Playground

This recorded test compares real Jev scores with human answers.

PREBUILT JEV TEST · 12 EXAMPLES

Scraped-data relevance

Keep web pages that answer a narrow research question.

A keyword match is kept, or a first-hand account is dropped.

WHAT ARE WE TESTING?

Does this page excerpt contain first-hand evidence of a user's specific difficulty during SaaS onboarding, rather than generic advice, a vendor pitch, or unrelated news?

YES

The page gives first-hand evidence about SaaS onboarding friction.

NO

The page is generic advice, a sales pitch, or unrelated news.

Where did Jev disagree?

Compare Jev’s recorded answers with these 12 human labels. Moving the line only explores the tradeoff on this set. It is not a holdout.

Recorded September 27, 2026. Opening it makes no new model call.
THE ANSWERS WE ALREADY KNOW

Examples behind the result

These human labels stay fixed while you explore the score and decision line.

6 YES · 6 NO

YES examples

Forum post: I tried to invite our team but the verification email never arrived, so we abandoned setup.

Review: The import wizard failed on row 3 and gave no error, so I couldn't complete onboarding.

Customer comment: We could not find the API token during setup and support took two days to answer.

User interview: I stopped at the billing step because I didn't know if a trial would charge me.

NO examples

The ultimate guide to SaaS onboarding best practices in 2026.

Our agency builds onboarding flows for growing SaaS teams. Book a call.

Company announces a new Series A round to expand its sales team.

Ten onboarding email templates every startup should copy.