Every research paper begins with the same problem: choosing a question worth answering. Authors often start with a broad topic they find interesting and struggle to narrow it into a study. The missing step is usually the research gap — the specific piece of knowledge that existing studies have not provided. A paper built on a clearly identified gap has a ready-made justification, a focused design, and a natural place in the literature. A paper built only on a broad interest tends to wander, because it never decided what exactly it was adding.
This guide explains how to find a research gap by reading strategically, describes the main types of gap, shows how to turn a gap into a well-formed research question, and gives examples from veterinary and animal sciences. It is written for authors planning their first studies as well as for experienced researchers starting work in a new area, and it connects directly to the later steps covered in our guide on how to publish a research paper.
A research gap is not simply a topic nobody has written about. Very few topics are entirely untouched, and claiming that nothing exists on your topic usually means the search was too narrow rather than the field being empty. A gap is a specific, defensible statement about what is missing: a population that has not been studied, a method that has not been applied, a contradiction that has not been explained, or a finding that has not been tested in a new setting. The more specific the statement, the more useful it is, because a specific gap points directly at a study design.
A gap must also matter. A missing detail that no reader, practitioner, or policymaker would care about is not a useful gap, even if it is real. Before committing to a gap, ask who benefits if it is filled: does it change practice, resolve a disagreement, extend a finding to a new population, or improve a method? If the honest answer is nobody in particular, keep looking. Reviewers judge a paper's contribution by the gap it fills, so choosing a gap that matters is the highest-leverage decision in the whole research process.
Gaps rarely announce themselves in a single paper. They appear when you read across papers and compare them. The most reliable strategy is to read the discussion and limitations sections of the most relevant studies carefully. Authors routinely state what their study could not do, which populations it did not cover, and which questions remain open. Those statements are the field's own list of candidate gaps, written by the people who know the work best. Collect them as you read, and patterns will emerge: the same limitation mentioned by several studies is a strong candidate for a new study.
A second strategy is to read for disagreement. When two good studies report different results, the contradiction itself is a gap: something explains the difference, and nobody has yet shown what it is. Common explanations include different populations, different measures, different settings, or different methods. A study designed to test which explanation holds is often more valuable than another study repeating one side of the debate, because it resolves the disagreement rather than adding to it.
A third strategy is to read recent review articles and editorials with a gap-hunting eye. Reviews summarise the field and usually end with a section on future research directions; editorials often call for work on neglected problems. Treat these as starting points, not conclusions: verify each suggested gap against the original studies before adopting it, because a review's suggestion may already have been addressed by work published since the review appeared. Our guide on how to write a literature review explains the searching and synthesis skills this strategy depends on.
Knowing the types helps you recognise a gap when you see one, and helps you describe it precisely to reviewers. Most gaps fall into one of the following types, and many real gaps combine two of them.
A finding established in one population has not been tested in another. For example, a feeding practice studied on large commercial dairy farms may never have been examined on smallholder farms, where herd size, resources, and management differ in ways that could change the result. Population gaps are common and practical, because they extend existing knowledge to the readers who need it most.
A question has been studied with one method but not with another that could answer it better. A topic studied only through surveys, for instance, may benefit from direct measurement; a topic studied only in laboratories may need field testing. Method gaps are strong when the new method removes a known weakness of the old one, rather than merely repeating the study in a different way.
Evidence gaps occur when too few studies exist to support a conclusion, and contradictory evidence occurs when the studies that exist disagree. Both point at the same response: a study designed to add reliable evidence or to explain the disagreement. These gaps are persuasive to reviewers because they arise from the literature itself rather than from the author's preference.
A finding from one region, production system, or period may not hold in another. Animal disease patterns, for example, vary by climate and management system, so a control measure validated in one setting may need testing in another before it can be recommended there. Time gaps appear when an older finding predates changes in practice, technology, or conditions, and needs re-examination.
A gap is a statement about what is missing; a research question is a statement about what your study will find out. Turning the first into the second takes three steps. First, state the gap in one sentence: what exactly is unknown, and in which population or setting? Second, narrow it to what one study can answer: a question that needs five studies is a programme, not a paper. Third, phrase it so that the answer is observable: the question should point at data you can collect and an analysis you can perform, not at a general discussion.
Consider a population gap as an example. The gap statement might be: "The effect of a specified feeding practice on milk yield has been studied on large commercial farms but not on smallholder farms." The research question follows directly: "Does the specified feeding practice affect milk yield on smallholder dairy farms, and if so, by how much?" Notice that the question names the population, the exposure, and the outcome — the three elements any reader needs to judge whether the study answers it. A question missing any of the three is not yet ready to design around.
Test your question before you commit to it. Ask whether the answer is already known (if so, the gap was not real), whether one study can answer it (if not, narrow it further), and whether the answer would change anything for a reader (if not, the gap does not matter enough). Also check feasibility honestly: can you reach the population, collect the data, and complete the analysis with the resources you have? A well-formed question that cannot be answered with available resources is a plan for a stalled project, not a paper.
Finally, test the question on a colleague or supervisor before designing the study. Explain the gap and the question in plain language, without the paper's jargon, and listen to what they ask back. If they immediately understand what is unknown and why it matters, the question is ready. If they ask which population you mean, what exactly you will measure, or why anyone should care, those questions point at the parts of the question that still need tightening. This short conversation often saves months of work on a study whose question was never quite clear.
Suppose several studies report that a vaccination schedule reduces disease incidence in intensively managed poultry flocks, but none has tested the same schedule in free-range systems, where exposure patterns differ. The gap is a population and setting gap combined. The research question becomes: "Does the vaccination schedule reduce disease incidence in free-range poultry flocks, and how does the effect compare with intensive systems?" That question is specific, answerable by one comparative study, and useful to veterinarians and producers working in free-range systems.
Another example: studies of a diagnostic test may all have been conducted in referral laboratories with high case volumes, while most practitioners use the test in field clinics with different sample handling. The gap is a setting and method gap: the test's performance under field conditions is unknown. The research question — "What is the diagnostic accuracy of the test under field clinic conditions?" — leads naturally to a field validation study, and its answer directly affects practice.
A third example involves contradictory evidence: two studies of a nutritional supplement report opposite effects on growth rate, one using one breed and one using another. The gap is the unexplained contradiction. A study designed to test the supplement in both breeds under the same conditions would resolve the disagreement, and its result would be cited by everyone who works on the topic afterwards. Resolving contradictions is often the fastest route to a paper that matters, because the field is already waiting for the answer.
The most common mistake is choosing a gap by personal interest alone, without checking whether the literature supports it. Interest is a good starting point, but the gap must survive contact with the literature: if a search shows the question already answered, the gap is closed, and persisting with it wastes the study. The second mistake is choosing a gap so broad that no single study can fill it, such as "improving animal health in developing countries" — that is a field, not a gap. The third is choosing a gap so narrow that the answer interests nobody, such as a minor variation on a well-settled finding.
Another frequent error is mistaking a gap in your own reading for a gap in the field. Before claiming a gap, search thoroughly, including the reference lists of key papers and the most recent work in the area. A gap that disappears on a second, wider search was never a gap. Finally, avoid building a study on a gap you cannot defend in one paragraph: if you cannot explain, in plain language, what is missing and why it matters, reviewers will not be convinced either. Write the gap paragraph early — it will become the core of your introduction and your abstract.
| Gap type | What is missing | Example research question |
|---|---|---|
| Population | Finding untested in another population | Does the practice affect yield on smallholder farms? |
| Method | Question studied with only one method | What do direct measurements show, compared with surveys? |
| Evidence | Too few studies to support a conclusion | Does the intervention work when tested at larger scale? |
| Contradiction | Studies disagree without explanation | Which factor explains the different results? |
| Setting or time | Finding untested in a new setting or era | Does the measure still perform under field conditions? |
These authoritative external resources explain the standards and guidance mentioned in this article:
How do I know if a gap is real and not just something I have not read yet?
Search thoroughly before claiming a gap: use several databases, follow citation trails, and check the most recent work. If a wider search closes the gap, it was a gap in your reading, not in the field.
Can a gap be too small to base a paper on?
Yes. If the answer would interest nobody beyond yourself, the gap is too narrow. A good gap matters to readers, practitioners, or the field's understanding — test it by asking who benefits if it is filled.
Should my research question mention the population, exposure, and outcome?
Yes. A question naming all three tells the reader exactly what the study will test and makes the design easy to judge. A question missing any element is not yet ready to design around.
What if two studies contradict each other on my topic?
That contradiction is itself a valuable gap. Design a study that tests which explanation accounts for the difference — populations, measures, settings, or methods — rather than simply repeating one side of the debate.
Do I need a gap before I start collecting data?
Yes, in almost all cases. Data collected without a question tends to be unfocused and hard to write up. State the gap and the question first, then design the study around them.
Can one paper fill more than one gap?
It can, but the paper should still have one primary question. Secondary gaps can be addressed as supporting aims, but a paper with several equal questions usually lacks focus and is harder to review.