Shaping Performance: Inside Hogarth's AI Model for Relevance at Scale
Hogarth Marketing Implementation Agency
London, UK - 29/11/2025
The real risk in modern marketing isn't spend. It's being ignored.
Hogarth's data is blunt: 78% of people skip generic messages, 63% get annoyed by irrelevant ads, 89% notice brands that "get" them, and 80% trust those brands. That's the gap between reach and actual attention. As Kate Johnson, group strategy director, puts it, if engagement is flat while impressions grow, alarms should be ringing.
Platforms also push freshness. Meta will down-weight ads if creative isn't refreshed roughly every two weeks, even if the content is strong. As global agentic solutions consultant Matt Vertigans notes, that means more output, more often, while staying coherent - or you fade fast. For context, see Meta's guidance via the Business Help Center.
The Shift
Hogarth's model reframes data from "what happened" to "what to do next." It's a loop, not a line. Insights guide ideas before production, and pre-flight analysis estimates performance before anything ships.
That changes the flow. Strategy, production, and optimization merge into a continuous system where each stage informs the next. Less guesswork, more signal.
How It Works
AI Agents
Autonomous agents handle heavy lifting across data analysis, creative versions, and formatting. Kate Johnson says the team built agents that support everything from strategy through production. The compounding effect is real: dozens of small process wins add up to serious time saved, without cutting corners.
Synthetic Personas
Hogarth builds synthetic audiences to test routes and spot cultural friction fast. One test flagged a thumbs-up icon as rude in Turkey - a small change that avoids a big issue. Result: quicker validation, safer creative, smarter global nuance.
Agentic Workflow
A master "orchestrator" agent coordinates specialist agents and surfaces human check-ins at the right moments. Strategists get an auto-ping when a draft is ready. Work moves adaptively instead of waiting on handoffs, while creative control stays with people.
Human Shapers
AI accelerates. Humans originate. Matt Vertigans puts it plainly: the human brain creates things that haven't existed before. Kate Johnson adds that AI can spot what's trending, but humans figure out why it matters. That's how empathy and taste stay front and center.
The Process
1. Dig Deep
Start by pulling first-party data, third-party data, and social listening into one picture. Then find the human story inside it. That truth is the base layer for everything after.
2. Build Connections
Link the audience problem to the business objective. That bridge becomes the strategic north star. It keeps work focused, coherent, and measurable.
3. Craft Relevance
This is where AI does its best work. Agents trained on brand DNA, guidelines, and voice turn strategy into clear narratives and variants. You get speed without losing consistency.
4. Make It Sing
Scale the idea into thousands of context-specific executions by audience, region, and funnel stage. Hogarth has gone down to regional nuances inside a single country - like different attitudes across North, South, East, and West Germany. One long-form piece can become a full media system of creative touchpoints.
The Proof
Hogarth reports 95% faster strategy creation, up to 1,000x more creative versions for precision, and an average 17% lift in engagement. In a recent demo, a full-funnel, multi-audience plan for a fictional fragrance brand was built in under three minutes, with messaging tuned to values, funnel stage, and platform.
A lot of the gain is workflow. "Checking a spreadsheet for a thousand ads - does a person need to do that?" asks Matt. Pre-flight analysis also trims wasted spend by pressure-testing creative before launch.
As Kate says, relevance drives performance. If the ad doesn't feel meant for you, you swipe. If it hits something true, you pause.
Why It Matters
The value here isn't automating creativity. It's freeing creatives to do creative work. AI handles volume and versioning; humans handle originality, cultural nuance, and emotion. Kate is clear: teams need empathy to find the human story that threads through a campaign.
AI brings the pieces together with less friction, so decisions happen earlier and ideas land cleaner. The payoff is work that's faster, sharper, and far more relevant.
For Creatives: What To Do This Week
- Audit your engagement signals. If impressions are up and comments, likes, and saves are flat, fix relevance first.
- Set a two-week refresh cadence per platform. Build lightweight variants in advance so you never scramble.
- Add a pre-flight check. Score ideas and assets for likely performance before production and before launch.
- Draft 5-7 synthetic personas per market. Include cultural flags (gestures, colors, humor, idioms) to avoid missteps.
- Codify brand DNA for agents: voice, tone, no-go zones, examples of "this, not that." Keep it short and living.
- Map the workflow: which agent does what, and where humans review. Automate handoffs; never automate taste.
- Pilot on one campaign. Measure: time saved, versions produced, engagement delta, and media saved from pre-flight kills.
Upskill and Next Steps
If you want a structured path to build these skills into your team, explore the AI Certification for Marketing Specialists. Practical frameworks, creative workflows, and brand-safe guardrails - the useful stuff you'll actually use.
The takeaway is simple: move fast, stay human, and make every piece mean something to someone. It's about making human connection.
Your membership also unlocks:
