Skip to main content
Check what each figure measures.

Who can do this?

Owners and Managers can review promotion details for their shop.

Where to do it

Open Promotions in the web app. Select the offer; check Redemptions and Promotion impact.

Read the headline figures

Redemptions includes uses through codes and offers applied without a code. It also counts offers already applied to orders that are not yet finished. Do not treat this number as a count of completed sales or unique customers. A total limit appears beside the count; otherwise, there is no cap. Minimum spend is an offer condition, not customer spending. Individual code counts cover only finished orders and exclude offers applied without codes, so they need not match the headline.

Read the impact report

The report can show separate results for Net revenue and Order count. Review each result with its comparison dates and last-updated date; revenue and order count can update on different days. An estimated effect is shown with a range. The number is not an exact amount caused by the promotion. Within normal variation means the analysis did not identify a measurable effect beyond ordinary variation for that metric. It does not prove that no customer used the offer. When a result cannot be measured, read the reason shown. There may be too little history, insufficient accuracy or another change that makes the comparison unclear.

Check timing and costs

The impact card shows discount cost, redemptions and the evaluated period alongside each result. Keep those dates in mind: the card is a periodic report, while the headline count can reflect more recent activity. If the report says it has not been evaluated yet, do not read that as zero impact. The report updates overnight; check again later and read any notice about missing or older data. No report is a promise of increased sales.

Compare Wingman results

An approved Wingman suggestion shows the same impact card in its Wingman details. Its short result label summarizes net revenue, not order count, so review both measurements. Decided suggestions stay in Wingman for 7 days after your decision. After that, check the result under Promotions > Promotion impact. While the offer runs, the summary shows Tracking above what a normal week would explain. for any measured effect, even a negative one, So far, within the range a normal week could explain. for other results, including unmeasurable ones, and Too early to say — the first measured night hasn’t landed yet. before any data. Check the detailed estimate’s sign, range and notices. The day counter tracks the measurement window; it does not confirm that the offer is active. Read the chart values; colour alone does not show whether an effect is positive. No effect we could confirm means the analysis could not establish a measurable effect. Couldn’t measure this one means there is no usable conclusion. Something else overlapped this window means other activity made the comparison unclear. None of these labels proves the promotion was unused. Wingman sometimes schedules several offers one after another. You can recognize one of these plan offers when, after approval, its promotion stays Draft until its start date, or when Promotions refuses to activate or change it. Its result label can reflect several overlapping offers measured together, and the card does not always say so. Do not attribute that result to one offer alone. If the label and detailed figures differ, compare their dates first. Give result feedback in Wingman while the suggestion is still listed; see the related article on answering Wingman questions. A reaction does not change the measured effect.

Platform and market notes

These web-app steps apply in Great Britain and Germany. Check shop, currency, comparison period and comparison shops before comparing figures.

Troubleshooting

If the impact card fails to load, use its retry action. A loading error is different from a report saying the effect cannot be measured. Contact Tapp support if results remain unavailable or the dates and figures appear inconsistent.