A comparison matrix often reduces a feature to a check mark. That can be useful for a simple presence question, but payment features may depend on the applicable program, issuer, verification or other terms. If the matrix removes those conditions, its visual simplicity changes the meaning. A checked box can look like a universal promise the source never made.
Turn the sentence into a conditional record
The public Corpay Prepaid page attaches conditions to some feature statements and notes issuer-specific limits. An analyst should preserve those qualifications rather than treating the page as a list of unconditional capabilities. This article does not verify any individual account’s status or access.
Use a research row with statement, condition, source and unresolved applicability. The condition belongs in the same row as the feature. A distant general disclaimer cannot reliably repair a table that marks every account as having the same capability.
Test three cases instead of two
A binary yes/no field can force an unknown situation into a false answer. Use at least three states: source supports the feature under stated conditions, source rules it out for the described case, and applicability is not established. These states distinguish missing information from negative evidence.
In a fictional example, a feature is described as available after a specified verification step. A researcher who has not observed the account’s verification status cannot mark the feature available for that account. The correct research state is that personal applicability is unknown. No private account information needs to be collected to make that methodological point.
Keep comparisons symmetric
If one provider’s feature is recorded with every condition while another is reduced to a broad promotional headline, the matrix is not comparing like evidence. Apply the same extraction rule to both. This does not require assuming that their terms match; it requires treating the evidence with the same discipline.
Record the document type as well. A general homepage, an agreement and an employer instruction may each answer a different question. A comparison that mixes them should explain the mismatch rather than presenting the resulting cells as equivalent proof.
Use a question when a verdict is unavailable
A useful outcome can be a precise verification question: ‘Which current program document establishes whether this feature applies in this context?’ That is more actionable than an unsupported yes. It also avoids turning independent editorial research into unofficial account support.
Before finishing a feature table, read every check mark as a full sentence. If the sentence sounds broader than the source, replace the symbol with a short conditional label or an unknown state. The table should help readers see the evidence structure, not hide it behind a cleaner design.
Continue the investigation
Sources and scope
- External official source: Corpay Prepaid: public product and issuer context — Name transition, PayCard/FlexCard distinction and conditions stated on the public page. Not an individual agreement or outcome study. Checked 2026-10-04.