Why Speed, Automation, and AI Are Forcing Us to Pay Back Our Technical Debt
Data Is Still the Center of Everything
The start of a new year is the perfect opportunity to reflect on 2025 and make thoughtful predictions for 2026. It’s very tempting to talk about tools — AI, platforms, automation, decision intelligence, speed. Different words but the same ambition. Everyone wants better decisions, faster. I won’t focus on the tools themselves, at least not directly.
Putting aside the buzzwords, one thing hasn’t changed — data is still the center of everything. Not in a philosophical way. In a very practical, sometimes uncomfortable way. No matter whether we’re talking about reporting, advanced analytics, AI-driven workflows, or executive decision-making, the same question keeps resurfacing: Can we trust the data underneath?
Most organizations still can’t answer that confidently. That’s not a new insight. What is new is that fewer people are pretending otherwise.
2025 Didn’t Change the Problem, It Gave It a Spotlight
If 2025 taught us anything in the industry, it’s that layering new technology on top of old data problems doesn’t solve them. It exposes them faster. At many events last year, I heard impressive stories about automation, analytics platforms, and AI-assisted workflows. But underneath, the same tension kept resurfacing:
- Data exists, but it isn’t trusted consistently
- Validation processes focus on documentation over confidence
- “AI-ready” pipelines still rely on manual exports and handoffs
- Apps and dashboards are built faster than the data they depend on can evolve
When decisions were slower and more manual, imperfect data could hide. For years, we optimized outputs and quietly ignored inputs. Now speed, automation, and AI are forcing us to pay that debt back.
Data as an Asset Is a Nice Idea. Data as a System Is the Reality
For years, we’ve talked about data as an asset. Something to collect, standardize, lock, and archive. That framing made sense when the main output was a table, a listing, or a static report. But it breaks down in modern analytics environments. Today, data behaves more like a living system. Yet many organizations still manage data as if its job ends at database lock.
It doesn’t matter whether your organization is focused on clinical reporting, exploratory analytics, real-world evidence, AI-assisted review, or operational decision support. If data is the foundation, then quality, traceability, and context aren’t optional features; they’re structural requirements.
This is where a lot of 2025 conversations became uncomfortable. Because many teams have invested heavily in:
- Dashboards that look good but explain little
- Models that perform well on paper but struggle in practice
- AI pilots that work in isolation and collapse at scale
The issue here is rarely ambition. It’s unclear ownership. And it’s data that is treated as something to be consumed, not something to be managed as a product over time.
One encouraging trend I noticed in 2025 isn’t about tools — it’s about questions. More teams ask:
- Where does this data come from, really?
- What happens to it after we export it?
- Which checks reduce risk, and which just satisfy tradition?
This might look subtle, but to me it’s important. It’s a move away from “What can this tool do?” toward “What decisions are we trying to support, and what data confidence do those decisions require?”
Why JPM Still Matters
With the J.P. Morgan Healthcare Conference around the corner, it’s tempting to focus only on deals, valuations, and strategic announcements. And yes, those matter, especially in a market still looking for momentum. But JPM has always been about more than transactions. It sets the tone for what leaders prioritize and what they signal:
- Where investment is flowing
- What capabilities are being acquired instead of built
- Which operational problems are being quietly acknowledged
When I listen to JPM announcements, I pay less attention to buzzwords and more to patterns. Those shifts don’t happen by accident. Because investors, regulators, and partners are getting better at spotting the gap between vision slides and execution. Behind every acquisition or platform strategy announcement is also a hidden assumption about data maturity, whether directly stated or not.
What to Watch Going Into 2026
Looking ahead, I don’t think the biggest changes will come from new tools. They’ll come from how organizations rethink ownership and workflows. A few signals to watch:
- Data quality metrics that persist over time, not just at lock
- Validation strategies focused on risk and confidence, not volume
- Analytics apps designed around users, not outputs
- Fewer tools, better integrated, and more willingness to retire what doesn’t work
None of this is glamorous. Most of it won’t make headlines. But teams with strong data foundations can adopt new tools faster, discard bad ideas earlier, and scale good ones without rebuilding everything. Teams without them will keep restarting.
Another thing I appreciated in 2025, and hope continues in 2026, is a growing willingness to be honest. Progress in this space doesn’t come from pretending everything is solved. It comes from acknowledging where it isn’t, and designing systems that reflect reality, not paper aspiration.
Looking Ahead
Working closely with pharma and life sciences teams, I have seen patterns like advanced ambitions constrained by data processes, people compensating manually for systemic gaps, and innovation happening despite infrastructure, not because of it. The most successful teams are not the ones chasing every trend. They’re the ones investing in data as an operational backbone, across reporting, analytics, applications, and decision support.
As we head into JPM and 2026 planning cycles, I’d encourage everyone to ask a few uncomfortable but necessary questions:
- If our output doubled tomorrow, would our data processes hold?
- Are we building intelligence on top of stability, or complexity?
The answers matter more than the tools you choose in the next year. Because no matter how advanced analytics, AI, or decision frameworks become, data remains the center. That’s where real progress starts.
Aga Rasińska
Associate Director of Strategy, Atorus
With more than a decade of hands-on experience in bioinformatics, data science, and program leadership, Aga Rasińska brings a dual perspective that bridges business strategy and technical innovation, transforming complex clinical and omics data challenges into scalable, results-driven solutions. Her expertise spans project and product management, change leadership, and data-driven decision-making within regulated life sciences environments. She is also a frequent industry speaker, panelist, and contributor focused on the intersection of science, data, and business strategy.
References & Resources
TransCelerate Biopharma Inc. Modernization of Statistical Analytics. Published 2020.
U.S. Food and Drug Administration. Real-World Evidence: Considerations Regarding Non-Interventional Studies for Drug and Biological Products. Published March 2024.
International Council for Harmonisation. Integrated Addendum to ICH E6(R1): Guideline for Good Clinical Practice E6(R2). Published 2016.
Frequently Asked Questions
Q: What is the difference between “data as an asset” and “data as a system”?
A: Viewing data as an “asset” often implies a static resource to be stored and archived. Viewing data as a “system” acknowledges that it is a living, evolving entity that requires continuous management, traceability, and validation across various workflows and life cycles.
Q: Why is a data foundation more important than AI tools in 2026?
A: Advanced tools like AI and automation amplify the quality of the data they are fed. Without a strong, validated data foundation (“data confidence”), AI tools will only accelerate errors or produce untrustworthy insights.
Q: How does Atorus support clinical data strategy?
A: Atorus helps organizations build “Data That Does,” moving beyond simple reporting to create validated, scalable data ecosystems that support advanced analytics, AI adoption, and regulatory compliance.