## Citations follow discoverability, not luck

Most advice on how to increase citations skips the part that actually matters. It tells you to publish in high-impact journals and network at conferences, which is fine, but it treats citations as a reward for prestige. The truth is more mechanical and more in your control: a paper gets cited when the right researcher finds it, reads enough to see it is relevant, and can pull the reference without friction. Break that chain at any point and the citation never happens, no matter how good the work is.

So the single most useful reframe is this. Before you think about impact factor or your follower count, ask whether a researcher working on your exact problem could find your paper through a normal search. If the answer is no, every other tactic is wasted effort. This post walks through the levers that genuinely move citation counts, in rough order of how much they matter, and it is honest about the ones that mostly generate noise.

## What actually drives citations (and what doesn't)

There is a comforting myth that you can tweet your way to a well-cited paper. The evidence does not support it. In a controlled study where scientists with large followings shared papers over ten months, the tweeted articles earned 81% higher altmetric scores after three years but were cited only 7% to 12% more, a difference that was not statistically significant. Social media reliably buys you attention. It does not reliably buy you citations.

Two things show a stronger, more consistent link to citations. The first is open access: when anyone can read the full text without a paywall, more people cite it. The second is preprints: posting a preprint is associated with both more attention and a measurable citation bump for the eventual journal version, partly because the work becomes findable and citable months earlier.

The pattern underneath all of this is simple. Tactics that put the actual full text in front of more of the right readers work. Tactics that only generate visibility metrics mostly do not. Keep that test in mind for every suggestion below.

## How to increase citations before you ever submit

The highest-leverage citation work happens before submission, when you write the title, abstract, and keywords. These three fields are what indexing engines read, and they are what a researcher skims in a results list to decide whether to click. Get them wrong and your paper is invisible to search even if it is brilliant.

This is academic SEO, and it is not a dirty word. It means writing the metadata of your paper so that the people searching for your contribution can find it. A few concrete rules:

1. Put your main keyword phrase in the first 65 characters of the title. Search engines weight early words, and a researcher scanning results sees the front of your title first.
2. Repeat the primary term naturally in the abstract, ideally in the first two sentences. The abstract is the most heavily indexed text in your paper.
3. Prefer a short, descriptive title over a clever one. Papers with concise titles tend to be cited more, and a pun rarely matches the words people actually search.
4. Test your keywords in Google Scholar before you commit. If a term returns ten thousand results, you are competing with the entire field; a more specific phrase puts you in front of the readers who need exactly your work.

Our [guide to writing an abstract that gets read](/blog/how-to-write-a-research-abstract-that-gets-read/) goes deeper on structuring that section so it satisfies reviewers and surfaces in search at the same time.

## Choose keywords that widen your reach

Keywords are a free, underused channel. Most journals let you list five to eight, and most authors waste them by repeating words already in the title. Those words are indexed regardless, so the slots are better spent on terms that broaden discovery.

A good keyword set mixes three layers: one or two broad terms that place your work in a field, two or three specific terms that describe your exact method or finding, and a couple of synonyms or abbreviations a reader might type instead of your preferred wording. If your subfield calls the same idea by two names, include both. If there is a common acronym, list the acronym and the spelled-out form. The goal is to catch the searcher no matter which phrasing they reach for.

When you are drafting and revising in a [manuscript editor like Alfred Scholar's](https://alfredscholar.com), it helps to keep a running list of the terms you would search for if you were hunting for this paper, then reconcile that list against your title and abstract before submission. The words you use to describe your own work and the words your readers use are not always the same.

## Make the paper free to find: open access and preprints

Open access is the discoverability lever with the strongest citation evidence behind it, and it does not require paying a large article processing charge. Green open access, where you deposit an accepted manuscript in a repository, is free and legal under most journal policies, and it puts a readable copy in front of anyone outside a paywall.

A preprint is a complete version of your paper shared in a public repository before peer review is finished. Posting one (where your field and journal permit it) does two things for citations: it makes your work findable months earlier, and it is associated with higher attention and citation counts once the peer-reviewed article appears. The earlier a relevant paper can cite you, the more of the citation window you capture.

If you are weighing the routes, costs, and policies, our [open access publishing guide](/blog/open-access-publishing-guide-for-researchers/) breaks down gold, green, and diamond OA with real numbers. The short version: at minimum, get a free full-text copy into an indexed repository, even if you publish behind a paywall.

## Build a scholarly identity that compounds

Citations are easier to attribute, and easier to find, when your identity is consistent across everything you publish. This is the least glamorous tactic and one of the most durable.

Register an ORCID and attach it to every submission. It disambiguates you from every other researcher who shares your surname and initials, which matters more than you would think; indexing errors and split author records quietly cost citations because the citing author cannot reliably link to your full body of work. Use one consistent form of your name on every paper. Claim and maintain a Google Scholar profile so your papers cluster under one author and new readers can see your related work in one place.

These identifiers create a flywheel. A reader who finds one of your papers and wants more can follow the trail to the rest, and each paper makes the others more discoverable. A messy, fragmented record breaks that loop.

## How do you reach the people who actually cite?

The people who cite you are other researchers writing papers in your area, and reaching them is different from reaching a general audience. Conference talks and posters work because they put your work in front of exactly that crowd, often before publication, which seeds early citations and collaborations. Field-specific channels, a relevant mailing list, a specialist newsletter, a working group, tend to outperform broad social media because the audience is composed of potential citers rather than passive scrollers.

This does not mean social media is useless. It is a fine top of funnel for attention, and attention occasionally converts. Just calibrate your effort to the evidence: spend the most time on the channels that reach citing researchers, and treat a viral post as a bonus rather than a strategy. Collaboration also helps in a concrete way, since co-authored papers tend to be cited more, partly because each author brings their own network of potential citers.

One more underrated tactic: cite well yourself. A paper with a thorough, fair literature review signals to reviewers and readers that you understand the conversation, and the authors you cite are more likely to encounter and cite your work in return. Our piece on [writing a literature review that proves novelty](/blog/how-to-write-a-literature-review-that-proves-novelty/) covers how to position your contribution against prior work without padding the reference list.

## How long until the citations come?

Patience is part of the strategy. Most papers do not collect their first citations for 12 to 24 months, because the papers that will cite you have to be conceived, written, reviewed, and published first. Citation accumulation usually peaks two to four years after publication and then tapers, though review articles and methods papers often have a much longer tail.

The practical implication is to stop checking your citation count weekly and to judge a paper's reach on a multi-year horizon. It also means the discoverability groundwork you lay at publication, the title, the keywords, the free full text, the clean author record, keeps paying off long after launch week. A paper that is easy to find in year three is still being found in year three. A clever launch that nobody can locate later is not.

## A realistic checklist to increase citations

Pulling it together, here is the order of operations that actually moves the number:

1. Write a short, descriptive title with your main keyword in the first 65 characters.
2. Front-load that keyword in the abstract and use all five to eight keyword slots on terms that widen reach.
3. Make a free full-text version available through open access or a repository.
4. Post a preprint if your field and journal allow it.
5. Register an ORCID, use a consistent name, and maintain a Google Scholar profile.
6. Present at conferences and share through field-specific channels, not just social media.
7. Write a fair, thorough literature review and keep your own citation metadata clean.
8. Give it two to four years before judging the result.

None of this is gaming the system. Every step makes genuinely good work easier to find and use, which is what citation counts are supposed to measure in the first place. Keeping your papers, references, and reading in one organized library, the way a [research workspace like Alfred Scholar](https://alfredscholar.com) is built to do, makes most of this checklist a byproduct of how you already work rather than a separate chore.