Map one competitor's entire visual system

Reverse their site, their ads and their social into one corpus, and find both the rules they run on and the places they break them.

Knowing what a competitor looks like is not the same as knowing their rules. Reversing every surface at once gives you the rules, and the places they fail to follow their own rules are the places you can take ground.

Why this will not look generic: It reads one competitor across every surface at once, so what comes out is their operating system rather than an impression of whatever campaign happens to be running this month.

What you walk away with

• A prompt for every asset across their surfaces, in one folder
• The rules their system actually runs on
• The surfaces where they contradict themselves
• A positioning note on where the gap is

Typical run: 20 to 50 across three or four surfaces. Breadth matters more than depth here: ten assets from four surfaces beats forty from one.

This one happens in one conversation

You bring: One competitor, captured across their site, their paid social, their organic and their packaging if it exists.

Open with: These all come from one competitor across different surfaces. What rules is their visual system running on, and where do they break their own rules?

What the thread holds on to:

• Which surface each asset came from, so an inconsistency is attributable rather than just noise
• The system it inferred, which you can test by asking it to predict an unseen asset
• Your own positioning, so the gap analysis is about you and not abstract

Then you can just ask:

• Predict what their next product launch image looks like
• Where is their system weakest, and is that weakness worth attacking?
• Which of their rules should we deliberately not share?

What the reverse actually gives you

Point it at: One competitor, captured across four surfaces in an afternoon.

What comes back (representative example): The system is tighter than it looks and it breaks in one place. Site and packaging agree completely: flat two-colour illustration, hard edges, no photography anywhere, and a single accent used only on calls to action. Organic social follows it. Paid social does not, and uses stock photography with a colour overlay in eleven of fourteen assets, which reads as a different company. The tell is the overlay: it exists to drag stock back toward the brand palette, which means paid is being produced by someone without access to the illustration system.

Things you would never have thought to type:

• no photography at all on site and packaging
• the accent reserved for calls to action, nowhere else
• paid social breaking the system in 11 of 14 assets
• a colour overlay used to drag stock toward the palette

What you change: Nothing of theirs. The finding is the inconsistency: their paid is produced outside their own system, and that is the surface where consistent work from you will read as more established.

How it works

1. Capture across every surface (In Chrome) - Breadth is the point. Site, paid, organic, packaging, email if you can get it.
2. File by surface - Tag each one with where it came from. The tags are what makes an inconsistency attributable later.
3. Infer the system, then find the cracks (In Claude) - Test the inferred system by asking it to predict an asset it has not seen.

Cost and time

This workflow runs at 2 reverses + about 1 minute. Your result carries automatically from each step into the next, so there is no re-uploading between tools. Reverses are billed from your reverse meter and generation steps from your AI Credits, both bundled in every pay-as-you-go credit pack.

Browse all guided workflows or all individual tools. Pay as you go, no subscription, AI Credits never expire.