Scientific Poster: How to Make One People Stop and Read

Poster sessions are run by meetings, not journals, and the rules change between them. Board dimensions, landscape or portrait orientation, whether the meeting prints for you, whether the session is digital, and how your abstract number must appear all vary. A poster designed at the wrong shape cannot be fixed by scaling.

Before you open any software, find these six things on the meeting's presenter page:

• Board size and orientation. Set your document to that size on day one.
• Print or digital. A digital e-poster may have a fixed screen ratio, a file-size limit and sometimes a narrated version.
• Session date and length. This tells you how many conversations you are preparing for.
• Abstract or poster number placement. Some meetings ask for it in a corner so visitors can find you from the program.
• Embargo or press rules. Some meetings restrict what you or reporters can share before a set time.
• Mounting. Pins, tape or Velcro decides whether a fabric or paper print is practical to carry.

Searching for a "standard poster size" will give you numbers. Your meeting's number is the only one that matters.

Erren and Bourne's second rule is "sell your work in ten seconds." Their suggestion is to pose your work as a decisive question and then answer it. We would push one step further: write the answer as a single sentence before you design anything.

A usable claim sentence names the finding and the evidence type: "Loss of gene X slows wound closure in mouse skin by about half, and restoring X rescues it." A weak one names a topic: "The role of gene X in wound healing." The test is whether a visitor could repeat your sentence to a colleague after walking away.

That sentence becomes three things:

1. Your title. State the finding, not the topic. Erren and Bourne call the title "your equivalent of a newspaper headline."
2. Your conclusion panel, in plain words, placed where the eye lands after the title.
3. The filter for everything else. If a panel, a method detail or a paragraph does not help a reader accept that sentence, it comes off the poster and goes into a handout.

If you already wrote the meeting abstract, start from its last sentence. Our conference abstract guide covers how to make that sentence specific enough to carry a poster.

A poster figure is not a paper figure printed larger. On paper the reader is 40 centimetres away and can study a six-panel figure. At a poster session they stand a few steps back and give each figure a few seconds.

Erren and Bourne suggest letting a figure be read "in both a superficial and a detailed way": a bold trend line with its interpretation stated, plus the individual points with error bars for the visitor who wants them. That is the right standard. The general rules for readable figures are on our scientific figure rules page.

A paper is linear because pages follow each other. A poster is not, and visitors wander. Give them one obvious path.

Columns. Two to four columns read top to bottom, then left to right, are the easiest path for most visitors. Number the panels if the order is not obvious.

Where things go.

White space. Erren and Bourne quote Keith Richards on painting: a good painter "will never cover all the space." Space between panels is what makes the path visible. If the poster looks full, it is too full.

Type size. Erren and Bourne's rule is plain: "never use less than a size 24 point font, and make sure the main points can be read at eye level." Put the claim and conclusion at eye level, not at knee height in the bottom row.

Two tests catch most problems, and both are free.

The one-page test. Print the whole poster on a single A4 or letter sheet. A4 is a quarter of A0 along each side, so an A4 printout of an A0 poster is the poster at one quarter scale. If you can read every word at arm's length, a visitor can read the full-size print from about four times that distance. If you squint, so will they.

The cold reader test. Give the one-page print to a colleague outside your project for one minute. Then ask what the finding was. If they cannot say your claim sentence back, the layout or the title is hiding it.

Erren and Bourne's fourth rule is "poster acceptance means nothing": a poster has been accepted, not endorsed. The endorsement comes from the conversations.

Prepare three versions of your talk:

• Ten seconds. The claim sentence, said out loud. Most visitors decide here whether to stay.
• Two minutes. Walk the figures in order: the question, the key result, the control that makes it believable, what it means.
• The long one. For the visitor who works on the same problem. This is where your best feedback comes from.

Bring printed copies of the poster on standard paper or a QR code to a handout, and a way for visitors to leave contact details. Erren and Bourne also note the point many presenters miss: the session "is likely your opportunity to get feedback on the work before it goes to publication. Better to be tripped up in front of your poster than by a reviewer of the manuscript."

Usually not, and the major policies say so directly. This is the question the poster guides tend to skip, and it is the one that matters if the work is not yet published.

Two cautions. First, these are the ICMJE and Nature positions; your target journal may have its own policy, so read it before the meeting if the result is sensitive. Second, ICMJE also reminds authors that dissemination outside scientific presentations "may diminish the priority journal editors assign to their work." A poster is fine. A press release about the poster is a different decision.

A poster session is the cheapest peer review you will get. Most presenters waste it by not writing anything down. Bring a notebook and record each question exactly as asked, then sort the list the evening after the session.

In our pre-submission review work, the gap we check every manuscript for first is a claim stronger than its evidence, and posters make that gap easy to create. On a poster, a confident title over one representative image passes in a crowded hall. In a manuscript, the first referee asks for the quantification, the second control or the replicate the poster never showed. If visitors kept asking "how many?" or "compared with what?", that is the referee's first question, already in your notebook.

Three other poster-to-paper problems are worth checking for:

• The poster figure becomes the paper figure. A figure simplified for distance usually drops the n, the individual points or the statistics a journal needs. Rebuild from the full data, not from the poster file.
• The claim drifted. Months later, the manuscript's abstract claims more, or less, than the poster did. Decide which version the data support.
• New data never reconciled. Results added after the meeting change a number that the poster's conclusion was built on, and the text still uses the old one.

When the poster work becomes a manuscript, our AI engine (trained by 35+ current top-tier-journal reviewers) can read it before a journal does. The free manuscript readiness scan takes a few minutes and shows the top submission risk. For $39, the Full Review goes further: it analyzes each section, verifies citations against live scholarly sources, comments on figures, methods and reviewer risk, and ends with a prioritized revision plan. It reviews manuscripts, not poster files.

• A topic for a title. "Studies of X in Y" tells a passer-by nothing. Put the finding in the title.
• The abstract pasted in full. It repeats what the poster should show, in the smallest text on the board.
• Every panel from the paper. Eight figures at postage-stamp size prove less than two at full width.
• Methods as a paragraph. A schematic with three labelled steps is read; a 150-word paragraph is not.
• The conclusion in the bottom right corner at knee height. Visitors read the title, then look for the answer. Put it where they look.
• Colour that carries the result alone. A red versus green comparison disappears for readers with colour-vision deficiency and in greyscale handouts.
• No way to follow up. No email, no QR code, no handout, and the best conversation of the session ends there.