After a decade of digital transformation projects, I’ve learned a key lesson: the most valuable opportunities don’t come from building more apps or dashboards. They emerge from solving what I call Deep Domain Problems (DDPs)—complex, high-impact challenges that demand domain expertise and technical innovation. Here’s how to spot them and decide which ones are worth your effort.
What Makes a Problem "Deep"?
DDPs aren’t just complicated—they’re significant, systemic, and stubbornly resistant to quick fixes. They share these traits:
1. They Demand Specialized Domain Knowledge
You can’t solve a DDP with tech alone. You need to grasp:
- Industry-specific processes honed over decades
- Regulatory frameworks shaping the field
- Undocumented "tribal knowledge"
- Nuanced stakeholder dynamics
Example: Based on my experience, which can be explored further at WiredUp, in corporate forex risk management, professionals navigate thousands of variables—currency exchange rates, market volatility, and regulatory frameworks—customized to the unique needs of global businesses and their financial objectives.
2. They Rely on Tacit Expertise
When experts say, "It depends" or "You learn it with time," they signal tacit knowledge—skills not found in manuals. Look for:
- Decisions hard to explain step-by-step
- Roles taking years to master
- Gaps between documentation and practice
- Context driving different outcomes
Example: To tackle oil and gas industry challenges, my technical expertise combined with Evolve’s domain knowledge relied on tactical, hands-on experience, deeply grasping how the industry functions in specific regions and adeptly abstracting that insight for global solutions.
3. They Span Systems and Stakeholders
DDPs cross boundaries:
- Legacy systems that don’t play nice
- Stakeholders with clashing priorities
- Organizational silos
- Trade-offs like efficiency vs. compliance
Example: Creating a loan against Mutual Funds product involves weaving together multiple interconnected elements: uniting lenders and regulated entities to enable lien marking, collaborating with product designers to craft a seamless UX for borrowers, and leveraging developers to build modern tech atop legacy systems— based on my experience at Finezzy.
4. They Defy Simple Automation
Past attempts at solving DDPs often yield:
- Partial fixes for routine cases
- Workarounds that spawn new issues
- High exception rates needing human intervention
- Tools users ignore despite the functionality
How to Spot High-Value DDPs
A concise checklist to identify Deep Domain Problems worth solving:
☑ Are there persistent or emerging manual processes resisting automation or strained by new tech?
☑ Do experts use phrases like "it depends," "there’s an art to it," or "you develop a knack," hinting at gaps between tools and reality?
☑ Are there "translation" roles bridging systems, teams, or procedural divides?
☑ Do standard workflows frequently break down with exceptions with high costs or impact?
☑ Are expert judgments driving significant outcomes that could be refined for greater precision or scale?
Evaluating DDP Opportunities
Not every DDP is worth solving. Weigh them against:
1. Impact Potential
- Who’s affected, and how many?
- What’s the cost of the status quo?
- Incremental fix or game-changer?
2. Readiness to Solve
- Can you access domain know-how? (If not, try reports or proxies like retirees.)
- Are experts open to collaborating?
- Is tech up to the task?
3. Market Forces
- Persistent problem or new pain point?
- Are organizations hunting for answers?
- Is pressure (regulation, competition) mounting?
4. Complexity Balance
- Is the hurdle more domain or tech?
- Can current tools model the knowledge? (e.g., NLP for regs, but not "gut feel")
- Does partial progress add value?
Tools to Get Started
- Process Mapping: Diagram exceptions.
- Interviews: Record experts to capture tacit gems.
- Data: Measure exception rates or downtime costs.
Watch the Risks
Don’t chase a loud stakeholder’s pet peeve—cross-check with data and diverse voices to confirm it’s a true DDP.
Conclusion
Deep Domain Problems hide in plain sight, masquerading as "just how it works" in an industry. They’re where complex knowledge meets technical limits—and where solving them sparks transformative value. As AI races ahead, spotting DDPs separates shallow wins (digitizing the obvious) from deep innovation. Pair AI’s power—say, pre-flagging risks—with human insight to define what matters, and you’ll unlock breakthroughs that reshape industries.