Visual Design Systems for Badges

Badge images that carry meaning consistently across a portfolio.

This workshop course (ICB-P04) teaches you to design the image side of a badge portfolio as a system, not as a series of one-off pictures. You practise on realistic tasks in a controlled setting: you build a visual taxonomy, apply accessibility rules, write an image brief and review an existing badge set.

Who it is for

• L&D and corporate academy leads who run badge programmes for employees, partners or customers.
• Learning designers and teaching staff in universities, colleges and training providers.
• Professional bodies and sector associations that issue badges to members.

Prerequisite: ICB-F01 Digital Badges and Credentials Essentials.

Framework alignment: ICoBC Credentialing Skills Framework (ICSF), skill CRDS "Credential design" (VRITE pillar: Rigor), level L2 "Operate", stage A "Ability". At this level you draft credential specifications from templates, including outcomes, criteria and required metadata. This course applies that skill to the visual part of the specification.

Learning outcomes

After this course you can:

1. Build a visual system that shows type, level and domain consistently.
2. Apply accessibility rules to badge images (contrast, text, alternatives).
3. Write an image brief for designers.
4. Review an existing badge set for consistency and meaning.

Workload: 4 hours notional learning in a workshop format. This file covers the self-paced part. Live or workshop elements appear as facilitator notes.

Quality focus: VRITE Experience. A badge image is part of what the earner and the viewer experience. It should be clear, fair and usable by everyone.

How it is assessed: You produce a visual taxonomy sheet and three badge drafts or design briefs. A person assesses them with the rubric in "Assessment and badge". A final knowledge check of 10 questions is also included. Nothing is graded inside this file.

Badge: "Visual Design Systems for Badges (Ability, L2)", alignment code ICSF-CRDS-L2-A.

What images can and cannot carry

The job of a badge image

A badge image is small. It appears as a thumbnail in an email, on a profile page, in a wallet and on a certificate wall. Often the viewer sees it at a size of 64 to 128 pixels. At that size the image has a limited job. It helps people recognise the badge, tells them roughly what kind of badge it is and gives the earner something to be proud of.

The image is the front door. The real information sits behind it in the badge metadata: the issuer, the criteria, the evidence, the date and the outcomes. A viewer who wants to know what the badge means must be able to reach that information. The image supports this, but it cannot replace it.

A badge image supports recognition. The meaning of a badge sits in its metadata and criteria.

What an image can carry well

Images can carry a small number of signals very quickly:

• Who issued it. A logo or a consistent frame points to the issuer.
• What type it is. For example a participation badge, a skill badge or a pathway badge.
• How advanced it is. A level, shown by a number, a count of marks or a shape.
• Which domain it belongs to. For example data, leadership or safety.
• A short title. One to three words, if legible.

Each signal needs a code that people can learn. The best systems use few signals, each with a clear rule.

What an image cannot carry

An image cannot show the criteria, the assessment method, the evidence or the standard of quality. It cannot prove that the earner did the work. It cannot show that the issuer is trustworthy. A very elaborate image can even mislead: it may look more important than the badge really is.

Be careful with these cases:

• Overloaded images. Eight signals in one icon leave the viewer with none.
• Borrowed prestige. Seals, crests, laurel wreaths and medals suggest a formal qualification. Use them only when the badge has the matching status.
• Hidden level. If one badge shows a level and another does not, viewers will compare them wrongly.
• Text that is too small. Words that cannot be read at thumbnail size carry nothing.

Image and metadata work as a pair

Think of the image and the metadata as two halves of one message. The image says "this is a level 2 data skill badge from our academy". The metadata says exactly what the earner did and how it was checked. When both halves agree, viewers trust the badge. When they disagree, for example a gold image on a badge with a light criterion, trust drops.

Many badge platforms store the image as a file linked to the badge data. In that case the visual system is a design decision made once and applied many times. This is why a system matters more than individual pictures.

Reviewing meaning in a set

Outcome 4 asks you to review an existing badge set for consistency and meaning. Start with two questions for every badge: "What does this image claim?" and "Does the metadata support that claim?" Then compare badges side by side. Do similar badges look similar? Do different types look different? Does a higher level look higher?

A simple review sheet has one row per badge and columns for type, level, domain, claim of the image, support in the metadata and a verdict (consistent, unclear or conflicting). You will use this sheet again in Module 4.

Worked example: The academy "gold" problem

Context 1: corporate academy. The learning and development lead of a logistics company's academy reviews her 12 internal badges. Three badges use a gold shield: "Safe Handling Basics", "Customer Call Handling" and "Team Leader Programme". The first two are short participation badges with a 30-minute quiz. The third needs 40 hours and a workplace assessment (all figures illustrative).

She asks the two questions. The gold shield suggests a high status. For the first two badges the metadata does not support this, so the verdict is "conflicting". She notes the issue: the shield colour does not encode anything, so viewers will treat all three as equal. Her fix is to reserve gold-style framing for the highest level and to show participation badges with a plain frame. She records the rule in one sentence and applies it to the whole set later.

Context 2: university. A similar check in a university shows the opposite problem: the continuing education unit gives each of its 15 short courses a different border, so a viewer cannot see which badges belong to the same programme. The verdict is "unclear", and the fix is a shared frame with a domain colour.

Practice activity: Claim check on a sample set

Take the sample set of six badge descriptions below (illustrative), or use six badges from your own organisation.

1. For each badge, write what the image claims in one short phrase.
2. Mark the metadata as supporting, partly supporting or not supporting the claim.
3. Give a verdict: consistent, unclear or conflicting.
4. Write one sentence on which pair of badges is most likely to be confused, and why.

Keep your notes. They form the first part of the review section in your assessed taxonomy sheet.

Check your understanding

When the image claims more than the metadata supports, trust drops and badges are compared wrongly. Colour and file size are not the issue here.

Criteria and evidence are detailed and need text. The image can carry short signals such as level, title and issuer cue.

A visual taxonomy

What a visual taxonomy is

A visual taxonomy is a short set of rules that links what a badge means to how it looks. It answers three questions for every badge:

1. Type: what kind of badge is this?
2. Level: how advanced is it?
3. Domain: which subject area does it belong to?

Each answer is shown by one visual variable. The rules are written down so that any designer can follow them and any viewer can learn them.

Keep the taxonomy small. A good rule of thumb is three to five types, three to five levels and four to eight domains. More categories are possible, but each one must be easy to tell apart at a small size.

Choose one variable per dimension

A visual variable is a property you can change on purpose: shape, colour, pattern, symbol, number of marks, position or size of an element. The key rule is to use one variable per dimension. If shape shows type, do not also use shape for level. This keeps the code readable.

A common and robust setup:

• Type is shown by the outer shape (for example circle, hexagon, rounded square).
• Level is shown by a number of marks or a numeral in a fixed position (for example one, two or three bars along the bottom).
• Domain is shown by a colour band plus a symbol in the centre.

Notice that domain uses colour and a symbol together. This is deliberate: colour alone fails for viewers who cannot tell colours apart, and it also fails in greyscale print. The symbol and the written domain name back up the colour.

Keep the meaning of colour stable

Choose a small palette and give each colour one job. If you use colour for domain, do not use the same colour as a level signal in another badge. Document the colour values in the taxonomy sheet, together with the contrast result (see Module 3). Check that domain colours remain distinguishable from each other.

Many sets also use a neutral base (for example white and navy) that does not carry any meaning. This holds the frame together and keeps the set recognisable as one family.

Fixed zones

Every badge in the system uses the same layout zones. For example:

• Outer frame: type shape and issuer cue.
• Centre: domain symbol.
• Lower band: level marks.
• Optional text zone: a short title, never more than three words.

Fixed zones make a set feel like a family. They also make briefing easy, because a designer only needs to fill the zones with the right values.

Writing the taxonomy sheet

The taxonomy sheet is the working document of your visual system. It is one to two pages and contains:

1. Purpose and scope: which badges the system covers.
2. Dimensions and variables: the three dimensions and the variable chosen for each.
3. Value tables: each type, level and domain with its visual value and a short meaning.
4. Zone layout: a simple drawing of the fixed zones.
5. Colour and text rules: palette with codes, minimum text size, font family.
6. Rules for exceptions: how to handle a badge that fits none of the values (usually: ask the system owner).
7. Do and do not examples: at least two of each.

This sheet is the first item of your assessed evidence. It is the same kind of template-based specification work that ICSF level L2 expects: you fill a structured document with clear outcomes and criteria rather than invent a format each time.

Worked example: Two taxonomies for two settings

Context 1: corporate academy or HR department. The HR department of a retail group (illustrative) issues badges for internal training. It chooses three types: Participation (circle), Skill (hexagon) and Pathway (rounded square). It chooses three levels shown by one, two or three bars. It chooses five domains: Customer, Operations, Leadership, Compliance and Digital. Each domain has its own colour band and symbol, and each symbol is stored in a shared icon set. A short rule states: "Compliance badges always show the level as a numeral as well as bars, because auditors read them." The HR lead records this exception in the sheet so that it is a rule, not a one-off.

Context 2: university, college or training provider. A college (illustrative) issues badges for a short course programme. It uses the same three types but names them Attendance, Skill and Programme. Its domain colours follow its faculties. Because many earners study in two faculties, the college adds a rule that a badge shows only one domain, the faculty that awards it. The level bars map to the national qualifications level used locally, and the sheet notes that the mapping is stated in the metadata.

Both taxonomies are small, legible and documented. Neither tries to encode the duration or the delivery mode in the image: that information stays in the metadata.

Practice activity: Draft your visual taxonomy

Build the first draft of your taxonomy sheet. You may use your own organisation or a fictional one.

1. List your types (three to five), levels (three to five) and domains (four to eight). Write a one-line meaning for each.
2. Pick one visual variable per dimension. State why you chose it.
3. Sketch the fixed zones on paper or in a drawing tool.
4. Choose the palette. Write the colour codes and note that each will be checked for contrast in Module 3.
5. Write three rules for exceptions.
6. Test it: write the code for two imaginary badges, for example "Skill, level 2, Digital" and "Pathway, level 3, Leadership". Can a colleague draw each from your sheet alone?

Check your understanding

Each dimension has its own variable, so the code stays readable. Reusing one variable for several dimensions mixes the signals.

Accessibility

Why accessibility matters for badges

A badge image reaches people who see it in many ways: on a small phone screen in sunlight, in greyscale print, with a screen reader, with low vision or with colour vision deficiency. A badge that excludes some earners or viewers fails the VRITE Experience focus of this course.

Accessibility is also a legal and standards matter in many contexts. The European Accessibility Act applies since 28 June 2025. WCAG 2.2 is the current W3C recommendation. EN 301 549 is the European standard for accessibility requirements for ICT products and services, which is used with the Act. Which rules apply to your badge programme depends on your organisation, your country and the products or services you offer. Ask your legal or compliance contact if you are unsure.

Status as of November 2026; check before publishing.

Contrast

Contrast is the difference in brightness between two colours. WCAG 2.2 sets minimum contrast ratios. Two thresholds matter for badge images:

• Text: a contrast ratio of at least 4.5 to 1 for normal text, and at least 3 to 1 for large text.
• Graphical objects: a contrast ratio of at least 3 to 1 for parts of graphics that are needed to understand the content.

Navy text (#02375B) on white (#FFFFFF) has a very high contrast. Navy text on orange (#F78222), cyan (#5DCAE7) or lime (#B1CF37) also works well. White text on those three accents does not reach the needed ratios, which is why the ICoBC style uses navy text on them. Always check your own colour pairs with a contrast checker rather than by eye.

Text in images

Text inside an image cannot be resized by the viewer and may not be read by assistive technology. Follow these rules:

1. Keep text in the image to a short title of one to three words, or none.
2. Use a plain, sturdy font at a size that stays readable at the smallest display size.
3. Keep text horizontal, with no text on curves.
4. Never put the only copy of important information in the image. Repeat it in text: the badge name, the issuer, the level and the criteria.

Do not rely on colour alone

Colour is a good extra signal and a poor only signal. Pair every colour meaning with a second cue: a symbol, a pattern, a label or a position. In the taxonomy of Module 2, the domain colour comes with a symbol and the level comes with bars, not with colour. Test your design in greyscale. If two domains look the same in greyscale and have no other cue, add one.

Alternative text

Every badge image needs an alternative text (alt text) that tells a screen reader user what the image is. A good alt text is short and states the useful meaning. It follows a pattern you can write once in the taxonomy sheet:

"[Issuer] [type] badge, [level], [domain]: [badge name]."

Example: "Northgate Academy skill badge, level 2, Digital: Data Visualisation Basics." (fictional names)

Do not start with "image of" or "badge showing". Do not describe decoration. Do put the same name and level that appear in the badge metadata. For fuller information the alt text can point to the criteria page, which is part of the badge data.

Other practical checks

• Size: test at the smallest planned size, for example 64 pixels wide.
• Motion: do not use flashing or fast animation.
• Background: test on light and dark backgrounds if the platform shows both.
• Print: test in greyscale.
• Metadata language: write the name, description and criteria in plain language so that screen reader users and others can use them.

Worked example: Fixing a badge that fails three checks

Context 1: corporate academy or HR department. An HR team has a badge "Customer Care Level 2": light grey text on a pale yellow background, the domain shown only by a yellow band, and the alt text "badge". A reviewer lists the failures. First, the text contrast is too low. Second, the domain relies on colour alone. Third, the alt text carries no information.

The team makes three changes. They set navy text on the pale background, which gives a high contrast. They add a heart-shaped symbol in the centre next to the yellow band, so the domain no longer depends on colour. They write the alt text with the pattern from the taxonomy: "Harbour Retail skill badge, level 2, Customer: Customer Care" (fictional). They also check the result in greyscale and at 64 pixels.

Context 2: university, college or training provider. A college has a badge image with a long course title in small letters curved around the edge. The text is unreadable at thumbnail size, and a screen reader cannot access it. The fix: shorten the title to two words placed horizontally in the text zone, and put the full title in the badge name and the alt text.

Practice activity: Accessibility audit and fix

Use the three badge descriptions below (illustrative). For each badge, complete the checks and write a fix.

1. Check contrast, text size, colour reliance and alt text for each badge.
2. Write one concrete fix for each failure.
3. Write the alt text pattern for your own taxonomy from Module 2 and apply it to two of your own badge codes.
4. Add a short "Accessibility rules" block to your taxonomy sheet with minimum text size, contrast targets and the alt text pattern.

Check your understanding

Navy on lime gives strong contrast. White, light grey and yellow on lime give weak contrast. Always confirm with a contrast checker.

Production and governance

From system to finished images

Once the taxonomy and accessibility rules exist, you can produce the images. Production has four steps:

1. Brief the designer or the tool.
2. Draft the images following the zones and values.
3. Check each draft against the taxonomy sheet and the accessibility rules.
4. Release the approved images with their alt text and metadata links.

Production is simple when the system is clear. It is slow and costly when each badge starts from a blank page.

The image brief

An image brief is a short document that tells a designer exactly what to make. It supports learning outcome 3. A good brief contains:

1. Badge name and issuer.
2. Purpose: one sentence on what the badge recognises.
3. Taxonomy values: type, level and domain, with the matching visual values.
4. Required text: the short title in the image, if any, with the exact wording.
5. Alt text: the final wording.
6. Palette and fonts: codes and sizes, or a pointer to the taxonomy sheet.
7. Technical specification: file formats, sizes, background rules, minimum display size.
8. Do not include: for example no logos of third parties, no seals suggesting accreditation, no photographs of people.
9. Rights and sources: who owns the final artwork, which licences apply to any element used.
10. Review and approval: who checks it and by when.

Write the brief so that the designer does not need to guess what the badge means. If the brief and the taxonomy sheet differ, the sheet wins.

Governance: keeping the set consistent

A visual system stays consistent only if someone owns it. Governance means a few simple rules:

• An owner: one named role is responsible for the taxonomy sheet.
• A change process: new values (a new domain, a new level) go through the owner. Each change is dated and recorded.
• A review rhythm: the set is reviewed at fixed intervals, for example once a year or when ten new badges have been added.
• A register: a simple list of all badges with their type, level, domain, image file and version.
• A retirement rule: old images are withdrawn or marked when a badge is retired or changed in meaning.
• A sign-off: each new badge image is checked against the sheet by a second person before release.

The review in outcome 4 uses the same method as in Module 1: compare each badge with the taxonomy, record consistent, unclear or conflicting, and propose a fix. In a mature system most findings are "consistent", and the governance process keeps it that way.

AI image tools: licensing, consistency and accessibility checks

Some teams use AI image tools to create badge artwork or icons. If you do, three questions need clear answers before release.

Licensing. Who owns the output, and under what terms may you use it? Terms differ between tools and can change. Check the terms of use for commercial and public use, whether the tool states rights over the output, and whether it requires attribution. Do not use prompts that copy a named artist, an existing logo or a protected symbol. If you cannot clarify the rights, do not publish the image. Ask your legal contact where needed.

Consistency. AI tools often produce a slightly different style each time. A badge set needs the same line weight, corner radius, palette and zone layout in every image. Practical options: use the tool only for a draft idea and rebuild the final image as clean vector artwork following the sheet; keep a written style prompt and a reference image; and compare every output against the taxonomy before accepting it.

Accessibility checks. AI output does not know your contrast rules. Text inside generated images is often malformed or misspelt. Run the same checks as for any badge: contrast, text size, greyscale, small size and alt text written by a person. Keep the text in the final badge as real, checked lettering.

A human reviewer approves every AI-assisted image and records in the register that AI assistance was used, which tool category and which terms were checked. This supports transparency and later audits.

Worked example: A brief and a governance rule

Context 1: corporate academy or HR department. The academy of a manufacturing company (illustrative) needs an image for "Machine Safety Skill, Level 3". The academy lead writes this brief:

• Badge: Machine Safety, issuer Eastfield Manufacturing Academy (fictional).
• Purpose: recognises assessed, safe operation of production machines at advanced level.
• Taxonomy values: type Skill (hexagon), level 3 (three bars), domain Compliance (lime band, circle symbol).
• Text in image: "Machine safety" (two words, navy).
• Alt text: "Eastfield Manufacturing Academy skill badge, level 3, Compliance: Machine Safety."
• Technical: square artwork, minimum display 64 pixels, transparent background.
• Do not include: third-party logos, hazard signs, photographs.
• Rights: artwork is owned by the company; no external elements.
• Review: second-person check against the taxonomy sheet before release.

For governance, the lead names herself as owner, starts a register and sets a yearly review.

Context 2: university, college or training provider. A college uses an AI tool for a first sketch of a "Community Health Basics" badge symbol. The designer rebuilds it as clean artwork in the college style, checks the tool terms with the legal office, writes the alt text and records in the register: "AI-assisted sketch; final artwork redrawn; terms checked 4 November 2026."

Practice activity: Brief, review and governance

Complete all three parts for your own portfolio or the sample set.

1. Write two image briefs using the ten-item structure above, for two of your three assessment badges. Make sure the taxonomy values, text and alt text agree with your sheet.
2. Review a set of at least six badges (your own existing set, or the sample set from Module 1) against your taxonomy. Use the review sheet with the verdicts consistent, unclear or conflicting. Write a one-line fix for each unclear or conflicting badge.
3. Write a short governance note of five lines: owner, change process, review rhythm, register fields and sign-off.

Add the review sheet and the governance note to your taxonomy sheet pack.

Check your understanding

Assessment and badge

The assessment task

A person assesses your work with the rubric below. Nothing is graded inside this file.

Task: Produce a visual taxonomy sheet and three badge drafts or design briefs for a real or realistic badge portfolio in a corporate academy or HR department, or in a university, college or training provider.

What to submit

1. Visual taxonomy sheet (one to two pages, plus a one-page review) with: purpose and scope, dimensions and variables, value tables for type, level and domain, zone layout, colour and text rules, accessibility rules (contrast targets, minimum text size, alt text pattern), rules for exceptions, and at least two do and two do not examples.
2. Three badge drafts or image briefs. Each uses a different combination of type, level and domain. Each includes the exact alt text. At least one is a full ten-item image brief.
3. A review of an existing set of at least six badges, with verdicts and fixes (from Module 4).
4. A short governance note (five lines).

Declare any AI tool use, the terms you checked and the human checks you applied.

Evidence submitted for the badge: the visual taxonomy sheet and the three drafts or briefs (with the review and governance note attached to the sheet).

Rubric

Scores: 0 = not shown, 1 = partly met, 2 = met, 3 = strongly met.

Pass rule

To earn the badge "Visual Design Systems for Badges (Ability, L2)" (alignment code ICSF-CRDS-L2-A):

• Score at least 2 on every criterion, or score 1 on at most one criterion and 3 on at least two others, and
• Pass the final knowledge check below with at least 8 correct answers.

If you do not pass, the assessor gives written feedback and you may resubmit the parts concerned. Completion of the course shows that you completed these practice tasks to the standard set here. It does not prove competence in a particular job.

Final knowledge check

Sources and further reading

Only sources that are certain to exist are listed. Search for the titles on the website of the publisher. Status as of November 2026; check before publishing.

• W3C: Web Content Accessibility Guidelines (WCAG) 2.2. World Wide Web Consortium, W3C Recommendation. This is the source for contrast, text and non-text content requirements.
• Directive (EU) 2019/882, the European Accessibility Act. Official Journal of the European Union, published on EUR-Lex. Applies since 28 June 2025.
• EN 301 549: Accessibility requirements for ICT products and services. European standard published by ETSI, CEN and CENELEC.
• 1EdTech: Open Badges specification. 1EdTech Consortium. Describes how badge images and metadata are linked in badge data.
• W3C: Verifiable Credentials Data Model. World Wide Web Consortium, W3C Recommendation. Background on credential data structures.
• ICoBC Credentialing Skills Framework (ICSF). International Council on Badges and Credentials, Berlin. Source of the skill, level and stage used by this course.
• ICB-F01 Digital Badges and Credentials Essentials. ICoBC Kursbibliothek. The prerequisite course for this one.