Hypotheses for A/B tests
Describe the page and the problem - you will receive a queue of hypotheses based on potential, formulated in a verifiable form and indicating which metric to look at.
The main limitation of an A/B test is not ideas, but power: to catch an increase of a fraction of a percent, tens of thousands of visitors are needed. Therefore, the less traffic, the larger the change should be. The tool produces a queue of hypotheses in a testable form, with a mechanism, a target metric and an assessment of whether you have enough traffic.
AI answers are free, but limited by the overall site limit. If the limit is reached, come back later: repeated clicks will not speed up the response.
Little traffic - change it a lot
The main limitation of an A/B test is not ideas, but power. To reliably catch a tenth of a percent increase in conversion, you need tens of thousands of visitors. Most sites don't have them, and a button color test will simply never amount to a meaningful result, honestly eating a month.
The practical conclusion is the opposite of intuition: the less traffic you have, the larger the change should be. A different offer, a different structure of the hero section, a different order of blocks make a difference that is visible in a small sample. Leave minor edits to those whose traffic counts in the hundreds of thousands.
Therefore, each hypothesis here is formulated with a mechanism: if you do this, the metric will change, because this will happen. A formulation without a mechanism is an idea, and it cannot be verified: it is not clear what is considered confirmation.
What the tool intentionally doesn't do
Does not insert numbers, deadlines, benchmarks and names of companies that did not exist in your input. If the output requires a number that you did not provide, in the answer will be marked “to clarify” rather than a plausible number. This limitation is worth in the system prompt and works against the main property of language models: It’s difficult to complete what’s missing.
The request text is sent to the AI provider, and the response is returned to the browser. The site does not save these texts in the database. To limit expenses separately the number of requests and the anonymized connection identifier are taken into account. If the free limit is reached, the tool will offer to come back later.