Research Gap Analysis
Discover meaningful research gaps and feasible entry points in your field, categorized by methodology, population, or mechanism.
- Fill in your research domain and focus area.
- Click AI Run — the research gap analyst will introduce itself and ask you to share your source material (paper abstracts, key passages, a DOI list, or a research topic) in the chat.
- Paste your source material into the chat.
- Receive a structured gap analysis — ask follow-up questions to explore specific gaps further.
Research gap analysis is the rigorous process of identifying what a body of literature has not yet answered — and why those unanswered questions matter. It is the intellectual foundation of every strong grant application and thesis proposal. Reviewers and funding committees consistently reject proposals that fail to convincingly establish why new research is needed; a well-executed gap analysis provides exactly that justification in a structured, credible form.
This tool systematically analyzes a provided set of papers, abstracts, or topic descriptions using a multi-dimensional gap classification framework. Rather than defaulting to the generic observation that "more research is needed," the system categorizes gaps by type: methodological gaps (existing studies used flawed or outdated designs), evidence gaps (a question exists theoretically but no empirical study has addressed it), cross-context gaps (findings from one population or setting have not been tested in another), and subgroup heterogeneity gaps (aggregate results mask important differences between patient subgroups).
A distinctive feature is the built-in anti-sycophancy mechanism. Rather than simply agreeing with the framing the user provides, the AI is instructed to actively search for evidence that each proposed gap may already be addressed in recent literature. This prevents investigators from building proposals around gaps that were closed by a 2023 multi-center trial they simply missed. Users receive not just a list of gaps but a confidence assessment for each, based on how thoroughly the gap appears to remain open.
The tool supports multiple output modes aligned to specific use cases: Grant Application mode emphasizes clinical significance and innovation framing; Thesis mode generates a broader range of entry points suitable for a multi-year project; Research Proposal mode provides a structured justification section ready to paste; General Exploration mode provides a landscape without prescriptive framing.
Best results are achieved when 3 to 10 high-quality source papers are provided as input. The AI merges these with its broader knowledge of the field to identify gaps that neither the user's papers alone nor the AI alone would surface. PDF uploads are supported for direct analysis of full-text articles.
All identified gaps should be independently verified in PubMed before being cited in a formal proposal. A fresh search session (separate from the AI conversation) is the most reliable verification method, as it avoids confirmation bias introduced by the prior context.