# The absorption test: seven scores for any AI product idea (free sample)

A free sample from Workflow, Not Wrapper by PaperChaseWebb. Works in Grok Bot, Claude, ChatGPT, Claude Code or Codex.

A clean screen over one model call is the easiest business to copy, and the model vendor can absorb it in a single release. This test tells you, before you build, how much of your idea would survive that.

## The quick test

If the platform added your feature tomorrow, what would your buyer still need you for? If the honest answer is "nothing", you have timing, not a business.

## The absorption test: seven scores

Score each line 0, 1 or 2. Write one sentence of evidence next to every score. A score with no evidence counts as 0.

| # | Score | 0 | 1 | 2 |
|---|---|---|---|---|
| 1 | Platform absorption | The vendor's default chat already does it | The vendor could add it as a button | It needs steps, inputs or rules the vendor will not build for one niche |
| 2 | Workflow depth | One input, one output | Several steps, but no review or delivery | Inputs, template, review, approval and delivery in the buyer's format |
| 3 | Data that improves results | None | Saved preferences | Buyer records or past outcomes that visibly make the next result better |
| 4 | Distribution access | You hope people find it | You can reach some buyers | You already sit where buyers gather, or they come back on their own |
| 5 | Model independence | Welded to one model or vendor feature | Swap possible with real rework | Model swap is a config change and a test run |
| 6 | Unit economics | Cost per job unknown | Known, thin margin | Known, with room for retries, refunds and support |
| 7 | Truthfulness and compliance | Output can mislead and nobody checks | A check exists but is optional | A check and a human approval are part of the product |

**Reading the total (0 to 14):**

- **0 to 5:** a wrapper. Build it only as a cheap experiment or a free tool that leads somewhere.
- **6 to 9:** a product with a hole. Name the lowest line and fix that first.
- **10 to 14:** worth a spec. The full pack has the one-page spec for this step.

Any 0 on line 7 stops the idea until it is fixed, whatever the total. A product that can mislead a buyer's customer is a liability with a checkout button.

## Two rules to paste

```
- Score every product idea with the seven-line absorption test before any build. Evidence for every score; no evidence counts as 0.
- Any 0 on truthfulness and compliance stops the idea.
```

## Check prompt

"Here is my product idea [paste]. Score it on this seven-line absorption test [paste the table]. Give one sentence of evidence per line, mark any score without evidence as 0, name the weakest line, and tell me what would move it up one point."

## What is in the full pack

The one-page workflow product spec, a problem mining loop with a complaint log, a channel translation table for launch tactics, the model column check, an offer ladder with six pricing rules, a four-question check for success stories, a fidelity gate for generated output, a failure table, a worked example, a rules block and 14 prompts.

## Source

Built in our own words from Ajay Yadav (@thevibefounder) on AI wrappers versus workflow products: https://www.instagram.com/reel/De2KuKxyfs1/