How to make an academic research poster

What size should an academic poster be?

Use exactly the size and orientation in the conference’s poster instructions. If the call says “A0 portrait”, a landscape poster may not fit the board. If it gives board dimensions instead, leave a margin of a few centimetres on every side for pins or tape.

Set your file to the final size from the start. The University of Birmingham’s print service puts it plainly: if you design at A4 and ask for it to be enlarged to A0, you may be disappointed with the print quality. Small images that looked sharp on screen become blurred when scaled up four times.

How should you lay out a research poster?

Use a grid. For a portrait A0 or A1 poster, three columns is the usual choice, with a full-width title band at the top. A landscape poster can take three or four columns. Readers start at the top left and finish at the bottom right, so arrange the sections in that order:

1. Title band: title, authors, affiliations, and institution or conference logos only if they are required.
2. Column 1: background (three or four sentences), the gap, and the research question or objective.
3. Column 2: method as a diagram, then the first result.
4. Column 3: the main result, the comparison with earlier work, and the conclusion in three short points.
5. Footer strip: key references (five or six), acknowledgements and funding, and a QR code to the paper or your email.

Leave space. NTU Library’s guide suggests a rough balance of 20% text, 40% visuals and 40% white space. That will feel empty when you design it and right when it hangs in a crowded hall. Align every box to the grid, keep the gaps between boxes equal, and number the section headings so the reading order is obvious.

Visual hierarchy

A visitor should see, in this order: the title, the main result figure, the section headings, and only then the body text. Make the main result the largest figure on the poster, and put a one-line takeaway above it in heading-size type (“Recall rose from 0.71 to 0.86 on the held-out set”). If someone reads only the title and that line, they should know what you found.

How many words should a poster have, and what font sizes work?

Fewer words than you think. The word budgets in the size table above are for the whole poster, including headings and captions. A methods section that took two pages in the paper gets perhaps 60 words and a diagram on the poster.

NTU Library’s legibility chart gives the distances: about 30-point text can be read from 1.8 metres, 48 points from 3 metres and 72 points from 4.2 metres. The 2014 IEEE Nuclear Science Symposium guidelines warned that text below 18 points is difficult for the audience to read. The ranges below follow from those figures.

• Use one or two typefaces. A clean sans serif (Arial, Helvetica, Calibri, Source Sans) for headings and body is easiest at a distance. Never more than three fonts on one poster.
• Left-align body text. Justified text leaves uneven gaps at large sizes.
• Avoid all capitals except for very short labels. Mixed case is faster to read.
• Write in fragments where you can. Bullets of one line each are easier to read standing up than paragraphs.

Which tool should you use, and how do you get the poster printed?

PowerPoint

The most common choice. Set the slide size to the poster dimensions (Design, then Slide Size, then Custom) before adding anything, turn on the grid and guides, and use tables and SmartArt sparingly. Insert figures as high-resolution images or vector files, not screenshots.

Canva

Quick for a good-looking layout if you are not comfortable with design. Start from a blank custom size, not a marketing template, and check that equations and figure labels stay sharp. Download as a print-quality PDF.

LaTeX with beamerposter

The beamerposter package on CTAN extends the beamer class to A-series poster sizes, with font scaling. It suits scholars who already write in LaTeX and have many equations. It takes longer to adjust the layout, so start a week earlier than you would with PowerPoint.

Printing

• Send a PDF with all fonts and images embedded. Open it on another computer to check nothing has shifted.
• Check images at 100% zoom. If a figure looks soft on screen at full size, it will look soft on paper.
• Proof a reduced copy first. Print the whole poster on A4 or A3 and read it with your supervisor. A typo found on an A4 proof costs a few rupees; one found on the printed A0 costs a reprint.
• Allow time. Campus print shops and local printers near most universities can print A0 on gloss, matte or fabric, but give yourself two or three working days before you travel.
• Think about travel. A paper poster needs a tube and can crease; a fabric poster folds into a bag and suits train and air travel.

What should you say at the poster session?

Stand beside the poster, not in front of it, for the whole session. Some IEEE conferences count a poster as not presented if the author is absent for most of the session; the IEEE SAM 2026 no-show policy says so explicitly. Prepare a two-minute pitch you can give to anyone who stops:

1. 1

   The problem (15 seconds)

   One sentence on the problem and why it matters, pointing at the picture that shows it.

2. 2

   What you did (30 seconds)

   The method in plain words, pointing at the method diagram. No detail the visitor didn’t ask for.

3. 3

   What you found (45 seconds)

   The main result, pointing at the one figure that shows it, with the number that matters.

4. 4

   Why it matters (20 seconds)

   What the result changes, and its main limitation, in two sentences.

5. 5

   Hand over (10 seconds)

   “What’s your own work on?” or “Would you like the paper?” Then listen.

Also prepare a 30-second version for people who are only passing, and a longer one for specialists who ask about the method. Keep a few A4 copies of the poster or the paper, or a QR code, for people who want to read more. Write down the names and questions of the people you speak to; poster sessions are where many useful research contacts are made.

What do poster judges look for?

Many conferences and university research days award best poster prizes, and the published judging rubrics cover much the same ground. Washington State University’s undergraduate research showcase, for example, judges poster design, presentation delivery, background and objectives, research methods and results. In practice, judges ask:

• Is the question clear within the first few seconds of looking?
• Does the method fit the question, and is it described well enough to judge?
• Do the results support the conclusion, with the key number visible?
• Is the poster readable: clear hierarchy, readable fonts, figures that make sense without the paper?
• Can the presenter explain it, answer questions about the method and admit the limits?

The last point decides more close contests than design does. A judge who asks “why this sample size?” and gets a clear, honest answer marks the work up. Whether the proceedings containing your poster abstract count for your PhD is a separate question; conference paper or journal paper? answers that.