Guide
Stress-test the work you already have
If the work already exists, paste it in: a marketing plan, a pitch deck, a set of ads, or a model. Brandium's Review mode reads the craft first, then the maths. It won't cheerlead.
Example on the desk
"Here is the deck. Where is the idea weak, and where are the unit economics a story?"
You leave with a stress test: gaps named, no applause.
What a stress test looks for
In the idea - Is there one clear message, or five competing ones? - Is the audience a person in a situation, or a demographic? - Is the proof evidence, or more claims? - Do the concepts differ, or are they tints of one idea? - Does anything in the deck contradict the positioning?
In the maths - Is CAC stated, and on what basis? - Is LTV built from an actual repeat rate, or assumed? - Does the margin survive fulfilment, returns and discounting? - Is the growth curve explained, or just drawn?
Common findings (illustrative)
| Finding | Type |
|---|---|
| "Our customers love us" with no review or retention data | Claim without proof |
| LTV based on a 40% repeat rate for a product launched last month | Story, not data |
| Three ad concepts with the same mechanism | One idea |
| Paid social CAC taken from an industry average | Borrowed number |
| Margin calculated before fulfilment cost | Missing cost |
Why a tool that won't flatter is useful
Pitch decks and plans are usually reviewed by people with a reason to be kind: colleagues, friends, an agency that wants the account. The questions an investor, buyer or board will ask are the unkind ones. It's cheaper to hear them first.
After the stress test
- If the idea is weak, go back to Build for positioning and codes.
- If the campaign is one idea, use Campaign for three distinct concepts.
- If the maths is a story, use Model with the costs you actually have, and Decide for a spend verdict.
Preparing a deck for a stress test
You'll get more from the test if you paste the deck with its numbers and say what it's for: investor, board, client pitch or internal plan. Say which slides you're least sure about. If the deck relies on a model, paste the model too, so the maths can be read against the story. Brandium reads what you give it. It won't fill in a missing CAC or repeat rate to make the deck hang together.
What you do with the result
Treat the findings as a to-do list, not a verdict on the idea. Fix the claims without proof first, because they're the cheapest to fix and the most damaging to leave in.
Questions
Will Brandium tell me my idea is good?
Only if the file supports it. It names the gaps, with no applause.
What can I paste in?
A plan, a deck, ads, or a model.
Which mode does this?
Review: paste existing work, and it reads the craft first, then the maths.