Create Safe Sample Assets Without Using Customer Information
Prepared with AI assistance by Alactic Editorial. Examples identified as hypothetical are illustrative, not customer results.
A colleague needs a sample statement for a message preview, so they duplicate a customer PDF and change the name. The layout looks convincing, but the account reference, signature image, document author, or second-page footer may still belong to the customer.
To create sample message attachments without customer data, start with a blank document and a written list of layout requirements. Fill it with deliberately fictional content, label it as a sample, and check both the editable source and the exported attachment. Use the same method for screenshots: build a separate demonstration scene instead of modifying a customer screen.
The goal is realistic structure, not a believable identity. A sample should exercise wrapping, pagination, image placement, and reference fields without carrying a real person's details or confidential business information. The filled worksheet below gives you a practical starting point.
1. Specify what the asset needs to demonstrate
Before writing sample content, identify the decision the asset will support. A document for checking a mobile preview needs different content from a document for testing page breaks. Neither requires a customer file.
Write a short asset brief containing:
- Format: the intended output, such as a PDF attachment or screenshot image.
- Layout questions: what must remain readable, wrap correctly, or fit within a page.
- Required fields: only the fields needed to answer those questions.
- Expected viewing context: a small preview, an opened attachment, or a presentation slide.
- Sample identification: where the fictional-content label will appear.
For example, a hypothetical statement preview might need a long recipient name, a multiline address, a reference, several line items, and a second-page footer. It does not need a real account number, payment link, signature, or customer photograph.
Keep a requirement only if you can explain what it tests. This prevents realism from becoming an excuse to import unnecessary detail.
2. Build from a blank source, not a disguised customer file
Create a new document, design canvas, or demonstration record. Reconstruct the necessary layout using generic headings, spacing, and fictional content. If an existing design template is available, use it only when its content and embedded assets are known to be suitable for sample work.
Changing visible text in a customer file is a fragile shortcut. Other pages, hidden spreadsheet cells, comments, embedded objects, or export properties may remain unchanged. Covering text with a shape also does not establish that the underlying information has been removed.
For screenshots, build the scene in a separate sample environment or a static mockup. Do not use a live customer view as the background. Browser tabs, profile pictures, notifications, recent-item lists, and neighboring records can reveal information outside the main panel.
If reconstructing the layout seems expensive, reduce the sample's scope. A single synthetic page that tests the key layout is more useful than a complete copied document whose contents are uncertain. When a full multipage asset is necessary, create each page deliberately rather than inheriting pages from an old file.
3. Use a filled sample-data substitution worksheet
Fictional worked example: The following worksheet describes a hypothetical two-page statement used to check an attachment layout. Every identity, address, reference, amount, date, and business detail below is invented for this example. It is not an Alactic customer document or evidence of a product feature.
Copy the worksheet into the asset's working notes. The field names make it reusable; the filled values make it usable immediately.
Asset identity and visible labels
- Purpose: Check a long recipient block, an itemized section, and a second-page footer in a PDF.
- Working filename:
SAMPLE_hypothetical_statement_layout_A.pdf. - Document heading: “Hypothetical sample statement”.
- Label on every page: “SAMPLE — FICTIONAL CONTENT — NOT FOR PAYMENT”.
- Use restriction in working notes: Demonstration and layout review only; not a customer communication.
Names and addresses
- Recipient name: “Sample Contact Alpha — Extended Layout Test”. The extra length tests wrapping without imitating a real person.
- Organization: “Fictional Sample Workshop”. Do not borrow a customer's trading name.
- Address line one: “Sample Building, Demonstration Floor”.
- Address line two: “Non-deliverable address for layout testing”.
- Locality: “Fictional District”.
- Postal field: “POSTAL-SAMPLE”. This intentionally does not represent a deliverable postal address.
- Email, if needed:
sample-contact@example.invalid. Do not substitute a staff member's address. - Telephone: “PHONE-SAMPLE”. Do not invent a plausible dialable number.
Account references and document content
- Account reference:
SAMPLE-ACCOUNT-ALPHA. - Document reference:
SAMPLE-DOC-LAYOUT-A. - Hypothetical statement date: “15 April 2030 — sample date”.
- Hypothetical line item one: “Demonstration item A”; quantity 2; unit price 12.00; line total 24.00.
- Hypothetical line item two: “Demonstration item B with an intentionally long description to test wrapping”; quantity 1; unit price 8.50; line total 8.50.
- Hypothetical total: “32.50 sample units — no amount due”.
- Payment details: Omitted. No bank account, card fragment, payment link, or payment QR code.
- Business narrative: “This fictional statement demonstrates document layout.” No copied order notes, pricing agreements, or internal comments.
Images and machine-readable elements
- Logo: A newly drawn rectangle containing “SAMPLE”, not a customer logo.
- Profile image: A geometric icon created for this asset, not a customer or staff portrait.
- Signature: Omitted; use the text “Signature area — sample only” if spacing must be tested.
- Barcode or QR area: A labeled placeholder box reading “CODE AREA — NO ACTIVE CODE”. It is not intended for scanning.
- Background: Plain color, with no reused office photograph, delivery photo, or map.
Metadata and export details
- Document title: “Hypothetical sample statement”.
- Author: “Sample Content”, where the editing tool permits an explicit value.
- Subject: “Fictional layout demonstration”.
- Comments and revisions: None intended in the distributed copy; inspect the output rather than assuming export removed them.
- Embedded filenames and paths: No customer names, account references, or confidential project names.
- Image descriptions: “Sample geometric icon”, not text inherited from another image.
Final accidental-information check
- Expected result: Every populated field matches this worksheet or an explicitly documented fictional addition.
- Pages to inspect: Both pages, including headers, footers, and any blank-looking areas.
- Links and attachments: None intended; investigate any unexpected clickable or embedded content.
- Review record: Leave reviewer and review date blank until someone actually checks the exported file.
4. Make the layout realistic without making the identity realistic
The worksheet deliberately uses obvious sample values. That can produce less natural-looking content, but it makes the asset easier to recognize when it leaves its original folder.
Add realism through shape and variation. Use a short and a long description, an address block that occupies several lines, differing item quantities, and enough repeated rows to test a page break. Keep fictional arithmetic consistent so reviewers can distinguish a layout problem from a content error.
Do not copy a customer's unusual transaction pattern, negotiated discount, complaint, or delivery instructions just because the name has changed. Those details can carry confidential business information independently of identity fields.
There is also a trade-off between safe placeholders and format validation. A field expecting digits may reject SAMPLE-ACCOUNT-ALPHA. For a visual mockup, show the placeholder as text. For an interactive test, use designated test values in a separate test environment where available. Do not solve the problem by borrowing a real identifier or disabling validation in a live workflow.
If a numeric-only value is essential, document the fictional test value and its purpose separately. Its numeric appearance alone should never be treated as evidence that it cannot match a real account.
5. Treat images and screenshots as content, not decoration
An otherwise synthetic document can still contain a real identity in a small image. Check profile icons, logos, stamps, signatures, scanned backgrounds, and image captions against the worksheet.
Prefer simple shapes or purpose-made illustrations when photographic realism is unnecessary. If the task is to assess image positioning, a labeled geometric image usually answers the question without introducing a person's face or a recognizable location.
For a screenshot mockup, populate every visible region deliberately. Include fictional navigation labels and a generic profile icon, and exclude unrelated tabs or notifications. If a reviewer needs to assess a narrow panel, create that panel directly rather than capturing a broader screen and trusting a crop.
Keep the sample label inside the asset's main content area. A label placed only in a surrounding slide or caption can disappear when someone downloads or crops the image. For a multipage document, repeat it on each page so an extracted page remains identifiable.
6. Check the exported attachment, not just the editor
The deliverable is the exported file, so review that file independently of the editing view. Export settings, font substitution, and pagination can change what recipients see. Properties may also differ between the editable source and the final attachment.
- Open the exported file afresh. Confirm the filename, page count, sample labels, and expected content.
- Read every visible area. Check headings, totals, footnotes, image text, page numbers, and second-page content.
- Search and select text where supported. Compare extracted or selectable text with the worksheet. Investigate unexpected text, including content that is not visually prominent.
- Inspect properties and embedded content. Review metadata, comments, attachments, links, and image details using tools suitable for the format.
- Check the intended viewing size. Make sure labels remain readable and the layout question can actually be assessed.
- Have another person compare the output with the worksheet. Record what they checked rather than asking for a vague sign-off.
These checks reduce avoidable mistakes; they are not proof that every possible hidden element has been detected. If the file's origin or structure is uncertain, rebuilding a simple synthetic source is preferable to relying on a quick visual review.
7. Recover from mistakes without preserving the risky shortcut
If real information appears during review, stop using that version. Do not keep replacing individual discoveries in a file that began as a customer document. Return to a blank source and rebuild from the worksheet.
If the asset has already been shared, identify the shared copies and follow the appropriate internal handling process. Replace accessible sample links or files where possible, but do not assume replacement removes previously downloaded copies. Keep the correction task separate from the new sample's layout review.
Other failures need different fixes:
- The sample label is unreadable in preview: increase its prominence and retest at the actual preview size.
- The layout looks too sparse: add fictional rows or longer descriptions, not copied transaction text.
- A reviewer mistakes it for a payable document: strengthen the title and no-payment wording; remove active payment elements.
- The export retains unexpected properties: adjust the source or export method and inspect the replacement file again.
8. Finish with a measurable release check
For the next sample attachment, complete the worksheet before opening a customer record. Build one synthetic source, export it, and ask a second reviewer to compare the output with the filled fields.
Use concrete acceptance criteria: every page carries a readable sample label; every populated identity and reference field is accounted for; no unexplained image, link, or embedded attachment remains; and metadata has been reviewed for the output format. Record unresolved findings rather than treating an incomplete review as a pass.
Then assess usefulness separately: can the reviewer see the long-name wrap, the multiline address, and the intended page break? A sample is ready only when it is both clearly fictional and capable of answering its layout question. Save the approved synthetic source with its worksheet so the next person can reuse it without reaching for a customer file.
Put the next conversation to work.
Explore Alactic Communications for WhatsApp conversations, campaigns and AI-assisted workflows. Choose the product and plan that fit your team.
Explore Communications