How to assess a promo code's dates and evidence

ShopTools AI helps you find promo codes, but an offer card does not establish eligibility for your purchase. Check the source, dates and meaning of any testing labels. Each tells you something different.

Keep three dates separate

• The offer period comes from the retailer or another source. It does not remove restrictions on products, countries or accounts.
• A card's update date tells you when the record changed. It is not necessarily the date a code was successfully applied.
• The last confirmed result concerns a particular observation. Its date does not make another shopper's order conditions apply to yours.

A future expiry date does not establish that a code fits your cart. If the deadline is missing or conflicts with the retailer's description, clarify the terms before budgeting for that discount.

Understand what an attempt count records

Consider an illustrative label, not a ShopTools metric: Worked 4 of 7. It describes seven recorded attempts and four outcomes that the source called successful. It does not promise the next attempt will succeed.

Ask what counted as success. An accepted-code message and a measured reduction in the payable total are different events. Without that definition, even comparing two counters can be misleading.

The age and composition of the observations matter too. Repeating a test in one cart is not the same as testing different conditions. Missing context cannot reasonably be filled in with an assumed probability.

A precise percentage can still tell you little

A percentage without a sample size, time period or explanation can look stronger than its evidence. The same proportion in a small and a large sample does not carry the same weight.

Application history describes the past. A prediction for a new cart would need a separately validated model. Neither a polished badge nor the term AI turns one into the other.

A missing label is ambiguous as well. There may be too little data, older observations or results that did not pass a quality check. An empty label proves neither that a code is fake nor that it is suitable.

Match the terms, not just the rating

• Is this the right retailer and, on a marketplace, the right seller?
• Do the shopping country, product category and minimum spend match the offer?
• Is the code limited to new customers or a particular payment or delivery method?
• Can it be combined with the current promotion, or would replacing that promotion make the order more expensive?

Two shoppers can face different conditions on the same day. Another person's recent result may help you choose a candidate, but it does not replace the retailer's decision on your order.

A failed code is not necessarily a fake

A genuine code may have expired, belong to a different account or require different items. One rejected attempt is not enough to establish that a publisher deliberately misled you.

For a useful error report, keep the source, offer terms and retailer response without personal details. They help distinguish an outdated record from missing conditions or an application problem.

Before paying, look for a lower total for the same order and delivery, retained after recalculation. If the result could not be read, treat it as unknown rather than successful.