Guide · 5 min read
How to Make Your Marketing Team More Efficient With AI
By Fredrika Frenkiel, Head of Studio & AI Creative Operations at Lunar, founder of Code of Alfred. About my work
The fastest way to make a marketing team more efficient with AI is to target the specific bottleneck costing the most time or causing the most rework, not to introduce AI broadly across every task at once. Efficiency gains come from redesigning a handful of high-friction workflows properly, not from sprinkling AI tools across an unchanged process.
1. Find where time is actually being lost, not where it feels slow
Ask the team directly: what gets redone most often, what takes the longest relative to its importance, and what causes the most last-minute scrambling? The answer is rarely “everything”, it's usually two or three specific workflows: brief intake, asset adaptation across markets, or repetitive QA and formatting checks.
2. Redesign the workflow, then decide where AI fits
For each bottleneck, write out the current steps plainly. Then identify which steps are mechanical and repeatable (a strong candidate for AI) versus which require judgment, taste, or a relationship (which should stay human). A useful rule: AI is a strong fit for generation, structuring, and scaling; it's a poor fit for prioritisation, final creative judgment, and anything where being wrong has real cost.
3. Build reusable systems, not one-off prompts
The biggest difference between teams that get lasting efficiency gains and teams that get a brief novelty boost is whether the AI work is treated as infrastructure. A one-off prompt saves one person time once. A structured, reusable prompt system, chained across a workflow, with defined inputs and outputs, saves the whole team time every time the workflow runs.
4. Automate the handoffs, not just the content generation
A large share of marketing inefficiency isn't the creative work itself, it's what happens between steps: briefs sitting unread, files being manually moved between tools, status updates chased over Slack. Automating triggers between systems (a brief being submitted automatically creating a project and notifying the right people, for example) often saves more time than automating the creative output itself.
5. Measure the efficiency gain honestly
Track a concrete before/after on the specific workflow you changed, time from brief to delivery, number of revision rounds, or hours spent on manual formatting, rather than a vague sense that things feel faster. This also protects the initiative: efficiency claims that can't be measured don't survive the next budget conversation.
6. Scale only what's proven
Once a workflow redesign demonstrably works, extend the same pattern to the next bottleneck rather than trying to overhaul everything simultaneously. Efficiency compounds when each fix is solid, not when many half-finished changes run in parallel.
This bottleneck-first approach, find the real friction, redesign the workflow, then build the technical layer, is the same method behind the systems I've built for marketing organisations. See how I approach this in practice.