Why Decision Making, Not Experiments, Is the Real Bottleneck in Drug Discovery

This article is adapted from a talk delivered by Allan Jordan at the Drug Discovery Chemistry (DDC) conference on April 14^th 2026.

In drug discovery, progress is often framed around experiments: the design–make–test cycle, assay throughput, or the latest platform technology. While these are undeniably important, they’re not usually where projects stall.

More often, delays arise after the data are generated.

What ultimately determines whether a project moves quickly, stalls, or fails is decision‑making: when decisions are made, how confidently they’re made, and whether teams are aligned on what the data actually means.

This perspective has become increasingly clear across the industry, particularly as timelines and costs continue to come under pressure.

Decision‑Making as a Competitive Advantage

Across drug discovery organizations, there’s growing recognition that speed isn’t just experimental. It’s strategic.

Recent analyses suggest that the average time from project inception to clinical candidate nomination is approximately four years, with best‑in‑class programs closer to three. Many organizations now view two years as an aspirational target.

The challenge is clear: How do teams compress timelines without sacrificing scientific quality or increasing risk?

A recurring theme in successful programs is not simply faster experiments, but faster transitions from data to action.

Reducing the “White Space” Between Data and Action

Much of the hidden delay in drug discovery sits in what can be thought of as white space: the gap between data generation and decision‑making.

One increasingly common best practice is to pre‑plan decisions before results exist:

• If the data are positive, what happens next?
• If the data are negative, where do we pivot?
• If the data are ambiguous, what additional information is required?

When decisions are defined in advance, data can immediately drive action rather than trigger prolonged debate.

Organizations aiming to minimise white space often focus on:

• Cross‑disciplinary integration across chemistry, biology, DMPK, in vivo, and computational science
• Rapid, shared access to project data rather than siloed ownership
• Interpreting compounds using multiparametric datasets rather than single‑assay performance

A particularly important mindset shift is pushing compounds until the data clearly say stop. This doesn’t mean assuming every compound will succeed, but ensuring that each one is evaluated rigorously enough to support a confident decision, positive or negative.

Finding the “Goldilocks” Compounds

Not every successful compound excels in one metric.

Many of the most promising candidates are “Goldilocks” molecules: not exceptional in isolation, but well‑balanced across exposure, free drug levels, functional activity, and developability. These compounds often emerge only when teams look holistically rather than optimising one parameter at a time.

Finding these molecules earlier requires faster, more meaningful translation from chemistry into biology.

Direct‑to‑Biology: Going Beyond Affinity

Direct‑to‑Biology approaches are now widely adopted, particularly for early binding assessment using techniques such as SPR. While powerful, affinity alone does not always predict functional or cellular activity.

A more challenging question is:

Can compounds be assessed directly from crude reaction mixtures using functional biological assays?

This requires careful assay design. Many biochemical and cellular systems are sensitive to residual solvents, catalysts, or by‑products. Where assays are engineered to tolerate these components, strong concordance can be achieved between:

• Crude and purified compounds in biochemical assays
• Crude materials and cellular activity
• Crude mixtures and early developability metrics (e.g. ChromLogD, EPSA)
• Crude mixtures and metabolic stability in human and mouse microsomes

The ability to assess binding, function, cellular relevance, and early DMPK without purification can significantly accelerate decision‑making, provided teams clearly understand when this approach is appropriate, and when it is not.

Extending Direct‑to‑Biology Decisions

Direct‑to‑Crystallography

In robust crystallography systems, crude reaction mixtures can be soaked directly to obtain bound structures, even from reactions with modest yields. Crucially, effort is focused only on datasets that provide new biological insight.

Pushing this upstream further, using screening stocks directly for structural hit validation, can remove months of resynthesis and reconfirmation, shortening hit‑to‑lead timelines by several months.

Direct‑to‑Discovery

When combined with high‑throughput chemistry platforms and AI‑guided design, these approaches create a workflow where thousands of compounds can be explored rapidly, informatively, and more sustainably without generating unnecessary waste or low‑value data.

Does It Work? A Real‑World Case Study

In a recently published collaboration with Storm Therapeutics, the goal was ambitious:
to deliver a preclinical candidate against a helicase target within two years.

By combining parallelised screening, Direct‑to‑Biology assays, direct‑to‑crystal SAR, early biomarker strategy, and proactive developability assessment, a best‑in‑class molecule was delivered in under 18 months, demonstrating complete and durable tumor regression in vivo.

This represented:

• A substantial reduction in timeline versus industry benchmarks
• A significant reduction in effective project cost
• Earlier delivery of a high‑quality candidate toward the clinic

What This Really Means

Faster discovery is not about sacrificing rigour.

Fast data are valuable.
Slow data can still be useful.
Poor data is useless.

When teams trust their data and align on decisions early, programs move with confidence instead of hesitation. While Direct‑to‑Discovery approaches won’t suit every target or biology, the underlying principle applies universally:

Better decisions, made earlier, are one of the most powerful levers for accelerating drug discovery.

Author Profile

Watch the talk on our YouTube Channel