What the documents look like
A brand concession inside a shopping mall differs from a brand-operated street store in its cashiering and settlement path. The mall collects payment centrally, and the brand settles with the mall under a joint-operation concession contract on a monthly cycle. The relevant data comes from the mall POS, the brand ERP, mall member campaigns, and the finance system, with daily and monthly update rhythms running side by side. The everyday documents are the joint-operation concession contract, mall campaign letters, sales receipts, and the monthly settlement statement.
For copy drafting, three facts matter. The campaign letter and the contract clauses define what a draft may cite. Sales receipts and the monthly settlement statement are the sources used to check any sales or price wording. Reconciliation, campaign participation, and concession fee handling consume most of the working time, so copy checking has to locate its basis across these documents rather than inside a single file.
Constraints at the interactive forms step
The form input node is a user interaction node. When it triggers, the conversation enters an interaction state, the workflow state is recorded, and execution waits until the user completes the interaction. After the user fills the required information and submits, the node collects the submitted form values and passes them to later nodes. The node itself does not act as a business database.
The user selection node is also a user interaction node. It triggers the same interaction state, records workflow state, and waits. When the user makes a choice, the node evaluates that choice and executes the corresponding branch.
Neither node performs field validation, business approval, write-back, or batch operations. Those need an external system or a later node. In copy drafting, the form collects the materials to be checked and the proposed sentence, while user selection confirms which handling branch applies. Interaction waiting and state recording follow the documented behavior. Submit-count limits, automatic judgement, and snapshot guarantees are not assumed.
Hypothetical paper exercise sample: Material text one reads, "This concession participates in the mall autumn member campaign from 1 September 2024 to 30 September 2024; the campaign letter states that all apparel items are sold at ninety percent of the tag price, and price wording during the campaign period follows the campaign letter." Material text two reads, "Clause 4.2 of the joint-operation concession contract provides that the brand supplies concession display and promotional materials, and any external copy involving price or discount must match the campaign letter for the current period." The proposed sentence to be judged reads, "During this concession's autumn member campaign, all apparel items are uniformly at ninety percent, and the original price resumes after the campaign ends." The judgement scope is limited to the same concession, the same campaign period, and the price and discount fields.
Configuration and implementation checks
The table below works from the sample above and lists node behavior, configurable fields, and implementation checks relevant to this page. The values are starting points for this hypothetical exercise, used for comparison against the material texts, and are not official defaults.
| Configuration item | Starting point or check method for this exercise | What it is compared against |
|---|---|---|
| Form input node fields | Three fill-in items: material text, proposed sentence, campaign period | Whether all three sample sentences are present, and whether the interaction state keeps waiting when one is missing |
| Required field check | Mark material text and proposed sentence as required | Whether a submission missing either item fails to reach later nodes |
| User selection branches | Three choices: consistent with material, unsupported, pending verification | Whether the judgement result falls into one of these branches |
| Interaction state observation | On trigger, record that the conversation enters the interaction state and the dialog box is hidden | Whether execution continues downward after the interaction completes |
| Downstream division of work | Field validation, approval, write-back, and batch operations marked as external system or later-node work | Whether database writes or approval actions appear inside the node |
Three common errors
- Reading "all apparel items at ninety percent of the tag price" in material text one directly as the proposed sentence's "uniformly at ninety percent, and the original price resumes after the campaign ends." Material text one does not state how price is restored after the campaign, so that part is unsupported and cannot be judged consistent with the material.
- Using the contract clause in material text two as the basis for price wording in place of the campaign letter. Clause 4.2 only requires external copy to match the campaign letter for the current period; the specific discount figure has to be checked back against the campaign letter, and the clause itself does not supply ninety percent.
- Treating the user selection branch result as a business approval conclusion. The node only executes a branch based on the choice. Approval and write-back require an external system, and selecting "consistent with material" does not constitute approval.
How to verify the setup
- Verification one: input material text one, material text two, and the proposed sentence, then compare whether the phrase "the original price resumes after the campaign ends" has a corresponding statement in the material texts. Neither material text one nor material text two states a restoration method. The expected result is "unsupported," and "consistent with material" is not satisfied.
- Verification two: input only material text one and the proposed sentence, without material text two, then compare whether the discount basis can be judged when the contract clause is missing. The expected result is "pending verification," recording the contract clause as an evidence gap, and it is not judged consistent with material.
- Verification three: input material text one, material text two, and a proposed sentence rewritten as "During this concession's autumn member campaign, all apparel items are at ninety percent of the tag price," then compare whether the discount figure and campaign period match material text one for the same concession, the same period, and the same field. The expected result is "consistent with material," without extending to price wording after the campaign ends. All three verifications use paper inputs and comparisons and were not actually executed.