How we test

A recipe is only worth trusting if someone actually ran it. Here is what that means here.

Who writes and tests

We write and test Plainly Guide recipes. We use AI to help draft, and we say so. A person then runs each recipe on realistic work, fixes what breaks, and writes down what happened. Right now most recipes are still drafts and say so. Read more on the About page.

The labels on a recipe

• Not tested yet: a draft. The steps are a starting point and may not work as written.
• Last tested: the date a person last ran it. If a tool changes, we retest and update the date. We aim to recheck every few months.
• Prices checked: the date we last looked at a tool's pricing. Always confirm on the tool's own site.
• Check this yourself: what a person should review in the AI's output before using it.
• Prompt, Tool, or Automation: what kind of recipe it is. Prompts are copy-and-paste. Tools and automations need an account with another company.

How we choose tools

• We start with the job a small business needs done, then pick the simplest tool that does it well.
• We favor tools with a free plan or a clear price, and we say what it costs.
• Affiliate programs do not decide which tools we pick. See the affiliate disclosure.
• We test with sample or made-up data, not real customer information.

What we will not do

• Promise results we have not seen.
• Tell you to paste private customer data into an AI tool.
• Use fake urgency, invented testimonials, or hype.
• Publish an AI draft without a person reading it.

AI can still be wrong

Even a tested recipe can give a bad answer on your data, and tools change without notice. Read the output before you use it, especially anything that goes to a customer or involves money, law, or health. See the terms.

Corrections

Found a mistake or a recipe that stopped working? Email contact@plainlyguide.com. We correct errors quickly. When a fix changes the advice, we update the "Last tested" date and note what changed.