How to Publish a Systematic Review: A Step-by-Step Guide

Doing a systematic review and publishing one are two different skills.

Plenty of researchers finish a technically correct review and then watch it collect desk rejections. The method was sound. The paper still was not publishable. Those are not the same thing, and the gap between them is where most reviews quietly die.

I have published systematic reviews, supervised dozens of them, and sat on the other side of the desk as a reviewer. So this guide is not about how to tick the boxes of the method. It is about how to publish a systematic review that an editor sends out for review and a journal accepts.

What most researchers get wrong

Most people treat a systematic review as a procedure. Define a question, search some databases, screen the results, extract the data, write it up. Follow the steps and you get a paper.

The steps matter, but they are not what gets you published. Editors see hundreds of methodologically fine reviews that say nothing new. A clean PRISMA diagram does not save a review that has no gap and adds no value. The work of publishing happens before you search and after you write, and that is the part almost nobody teaches.

The systematic review pipeline

Every review follows roughly the same pipeline. The point of it is rigour: a transparent, reproducible process that anyone could follow to reach the same result.

1. Define the question, including your inclusion and exclusion criteria.
2. Build the search strategy: databases, terms, Boolean logic.
3. Screen: run the search, deduplicate, screen titles and abstracts, then full texts.
4. Extract the data.
5. Synthesise and write up.

That is the machine. Now here is how to run each stage so the result is publishable, not just complete.

Step 1: Choose a topic that can actually be published

This is where the outcome is mostly decided, and where most people spend the least time.

I use what I call the Convergence Method: a strong topic sits at the intersection of feasibility, passion, and debate.

• Feasibility. Can you finish it in 6 to 12 months with the data and skills you have or can quickly get? A quick Google Scholar search should surface at least 8 to 10 relevant studies. Too few and your topic is too narrow or too new.
• Passion. Could you talk about this topic at a party without losing interest? Reviews demand sustained engagement. A topic you do not care about will stall.
• Debate. Does the topic connect to a live argument, where there is uncertainty, conflicting findings, or new evidence emerging? That is what makes a review worth reading.

Then run two quick tests. The Duplication Test: do up-to-date reviews already exist? If none exist or they are outdated, you may have found your gap. And look for low-hanging fruit: areas that are highly cited but understudied, where a review is overdue.

Finally, the Grandmother Rule. If you cannot explain your topic clearly in two or three sentences to someone outside your field, it is not focused enough yet.

Step 2: Turn the topic into a precise question

A vague question produces a vague review. In health and social sciences, the cleanest tool is PICO:

• Population: who is being studied?
• Intervention or exposure: what factor are you examining?
• Comparison: what is the alternative or control?
• Outcome: what effect are you measuring?

For example: in adults with type 2 diabetes, does intermittent fasting compared to standard calorie restriction improve glycemic control? That is a question a reader can hold in their head, and a question your whole review can answer.

Step 3: Build a search strategy a reviewer cannot fault

This is the stage reviewers probe hardest, because it is where bias creeps in. Your search has to be transparent and reproducible. Name your databases, write out your full search strings, and show how you combined terms with Boolean operators. If a reviewer cannot rerun your search and land in the same place, you have a problem.

If you lack access to paid databases, Semantic Scholar and OpenAlex are workable free options for building and checking coverage.

Step 4: Screen and extract so the work is defensible

Screen against your criteria, not against your hopes for the result. Record what you exclude and why.

For extraction, build your data table so it reads left to right and tells each study's story, from methods through to findings. For every study, capture what they did, what they found, the statistical significance, and the effect size. Record the null results too. Leaving them out is how reviews drift into bias and how reviewers catch you.

Step 5: Write it so it survives the desk

Here is the part that separates a finished review from a published one. Before you submit, run your paper through what I call the Publishability Formula:

Gap x Value Add x Alignment x Clarity x Fit

It is multiplicative on purpose. Any single zero takes the whole paper to zero.

• Gap. Is the hole in the literature stated plainly, and measured against your nearest-neighbour papers? If you cannot say what is missing, you have no reason to exist.
• Value add. What does your review give the field that those nearest-neighbour papers do not? Recombining what is already known is not enough.
• Alignment. Do your conclusions follow from your evidence? Overclaiming past your data is one of the fastest routes to rejection.
• Clarity. Can a busy editor grasp your contribution in the first paragraph?
• Fit. Is this the right journal for this question and this audience?

Most reviews that get desk rejected fail on gap and value add. The method was fine. There was just no clear reason for the paper to be published. Fix the weakest element before you submit, not after the rejection.

Get eyes on it before you submit

The single highest-return thing you can do is get expert feedback before an editor sees your work, not after. A reviewer's time is free and brutal. A mentor's time, before submission, is what actually moves a paper from rejected to accepted.

That is exactly what we built the Research Collective for. You can start with a one-week trial and work through the full step-by-step systematic review course at your own pace: https://courses.fasttrackgrad.com/offers/3aJwdxgT/checkout

Inside the community you can also post your own work and bring it to our live workshops, where you get my personal feedback on your review, your question, your gap, and your write-up. Try here it here for 7 days: https://courses.fasttrackgrad.com/offers/AxvzGF6g/checkout

Why I believe our system is the best in the world

I will be upfront. I am biased. I built this, so of course I think it works.

But here is the honest case. AI can now flood the field with reviews that look convincing and are quietly wrong. That does not make a properly executed review worth less. It makes it worth more. The durable skill is live expert judgment, named frameworks you can apply yourself, real accountability, and a track record of papers that actually got published. That is what we teach, and it is what a tool cannot give you.

I have spent twenty years publishing and supervising this work, across more than 400 peer-reviewed papers. The frameworks in this guide, the Convergence Method, PICO, the Publishability Formula, are the same ones I use on my own papers and the same ones our members use to get published. You do not have to take my word for it. You can test the whole thing for the price of a coffee.

Start here

If your review is sitting in a drawer, or you are about to submit and something feels off, do not guess. Run the Publishability Formula on it first, then bring it to us.

Start your one-week trial here and get the full systematic review course: https://courses.fasttrackgrad.com/offers/3aJwdxgT/checkout