Skip to main content
PlaybookPractitionerEU-wide17 min readTechnology, AI and Infrastructure#9

AI bootcamp for EU affairs, day 5: bring your campaign to life

Turn the approved work from day 4 into a controlled campaign-production workflow: visual concepts, short-form video, a seven-day content plan, measurement and multilingual adaptation, with human review before anything goes live.

Sebastián Rodríguez Pérez

Founder

European Campaign Playbook · Spain

Sebastián Rodríguez Pérez is a European campaign strategist with direct experience running and advising campaigns for some of Europe's largest pro-European organisations. He founded the European Campaign Playbook to codify practitioner knowledge across the EU's 27 political environments and make it accessible to campaign professionals, advocacy teams, and civic organisations working at the European level. His work spans electoral strategy, EU public affairs, digital campaigning, and the integration of AI tools into political communication workflows.

European campaign strategyEU public affairsDigital campaigningAI in politicsCross-border advocacyGEOSEO

Disclosure

Sebastián Rodríguez Pérez is the founder of the European Campaign Playbook and Campaign Intelligence Library. He has worked with pro-European organisations across multiple EU member states. He maintains editorial non-partisanship in all published content; where articles relate to campaigns or organisations he has directly worked with, this is disclosed within the article. He holds no current positions in active electoral campaigns.

What it covers

The production half of the campaign workflow: turning approved messages into visual concepts, short-form video, a channel plan, measurement and multilingual versions, while preserving factual accuracy, provenance, accessibility and human editorial control.

Who it is for

EU public affairs, advocacy and communications teams that have already agreed their campaign objective, evidence base and core message and now need to turn those decisions into publishable assets and a controlled rollout.

When to use it

Use this only after the campaign objective, evidence base, narrative and approved core message are clear. If those are still moving, return to day 4 before producing assets at scale.

Key takeaway

AI can accelerate production, but speed is useful only if every asset still traces back to an approved message and evidence base. Prototype with AI. Review with humans. Disclose synthetic media where required. Measure the campaign against its objective.

In brief

  • Day 5 converts the approved campaign framework from day 4 into production and distribution.
  • Start with a creative brief before generating images or video; the visual should serve the message you already approved.
  • Treat AI-generated images and video as production assets that require provenance, editorial and compliance review before publication.
  • Build the rollout around campaign objectives and audience information needs.
  • Measure exposure, attention, action and campaign outcomes separately; engagement alone does not prove impact.
  • Treat AI translation as a draft. Terminology, political nuance and local context require fluent human review.
  • Every published asset should have an owner, source trail, approval state and version history.

Framework

The production loop

Every asset moves through five stages. AI can accelerate Prototype and Produce. It does not remove Brief, Review or Measure.

LayerLabelDescription
BriefWhat must this asset communicate?Approved creative brief
PrototypeWhat are our options?AI-generated concepts or drafts
ProduceWhich option should become a real asset?Edited production asset
ReviewIs it accurate, accessible, compliant and on-message?Approved version
MeasureWhat did it actually achieve?Evidence for the next decision

Day five starts with a brief

By the end of this article

You will have:

  • a visual brief linked to your approved campaign message;
  • three prototype visual concepts;
  • a short-form video concept and production brief;
  • a seven-day content rollout;
  • a campaign measurement framework;
  • multilingual draft versions of approved content;
  • a pre-publication checklist for synthetic media, evidence, accessibility and EU compliance;
  • a completed advocacy canvas covering the full campaign workflow.

Bring the campaign into production

Day 4 established the foundation: evidence, organisational context, public narrative, stakeholders, inclusion and the approved core message. Day 5 begins only after those decisions are stable.

Now the job changes. You are no longer asking:

What should this campaign say?

You are asking:

How do we express the approved message consistently across formats, channels, languages and time?

That distinction matters because generative AI makes production extremely easy. It can also make inconsistency extremely easy.

One image introduces a new metaphor. A social post adds an unsupported statistic. A translated slogan changes meaning. A video creates a visual implication the policy does not support.

The objective of Day 5 is therefore not maximum content. It is controlled production.

Who will benefit most

This is especially useful for:

  • communications leads planning a campaign launch;
  • public affairs teams producing content from approved policy positions;
  • small organisations without a dedicated creative-production team;
  • campaign managers coordinating several channels and languages;
  • teams introducing AI into creative production without removing human editorial control.

The production loop

Every asset moves through the five stages in the framework above: brief, prototype, produce, review and measure.

Where day 5 fits

Five steps filling the how, when and where box of the advocacy canvas: 6 images, 7 video, 8 rollout, 9 tracking and 10 translation.
Today fills the last box of the advocacy canvas.

Day 4 answered why, who and what. Day 5 answers:

  • what does it look and sound like?
  • where does it appear?
  • when does it appear?
  • how do we know whether it worked?
  • how does it travel across languages?

Original working example

If you want to see the original experiment that inspired this exercise, the archived conversation is available. Treat it as a historical example.

Want to try it live?

If you want to go deeper and actually run all ten prompts with me, join one of our live workshops. You’ll get:

  • a walkthrough of each prompt in real time
  • space to ask questions, test your own use case and get feedback

Open the workspace or project where you keep your approved day 4 material, and work through the stages in order.

Every prompt from this bootcamp is collected in the AI prompt library.