How Do I Write Good Alt Text for Educational Images?

Good alt text for educational images describes what the image communicates in context, not just what it looks like. For a chart, describe the data trend. For a diagram, describe the relationships. For a photograph, describe what is relevant to the learning objective. The goal is to give a student who cannot see the image the same information a sighted student gets from it.

This sounds simple, but it trips up even experienced faculty — especially when the image is a complex graph, a chemical structure, or an anatomical diagram. This guide walks through every common type of educational image with concrete examples of what to write and what to avoid.

The Golden Rule: Describe Function, Not Appearance

The single most important principle of alt text is this: describe what the image does in your document, not what it looks like.

A sighted student looking at a bar chart does not think "I see blue bars of varying heights." They think "enrollment grew from 2,400 to 3,100 over three years." Your alt text should convey the same takeaway.

Bad: "Bar chart with blue bars"

Good: "Bar chart showing undergraduate enrollment growth from 2,400 in 2023 to 3,100 in 2026"

The bad example tells a screen reader user that a bar chart exists. The good example tells them what the bar chart means. That distinction is the entire difference between usable and useless alt text.

Simple Images: Photos and Illustrations

For straightforward images — a photo of a campus building, an illustration of a historical figure, a picture of lab equipment — keep your alt text concise. One to two sentences is usually enough.

Ask yourself: why is this image here? The answer is your alt text.

Photo in a biology lecture:

• Bad: "Picture of a frog"
• Good: "Red-eyed tree frog gripping a branch, showing the adhesive toe pads discussed in this section"

Portrait in a history course:

• Bad: "Image of a person"
• Good: "Portrait of Ada Lovelace, circa 1840"

The context matters. If the same frog photo appeared in an art appreciation course, the alt text might focus on composition and colour rather than toe pads. Alt text is not a universal label — it serves the specific learning context.

Charts and Graphs

Charts are where most faculty struggle, because the data is the point. Your alt text needs to convey the trend, the key numbers, and the takeaway.

Line graph in an economics course:

• Bad: "Graph"
• Good: "Line graph showing inflation rate rising from 2.1% in January 2024 to 6.8% in December 2025, with a sharp increase beginning in March 2025"

Pie chart in a survey methods class:

• Bad: "Pie chart of responses"
• Good: "Pie chart of student survey responses: 45% strongly agree, 30% agree, 15% neutral, 7% disagree, 3% strongly disagree"

Include the actual numbers when they matter for understanding. If students are expected to read specific values off the chart, those values must appear in the alt text. If the chart illustrates a general trend, focus on the trend.

For charts with dense data — dozens of data points, multiple series — a short alt text cannot capture everything. In those cases, summarise the key takeaway in the alt text and provide the full data in an adjacent table or long description (more on this below).

Diagrams and Flowcharts

Diagrams communicate relationships, processes, and hierarchies. Your alt text should describe those relationships, not the shapes on screen.

Flowchart in a research methods course:

• Bad: "Flowchart with boxes and arrows"
• Good: "Flowchart showing the peer review process: author submits manuscript, editor screens for fit, two reviewers evaluate independently, editor makes accept/revise/reject decision, author revises and resubmits if required"

Organisational diagram:

• Bad: "Org chart"
• Good: "Organisational chart showing the Dean at the top, three Associate Deans reporting to the Dean, and five department heads reporting to the Associate Dean for Academic Affairs"

For complex diagrams with many nodes and connections, alt text alone may not be sufficient. Consider placing a text description directly below the diagram in the document body, or using a structured list that walks through the relationships. This is more effective than trying to cram an entire system diagram into an alt attribute.

STEM Images: The Hard Cases

Chemical structures, mathematical notation, anatomical diagrams, circuit schematics, and engineering drawings are the hardest images to describe — and the most important to get right, because students cannot learn the material without them.

Chemical structure in organic chemistry:

• Bad: "Molecule diagram"
• Good: "Structural formula of aspirin (acetylsalicylic acid) showing a benzene ring with a carboxyl group and an ester group at adjacent positions"

Anatomical diagram in a physiology course:

• Bad: "Diagram of the heart"
• Good: "Cross-section of the human heart showing the four chambers: right atrium receiving deoxygenated blood from the vena cava, right ventricle pumping to the lungs via the pulmonary artery, left atrium receiving oxygenated blood from the pulmonary veins, and left ventricle pumping to the body via the aorta"

Mathematical equation: For equations, alt text should express the equation in words or a readable notation. For example, the quadratic formula image might use: "x equals negative b plus or minus the square root of b squared minus 4ac, all divided by 2a." Better still, use MathML or LaTeX markup instead of images whenever possible — these are natively accessible to screen readers without any alt text at all.

The guiding question for STEM images is always: what does a student need to learn from this image? Describe that, even if it takes a full paragraph.

Maps

Maps communicate geographic information, spatial relationships, or data distributions. Describe what the map shows, not its visual design.

Map in a political science course:

• Bad: "Map of Australia"
• Good: "Map of Australia showing 2025 federal election results by division: Labor holds most urban seats along the east coast, the Coalition dominates rural and regional divisions, and the Greens hold inner-city seats in Melbourne and Brisbane"

Campus map in an orientation guide:

• Bad: "Campus map"
• Good: "Campus map highlighting the route from the main library to the science building: exit the library heading north, turn right on University Avenue, and the science building is the third building on the left"

Decorative Images

Not every image needs alt text. Decorative images — borders, spacers, background patterns, stock photos used purely for visual interest — should be marked as decorative so screen readers skip them entirely.

In a PDF, this means marking the image as an artifact. In HTML, it means using an empty alt attribute (alt=""). In Word or PowerPoint, it means checking "Mark as decorative" in the alt text dialog.

The test is simple: if you removed this image, would the reader miss any information? If not, it is decorative. A stock photo of students studying that accompanies a paragraph about study tips is decorative — the paragraph says everything. A photo of a specific laboratory setup that students need to replicate is not decorative — it carries essential information.

Screenshots

When your course materials include screenshots — of software interfaces, websites, or application outputs — describe the relevant content, not every pixel.

Screenshot in a data science course:

• Bad: "Screenshot of RStudio"
• Good: "RStudio console output showing the result of the linear regression: R-squared value of 0.847, p-value less than 0.001, indicating a statistically significant fit"

Focus on what the student is supposed to learn from the screenshot, not a comprehensive description of the entire interface.

Common Mistakes to Avoid

Starting with "image of" or "picture of." Screen readers already announce that an element is an image. Writing "Image of a bar chart showing..." is redundant. Just write "Bar chart showing..."

Writing too little. "Graph" or "diagram" as alt text is nearly useless. It tells the student an image exists but nothing about what it contains.

Writing too much. A 500-word alt text is overwhelming through a screen reader because users cannot skim it the way they skim visual content. Be thorough but concise. For truly complex images, use a long description instead.

Using the filename. "IMG_4837.jpg" or "figure3_final_v2.png" as alt text helps nobody. This happens more often than you might think, especially in auto-generated content.

When Alt Text Is Not Enough: Long Descriptions

Some images carry too much information for a brief alt text. Data-heavy charts, detailed infographics, complex system diagrams, and annotated photographs may need a long description — a full text alternative that provides everything a sighted user would get from studying the image.

Options for providing long descriptions:

1. Adjacent text. Place a detailed description immediately before or after the image in the document body. This is the simplest approach and benefits all students, not just those using assistive technology.
2. A linked document. For extremely complex visuals, link to a separate page or appendix containing the full description.
3. An HTML aria-describedby reference. In web content, you can point to a longer description elsewhere on the page using the aria-describedby attribute.

The alt text for the image itself should still provide a brief summary and indicate that a longer description is available nearby.

Putting It Into Practice

Start with your highest-enrolment courses. Review the images in those materials and ask, for each one: if a student could not see this, what would they need to know? Write that down. That is your alt text.

You do not have to do this perfectly on the first pass. Reasonable alt text that conveys the key information is far better than no alt text at all. You can refine it over time as you get more comfortable.

For a complete walkthrough of making your documents accessible — including alt text, tags, reading order, and more — see our guide on how to make a PDF accessible. To check whether your current documents meet the standard, our PDF accessibility checker guide walks through the verification process.

Writing alt text for hundreds of images across an entire course catalogue is a significant undertaking. Aelira's AI generates context-aware alt text descriptions and flags uncertain ones for your review — see how it works. And if you are wondering whether AI-generated alt text is reliable enough, our analysis of whether AI accessibility fixes can be trusted addresses that question directly.