THE SHORT ANSWER
Ask where catalog data is processed, who can access it, how credentials are protected and how tenant boundaries are enforced. Review retention, export and deletion behavior before uploading data. A provider’s security claims should be supported by documented controls or demonstrable behavior, not reassuring badges alone.
Classify the data you intend to share
A catalog may contain public product facts alongside private supplier notes, unpublished pricing or internal references. Remove fields the workflow does not need. Product-feed evaluation ordinarily has no reason to include customer records or authentication secrets.
Check source URLs before sharing them. Some contain access tokens in query parameters or paths. Use the provider’s supported secret mechanism rather than placing credentials in a feed, screenshot or support message. Keep the scope of access appropriate to the task.
Understand where processing happens
Distinguish browser-local tools from server processing and third-party services. Ask whether uploaded files, logs or proposed values are retained and whether external AI services receive product data. A generic privacy statement may not answer those workflow-specific questions.
Feedaptor’s public free tools process their bounded inputs in the browser and do not intentionally upload those values. Consented usage events describe tool actions rather than raw inputs. That public-tool behavior should not be assumed to describe every private workspace feature or another provider’s product.
Test access and operational boundaries
For multi-user or multi-client software, ask how workspaces, jobs, exports and credentials are scoped. Review roles and what happens when access is removed. A shared dashboard is not evidence that background processing respects the same boundary.
Inspect support and administration practices. Ask which people can view data, how sensitive actions are recorded and whether you can retrieve an audit trail. Avoid requesting or accepting undocumented security certifications as a substitute for understanding the actual controls.
Plan retention and exit
Ask what is retained, for how long, and how the policy applies to source files, derived values, backups and logs. Clarify export formats and whether reviewed data remains usable outside the service. Put important contractual requirements in the appropriate agreement.
Use a bounded sample until the handling is understood. Record unanswered questions and decide whether they block the intended use. Security review should lead to concrete data and access decisions, not a collection of vague yes/no answers that nobody can verify.
ILLUSTRATIVE WORKED EXAMPLE
Make it concrete.
Illustrative example: a supplier feed includes a token-bearing source URL and private margin columns. Before evaluation, the merchant removes unnecessary columns and uses the provider’s supported credential configuration. The review then checks workspace access, exports and retention with the reduced sample.
Your practical checklist
- Remove unnecessary private fields and customer records.
- Keep credentials out of feeds and screenshots.
- Distinguish local, server and third-party processing.
- Review workspace, job and export access boundaries.
- Clarify retention, export and exit behavior.
Common questions
Are public product facts always harmless to upload?
Context matters. A file may combine public facts with private columns, unpublished offers or access-bearing URLs. Inspect the actual data.
Does Feedaptor claim a security certification?
Do not infer one from general trust copy. Review the published trust information and ask for evidence of any specific control your organization requires.
Sources & scope
Official references checked 29 October 2026. Examples are illustrative, not customer results. Platform requirements can change; use current destination documentation for implementation decisions.
Our editorial method & corrections ↗