutions that current technology can’t quite deliver, you know the demand will reward whoever can solve it first.
How To Build For Future Tech
Betting on technology evolution asks for specific architectural decisions. Design your system with the right abstraction layers to keep the emerging technology component modular and swappable. Build everything else around stable, proven tools. Have a plan B.
Test continuously against the current technology. Even if it’s too expensive or flawed for production use, validation ensures you’re ready when the improved version launches.
Stay close to the development community. Follow research releases, benchmark reports and competitive announcements. The best signals come from technical details, not marketing hype.
The Risk Balance
Building on immature technology does carry real risk. The anticipated improvements might arrive more slowly than expected. Costs might not drop as projected. Your architecture might need significant rework.
But waiting carries risk, too. Competitors will enter the market alongside you. Those who waited for “proven” technology enter later, facing established players with network effects and customer lock-in.
The question isn’t whether to take a risk. It’s which risk to take.
Making The Call
Strategic technology timing isn’t about predicting the future. It’s about spotting patterns, watching the right indicators and making informed bets on what you can actually see happening.
Ask yourself: Is money flowing into this technology? Are multiple strong players competing? Does the core capability exist, even if imperfect? Is market demand clear and growing?
If the answers point to rapid maturation, build your architecture now for the technology that’s six months away. Position yourself to capture the market when capability and economics align.
The first movers who time it right don’t just enter the market. They define it.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?