AI agent for creative directors
Pitch Competitor Positioning Scan Agent
A pitch concept set checked against recent work for the prospect and competing agencies, with overlaps resolved or knowingly accepted
What it does
Nothing is worse than presenting a big idea and learning the client saw nearly the same one from a competitor last month. Before a pitch, this agent collects the prospect's recent campaigns and the known work of competing agencies from public sources and the team's own notes. It extracts the themes, tones, slogans and visual approaches used, then checks each proposed concept against them. If a concept collides with something already out there, it flags the overlap with the evidence and proposes distinct angles. It then re-checks the revised set until all concepts are clear or the director decides to keep one on purpose. The director approves what goes in the deck. Edge case: a concept that matches only a competitor's old campaign from six years ago is rated low risk but still noted.
How it works
Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.
Read the steps as a list
- Pitch concepts drafted
- Collect the prospect's recent campaigns and competitor work from public sources
- Extract themes, tones, slogans and visual approaches with dates
- Compare each concept with the collected work
- Is each overlap supported by a dated example?If not: Search again for the example or drop the overlap claim. Back to step 2.
- Rate the risk of each concept as high, medium or low
- Propose distinct angles for the high-risk concepts
- Do the revised concepts now clear the overlap check?If not: Revise the angle and compare again. Back to step 3.
- Director decides which concepts go in the deckThe agent waits here for your OK.
- Positioning scan report attached to the pitch
How it decides
It rates overlap by theme, tone and visual idea against dated examples. Recent and close matches count as high risk, old or loose ones as low.
- High risk: same theme and tone used within the last 18 months
- Medium risk: same theme but different tone or execution
- Do not rely on an overlap claim without a dated example
- Stop after 3 revision rounds per concept
Make it yours
Every agent is a starting point. You choose these settings for your own situation.
- Competitor list
- Time window for recent work (default 18 months)
- Sources to search
- Risk rating rules
- Number of revision rounds allowed
What keeps you in control
It always asks you first
- Director approves the final concept set for the deck
Hard limits
- Use only public or the team's own material
- Never present an unsupported overlap claim as fact
It stops when
- Done: Every concept is cleared or knowingly accepted by the director
- Stop: Public evidence is too thin to judge, so the agent marks the concept as unchecked
Set it up
We guide you through the set-up, step by step
Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.
- One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
- The agent then walks you through connecting your own data, one source at a time
- A downloadable copy with the flow chart, the rules and the full guide
