AI in Marketing:
Questions, Answered.

The 30% rule in AI suggests that AI should handle no more than 30% of any given workflow before the quality-to-risk ratio becomes unfavourable. In marketing, this means AI takes on up to 30% of volume work — drafting, research, personalisation — while humans retain strategy, creative judgment, and final decisions.

Exceeding 30% often means AI is making judgment calls it isn't equipped to make reliably. The 30% figure isn't a scientific constant — it's a useful heuristic for keeping AI in an assistive role rather than an autonomous one.

The practical reframe: the question isn't whether AI is doing 30% — it's whether your AI is doing the right 30%. Most businesses have AI doing the wrong 30%.

Yes — 30% is a reasonable ceiling for most businesses starting with AI. But the question isn't really whether 30% is right. It's whether your AI is doing the right 30%.

Most businesses have AI doing 30% of things it shouldn't be doing — making tone decisions, choosing which message to send, approving creative — while their human team is doing the 30% that AI could handle better: research summaries, first-draft sequences, data compilation.

Start with AI handling the tasks you hate doing and measure the output quality. If it's acceptable, expand the scope. The risk isn't AI doing too much — it's AI doing the wrong things.

AI handles the repetitive; humans handle the meaningful.

Every AI implementation should free up human attention for higher-judgment work. If AI is doing your most creative, highest-stakes work, something has gone wrong — not with the AI, but with how it's been deployed.

The golden rule in practice: automate what you're tired of doing; protect what requires genuine expertise and relationship. If you dread doing it and it doesn't require your specific judgment — that's the AI's job.

AI is used in marketing across four main areas:

1. Content production — drafting copy, generating variations, repurposing long-form into social posts and emails at scale. One podcast becomes fifteen pieces in an afternoon.

2. Personalisation — dynamic email content, ad creative, and landing pages that adapt to audience segments in real time. Email sequences that respond to behaviour, not just send on schedule.

3. Prospecting and outreach — AI agents running autonomous sequences, qualifying inbound leads, and following up without manual intervention. Booking calls at 2am without anyone asking.

4. Analytics and insight — pattern recognition across data that identifies what campaigns actually moved metrics versus what looked impressive in isolation.

The teams getting the most from AI are using all four — not just content drafting.

A content creation framework: 3 core pieces of long-form content per month (pillar articles, reports, guides), broken down into 30 shorter pieces (social posts, email sequences, ad copy), distributed across 3 channels.

It solves the common problem of creating enough volume to stay visible without burning out a content team.

AI makes the 30× multiplier nearly effortless. One well-structured long-form piece can become fifteen quality derivatives in an afternoon. The rule's value isn't the specific numbers — it's the ratio: one strategic investment, many extracted returns.

Without AI, the 3-3-3 rule is aspirational. With AI, it's baseline output.

The 8 second rule states that the average attention span for digital content is approximately 8 seconds before engagement drops sharply.

Your content needs to deliver value or hook the reader within the first 8 seconds — a strong headline, a provocative question, or a clear outcome statement.

AI affects this in two ways: it makes it easier to test many variations of headlines and openings to find what hooks your specific audience, and it makes personalised opening lines at scale feasible where a human writer couldn't write 200 personalised first paragraphs.

The 8 second rule applies most acutely to cold content — ads, cold emails, cold social. Warm audiences and inbound content have more patience if the value proposition is clear.

For every 7 units of information, include at least 1 personal or emotional element. Pure information without human connection is forgettable; pure emotion without substance lacks credibility.

In AI marketing contexts, the 7×7 rule is a useful check on AI-generated content. AI is very good at information, very bad at genuine personal insight. The human element that makes content land needs to come from the brand voice, the writer's specific experience, or a genuine point of view.

AI drafts the 7; humans supply the 1. Without the human element, AI content is technically correct and completely forgettable.

The best current example is personalised video at scale. Brands use AI to generate thousands of personalised video variations from a single template — personalised by name, past purchase behaviour, or location.

A sports club can send 50,000 fans a video that says "Hi [Name], your nearest game is on [Date] at [Venue]" — each one rendered individually. The response rates on personalised video run 3-5× higher than generic equivalents.

AI handles the rendering; humans decide the creative strategy and approve the final output. That's AI doing what it does best: high-volume, rules-based personalisation at a cost that makes sense.

Other strong examples: AI agents qualifying inbound leads 24/7, and dynamic pricing optimisation for e-commerce. But personalised video is the most viscerally impressive demonstration of what AI + marketing looks like.