Scientific Posters: Designing a Figure-First Research Narrative
If you are a life sciences researcher preparing a conference poster, this guide is for you. The most effective posters lead with figures, not text. By designing a figure-first narrative, you help viewers grasp your key findings in seconds. I will explain how to structure your poster around visual hierarchy, choose and annotate figures, write readable captions, ensure accessibility, and plan your discussion. This approach draws on established principles of visual communication and resources from the NIH Office of Intramural Training and Education [1], which offers official guidance for scientific presentations.
Unlike a manuscript, a poster is a visual conversation starter. Your audience will scan the layout before reading any words. Therefore, your figures must tell the story independently. The rest of the poster supports them. This guide will walk you through the critical decisions for a figure-first poster, from initial planning to your presentation. Whether you are presenting clinical data as in the nationwide analysis of hospital discharge data for inflammatory rheumatic diseases [5] or a public health dissemination project [10], the principles remain the same.
At a Glance
| Element | Primary Goal |
|---|---|
| Visual hierarchy | Direct the viewer from the main figure to supporting figures in a logical order |
| Figure selection | Choose one central figure that captures your core result, then up to three supporting figures |
| Readability | Use sans serif fonts at least 24 point for body text, 36 point for captions, and 72 point for your title |
| Captions | Write self contained explanations that do not require reading the text panel |
| Accessibility | Choose colorblind safe palettes, add alt text for digital versions, and maintain high contrast |
| Discussion planning | Prepare three key takeaway statements that connect directly to your figures |
Why a Figure First Approach?
A figure first poster prioritizes your visual results. The typical viewer spends three to five seconds scanning your poster before deciding to stop. If they see a wall of text, they will walk away. By contrast, a strong figure catches attention and invites closer inspection. This approach matches how scientists actually communicate at conferences: conversations happen around figures, not around paragraphs.
The NIH Office of Intramural Training and Education provides explicit advice on poster design that emphasizes clear visuals and minimal text [1]. Following this guidance increases the chance that attendees will engage with your work. In my experience, a figure first poster also makes it easier to present your findings in a logical spoken narrative. You can point to each figure and explain the pattern without memorizing text.
Decision Criteria for Figure Selection
Choosing the right figures is the most critical design decision. Your central figure must be your strongest result. It should be interpretable within ten seconds. Consider these criteria:
- Importance. Does the figure convey your most novel or important finding? Prioritize a graph that shows your main effect or outcome.
- Simplicity. Avoid figures with more than three variables or multiple subpanels unless absolutely necessary. A clean scatterplot or bar chart often works better than a complex heatmap.
- Story fit. Will this figure make sense to someone from a related field? If you need a whole text block to explain it, choose a different figure.
- Data integrity. Use figures that represent your actual data, not idealized simulations. The figure you select should be reproducible from your analysis code.
For example, a study on hospitalisation patterns in rheumatic diseases might lead with a map showing regional variation [5]. A study on implementation strategies might lead with a flow diagram of the realist synthesis process [6]. The figure should preview the conclusion.
Practical Workflow for a Figure First Poster
Follow this sequence to build your poster from the figures outward.
Step 1: Draft your central figure first
Create or select the one graphic that best summarizes your main result. This figure will occupy the largest visual space on your poster, usually at eye level in the center or upper left quadrant. Ensure it has a clear title or label so even a distant viewer knows what it shows.
Step 2: Choose up to three supporting figures
Supporting figures should address secondary questions, methods validation, or subgroup analyses. Place them in a logical reading order: left to right, top to bottom. If you have more than four figures total, consider a handout or digital supplement. Cite your data management plan from the NIH Data Management and Sharing Policy [4] to show that full results are available.
Step 3: Write captions that work without context
Each figure caption should be a complete paragraph. It should state the takeaway, define the axes or labels, and note the sample size and statistical test. Do not use a title like "Figure 1: Results." Instead write "Figure 1. The treatment group showed a 30% reduction in symptom score compared to placebo (p=0.002, t test, n=45 per group)." The viewer can then discuss the figure without reading your methods section.
Step 4: Add text panels last
Your introduction and methods should be extremely brief, no more than two short paragraphs each. Use bullet points. The conclusion should be two to three sentences maximum. The text exists only to support the figures, not to repeat them.
Step 5: Design for accessibility and readability
Use sans serif fonts such as Arial or Helvetica. Body text should be at least 24 point, captions at least 32 point. Choose colorblind safe palettes like those from ColorBrewer. If you plan to share a digital version of your poster, include alt text for each figure. Your ORCID profile [3] is a good place to link to a permanent copy of your poster.
Step 6: Plan your discussion around the figures
Prepare a one minute overview that walks through each figure. For each figure, prepare one sentence that describes what it shows and one sentence that explains why it matters. Anticipate questions about outliers, error bars, and data limitations. Practice pointing to specific parts of the figure as you speak.
Common Mistakes and How to Avoid Them
The most frequent error is trying to cram every result into the poster. A figure first poster should not include more than four figures. If you have more data, create a supplementary QR code that links to a preprint or a data repository. The BLS career outlook resources [2] are not directly about posters, but they emphasize clear communication as a key professional skill, and that applies here.
Another mistake is using small or low resolution images. Export figures at 300 dpi. Check that text inside figures is at least 12 point. Never paste a screenshot of analysis output into the poster. The figure must be clean and labeled.
A third mistake is ignoring colorblind accessibility. Many scientists use red green color schemes. Replace red and green with blue and orange or use patterns and shapes. This improves comprehension for all viewers.
Finally, do not make the poster a miniature manuscript. Do not include long introductions or detailed methods. If someone wants those details, they will ask or look at your paper.
Limits and Uncertainty
No single poster design works for every audience. Conference halls have different lighting and viewing distances. A figure first poster that works at a small specialized meeting might fail in a giant poster session where people view from five meters away. Test your poster by printing a small version or showing it on a screen to colleagues. Ask them what they see from across the room.
The figure first approach assumes your results are visual. For fields that rely heavily on tables or qualitative data, you may need to adapt. For example, a course based undergraduate research experience highlighted in a recent study [8] might include a figure of a laboratory workflow but also a table of student learning outcomes. In those cases, treat a well formatted table as a figure.
Also note that a figure first poster does not guarantee high engagement. Your ability to greet viewers and explain the poster matters more than the layout. Prepare a friendly introduction that directs attention to your central figure.
Frequently Asked Questions
How many figures should a typical scientific poster include?
Three to four figures is standard for a poster that will be printed on a 48 by 36 inch board. One central figure and two to three supporting figures is ideal. If you have more figures, create a digital supplement and add a QR code.
Should I include a data availability statement on the poster?
Yes, especially given the NIH Data Management and Sharing Policy [4]. Add a small line at the bottom of the poster with a repository link or a statement that data are available upon request. This shows transparency and helps other researchers.
What font size should I use for figure captions?
At least 32 point for the caption text, and at least 24 point for any labels inside the figure. Your title should be at least 72 point. Remember that the poster will be viewed from two to three feet away, so everything must be larger than you think.
Can I use a template from my university for a figure first poster?
Many templates still default to text heavy layouts. You can adapt them by deleting large text boxes and expanding your figure area. Alternatively, start from a blank slide and place the central figure first. The NIH Office of Intramural Training and Education provides a sample layout [1] that can serve as a model.
References and Further Reading
- NIH Office of Intramural Training and Education: Poster Presentation Tips
- U.S. Bureau of Labor Statistics: Occupational Outlook Handbook
- ORCID: Persistent Researcher Identifiers
- NIH Data Management and Sharing Policy
- Changing Patterns of Hospitalisation and Orthopaedic Procedure Profiles in Major Inflammatory Rheumatic Diseases in Germany
- Exploring Change Mechanisms of Implementation Strategies in Inpatient Settings
- Design and Feasibility Trial of Interventions to Reduce Young Adult Alcohol Use
- Course Based Undergraduate Research Experience in HPV Associated Cervical Cancer
- Medical Claim Costs of Facioscapulohumeral Muscular Dystrophy
- Multi Level Community Dissemination of Public Health Research in Kenya and Malawi
Related Articles
- Data Science Careers in Biology: Building a Transferable Evidence Portfolio
- Bioinformatics Resume: Presenting Technical Depth Without a Tool Dump
- Bioinformatics Interview Preparation: How to Explain Your Analysis Decisions
- Research Statement for Bioinformatics Applications
- Personal Statements for Biology PhD Programs: Evidence, Fit, and Intellectual Direction