A graphic design system under control

How can you use AI to harmonise your communication materials without losing your brand’s distinctive character?

1960 € per session, excl. VAT

In short

A 7-hour day, excluding breaks, to turn brand positioning into usable graphic design rules. You use generative AI to explore visual directions, select an approach, formalise a concise visual style guide and apply it to three communication materials. Quality is checked using an assessment grid for consistency, readability and compliance.

Why this training?

Individually appealing images do not constitute a graphic design system: varying palettes, incompatible compositions and divergent photographic treatments undermine brand recognition. AI makes exploration easier, but requires a reasoned selection and stable rules. The day links each visual choice to the positioning, then tests how well it holds up across several formats. Running case study: Avelune Céramique, a fictitious artisan tableware workshop, is preparing a collection for independent shops. Its catalogue, social media posts and product sheets feature conflicting styles. You must harmonise these three materials around an understated, warm and contemporary positioning. Transfer to the workplace: On your return, compare the concise visual style guide against an existing anonymised communication material, identify discrepancies and have the corrections approved by the brand manager before publication.

Objectives of this training

  • Translate positioning into observable visual attributes
  • Compare two art directions using a reasoned assessment grid
  • Formalise the rules for colour, typography, composition and imagery
  • Apply the selected direction to three communication materials
  • Assess the consistency, readability and legal risks of the outputs

Who is it for and what are the prerequisites?

Roles and tasks concerned

  • Communication and marketing managers responsible for harmonising communication materials
  • Graphic designers and content creators integrating AI into their production

Prior skills

  • Be able to formulate and revise a prompt in a generative AI assistant.
  • Have a command of layout basics and common image formats.
  • Understand the target audiences and uses of brand communication.

What you will work on

  1. From positioning to a visual brief

    Objective: Establish verifiable graphic design benchmarks. Target audiences, personality, reading contexts; distinction between brand attributes and aesthetic preferences.

  2. Explore two art directions

    Objective: Produce distinct approaches without imitation. Structured prompts: intent, palette, materials, lighting, framing and exclusions; limits to reproducibility.

  3. Select references and check their legal suitability

    Objective: Select a relevant, documented approach. Copyright, trade marks, licences, individuals’ consent; required disclosures depending on context; risks of deepfakes and false attribution.

  4. Build the concise visual style guide

    Objective: Turn the approach into applicable rules. Colour codes, typographic roles, margins, hierarchy, image treatment; fixed rules and permitted variations.

  5. Adapt without drifting

    Objective: Maintain recognition across formats. Adapting framing and compositions; controlled addition of text; checking for invented product details.

  6. Check and prepare the handover

    Objective: Measure discrepancies and organise approval. Compliant/non-compliant assessment grid: palette, typography, composition, imagery, readability; human review before distribution.

What you take away

  • Visual brief and comparison of two art directions
  • Concise visual style guide with permitted and prohibited examples
  • Three harmonised mock-ups: catalogue, social media post and product sheet
  • Completed quality assessment grid and register of sources, rights and approvals

Method and assessment

  • Short teaching inputs followed by immediate practice on the running case study.
  • Individual work followed by work in pairs, with peer review of the outputs.
  • Guided group discussion: what works, what needs correcting and what remains a human decision.
  • Preparation: Arrange access to a tool authorised by your organisation and check its terms for the use of content. Gather a fictitious or anonymised brief and visual references that you are permitted to use.

Observable criteria: Each graphic design choice is linked to a positioning attribute. ; The three materials comply with the defined colour codes, hierarchies and composition rules. ; Text and contrast are checked at actual viewing dimensions. ; Each resource has a documented source and usage status. ; Identified discrepancies are corrected or explicitly submitted for approval..

Frequently asked questions

Does the logo need to be redesigned?

No. The work focuses on the graphic design system; an existing fictitious brand mark can be retained.

Can real portraits be used?

The exercises prioritise fictitious content. Any identifiable portrait requires appropriate permissions and consent.

Are generated images automatically royalty-free?

No. Check the terms of use, similarities and rights associated with the resources you introduce.

Must the use of AI be disclosed?

Determine the required disclosures according to the applicable regulations, the channel and internal rules, particularly for manipulated realistic content.

Last updated: