A research gap is a question that existing studies leave unanswered, or answer badly. It is the reason your paper needs to exist. Without a clear gap, a paper is a summary. With one, it is a contribution.
This guide explains what counts as a gap, shows the main types with examples, and gives you an eight-step method to find one in your own field. It ends with a template that turns a gap into a research question. If you are still planning your paper as a whole, start with our beginner's guide to writing a research paper.
Key Takeaways
A research gap is something the literature leaves unresolved: a missing answer, a contradiction, an outdated result or an untested context.
Few search results do not prove a gap. Search with synonyms across several databases before you decide.
The best places to look are the Limitations and Future Work sections of recent papers.
A strong gap is real, relevant, researchable and original. Test any candidate against all four.
State your gap in one sentence, then connect it directly to your research question.
What Is a Research Gap?
A research gap is a specific, evidenced weakness in what is currently known. It might be a question nobody has asked, a result that studies disagree on, or a finding that has not been tested in a new setting.
A gap is not simply "nobody has done this". Sometimes nobody has done a thing because it is not worth doing. A useful gap passes two tests: the literature shows that something is missing, and filling it would change what researchers or practitioners know or do.
Gaps usually appear for one of these reasons:
Thin or outdated evidence. The last serious study is years old, and the field has moved on.
Conflicting results. Studies reach different conclusions and nobody has explained why.
Overlooked contexts. Earlier work ignored a population, setting, dataset or variable.
Weak methods. Earlier studies used small samples, narrow benchmarks or untested designs.
New developments. A new technology or concept has not yet been studied properly.
Types of Research Gaps
Authors classify gaps in slightly different ways, so treat the types below as prompts, not a fixed standard. The examples are illustrative.
Type | What is missing | Illustrative example | First move |
|---|---|---|---|
Evidence gap | Studies contradict each other | One study finds deep models beat classical methods on a medical dataset, another finds classical methods match them | Compare data, methods and populations to explain the conflict |
Empirical gap | Real-world data or testing | A technique is proposed but only tested on small, old or synthetic datasets | Collect new data or run a fresh experiment |
Knowledge gap | Basic understanding | A new technology appears and its effects are still unstudied | Run an exploratory study |
Conceptual gap | Clear definitions | Researchers use one term for several different ideas | Compare definitions and propose a clearer framework |
Methodological gap | Better methods | Earlier work used only qualitative interviews or a single benchmark | Repeat the question with a different or stronger method |
Theoretical gap | A theory that fits | An older model cannot explain a newer setting, such as platform businesses | Extend, refine or combine models |
Practical gap | Theory applied in practice | A method works in the lab but has not been tried in the field | Test it in realistic conditions |
How to Find a Research Gap in Eight Steps
Choose a broad area, then confirm it fits your programme. Check your institution's brief or ask your supervisor what level of originality is expected.
Search widely. Use Google Scholar, your university library, and open sources such as DOAJ and arXiv. Combine many keywords and synonyms, and filter for recent years.
Build a reading list. Collect five to ten relevant papers per keyword. Prefer recent papers and review articles, which summarise a field quickly.
Triage each paper. Read the Abstract, the Introduction and the Discussion or Conclusion first. That tells you what the authors set out to do and what they found.
Scan for signal phrases. Search each paper for "future work", "further research is needed", "remains unclear", "has not been" and "limitations". Authors often point to gaps themselves.
Keep a gap log. Record every candidate in a table, as shown below, so patterns emerge across papers.
Look for contradictions and silences. Compare results, methods and populations across your papers. Note what authors do not mention as well as what they claim.
Verify before you commit. Search again with new keywords and check recent preprints to make sure nobody has already filled the gap. Then evaluate your short list and pick the strongest candidate.
Here is a simple gap log you can copy:
Paper | Main claim | Limitation or future work | Possible gap |
|---|---|---|---|
Author, year | What they showed | What they say is missing | Your reading of it |
Repeat the same limitation across several papers and you have found a strong signal.
How to Tell Whether a Gap Is Worth Pursuing
Not every gap deserves a paper. Test each candidate on four points:
Real. The literature, not your assumption, shows that something is missing.
Relevant. Answering it matters to your field, not only to you.
Researchable. You have the time, data and methods to address it.
Original. It adds something new, rather than repeating established findings.
If a candidate fails one of these, keep it in your log and move on. A smaller gap you can fill well beats a large one you cannot.
Turning a gap into a paper at an SAI conference?
Once you have a gap and a research question, the Call for Papers pages for IntelliSys, Computing Conference and CVC list the tracks you can submit to. Check the IntelliSys tracks
From Gap to Research Question: A Worked Example
Imagine you read several papers on a vision model that detects plant disease. Each paper tests it on leaves photographed in a lab. Two of them list "performance in field conditions was not evaluated" under Limitations.
You have found a candidate gap. Now write it as a gap statement:
Prior work has evaluated plant disease classifiers on controlled lab images. However, performance under real field lighting and backgrounds is unreported, because no standard field dataset exists. This study addresses that by testing [model] on field-captured images.
A simple template is: Prior work has [what is known]. However, [what is missing] because [why it matters or why it persists]. This study addresses it by [your approach].
The research question follows directly: "How much does classification accuracy drop when the model is tested on field images instead of lab images?" Notice that the question, the method and the gap all point at one another. If they do not, revise until they do.
Common Mistakes When Identifying Gaps
Treating a short search as proof. Few results can mean you used the wrong keywords. Try synonyms and neighbouring terms.
Assuming every limitation is a gap. A limitation matters only if filling it would teach us something worthwhile.
Changing the sample and calling it new. Swapping the location, population or dataset is not novel unless you can explain why the result might differ.
Ignoring conflicting evidence. Studies that disagree often hold your best opportunity, so compare them closely.
Stating the gap vaguely. Say what is unclear, why it matters and how your question resolves it.
Frequently Asked Questions
What is a research gap?
A research gap is a question or problem that existing studies leave unanswered, answer inconsistently, or have not tested in a relevant setting. It justifies why new research is needed.
Why is it important to identify a research gap?
The gap explains why your study exists and shapes your research question, method and contribution. Reviewers look for it in the Introduction, so a missing or vague gap is a common reason for rejection.
What are the types of research gaps?
Common types include evidence, empirical, knowledge, conceptual, methodological, theoretical and practical gaps. Authors label them differently, so use the types as prompts for what might be missing.
Can I use the limitations of earlier studies to find a gap?
Yes, limitations and future work sections are a good starting point. Check that the limitation is genuinely important and that nobody has addressed it since, then decide whether you can fill it.
How can I find a research gap quickly?
Start with a focused topic, read recent review articles, and scan the Limitations and Future Work sections of five to ten key papers. Log what you find and look for repeated signals.
Is a research gap the same as a research problem?
No. The gap describes what is missing in the existing literature. The research problem is the specific issue your study sets out to solve, and it is usually framed to fill that gap.
How many gaps should one paper address?
Aim for one primary gap. A paper that tries to fill several at once usually does none of them well, and secondary gaps can be saved for future work.
Ready to Turn Your Gap Into a Paper?
A clear gap, a focused question and a sound method make a strong submission. SAI Conferences organises three in-person events where you can present the work. Each Call for Papers page lists tracks, review details and current submission rounds.
SAI event | When and where | Submit |
|---|---|---|
13th edition | 2 to 3 September 2027 Mercure Amsterdam City Hotel, Amsterdam, Netherlands | |
15th edition | 8 to 9 July 2027 Clayton Hotel Chiswick, London, United Kingdom | |
Computer Vision Conference 2027 | 15 to 16 April 2027 Mercure Amsterdam City Hotel, Amsterdam, Netherlands |
Submit to IntelliSys 2027 | Submit to Computing Conference 2027 | Submit to CVC 2027
Related Guides
SAI organises IntelliSys, Computing Conference and CVC. SAI cannot guarantee that any paper will be published or indexed, and the reviewers and indexing databases make those decisions. Confirm all details on each event page before you submit, and see the Open for submissions panel on this page for live round status.