How we write and check the worksheets

Start with the output

We choose a different practical output for each entry. The inventory covers corrections, conditions, quantities, exclusions, responsibility and uncertainty through distinct scenes. We do not create new pages by changing city names in the same task. The field structure must fit the decision: a tray for an order, a route for directions and a status grid for a handover.

Write an original scene

German examples are written editorial exercises. They are not quotations, transcripts or descriptions of recordings. We supply an English translation and explain the phrase that controls the answer. Names, schedules, fares and arrangements in those scenes are invented. We do not attach synthetic audio to make a written exercise appear recorded.

Build a checkable key

Each key names the supported answer and a plausible error. It also identifies absent information. Unknown is valid when the scene gives no answer. We keep estimates, proposals and conditional actions distinct from confirmed facts. The transfer step asks readers to repeat the task on real speech, without importing our invented details.

Check scope and product facts

We checked the official product summary and machine-readable facts on 9 November 2026. The source fact review date is 28 October 2026. We make Jibber Jabber. App descriptions are not independent tests, and task suitability is an editorial judgement. We do not invent ratings, statistics or learner results.

What has not been established

No external native-speaker review, measured learning trial or exam alignment is claimed. Static build checks verify routes, metadata and link rules. They do not establish language perfection, accessibility in every browser or the quality of a chosen real clip. The final visual review is a separate publication step.

Use the source register for product evidence and the FAQ for scope questions. For a problem with the app, use its official help information.