Unused Loyalty Reward

Put this flow in place when the loyalty report shows a growing share of rewards that sit redeemable and then expire untouched. You need to track, by reward identifier, that the reward is genuinely redeemable, which member holds it, the redemption event and its real expiry date where one exists, and you need push and email consent. If members cannot find how to redeem, or the reward fails to apply at checkout, fix that screen first; a reminder does not make an unusable reward usable.

Incremental redemption rate vs control

Running the reminder pays only if rewards in the messaged group get redeemed more often than those in the control group that hears nothing; a push or email click is not a redemption.

How it is counted: Of all rewards assigned at entry to either group, the share with a redemption event matched to that reward record within the same follow-up window from assignment; the difference between the two groups is what the reminders added. Delivery or opens do not change the denominator.

Also track

• Redemption before any messageShare of rewards already used at the first check; these needed no reminder, and the figure helps you judge the length of the first wait.
• Redemption per touchTrack redemptions after the push and after the email separately; if the email adds none, revisit the expiry-based wait that follows it.
• Contacts per personFor members with several rewards, the total number of reminders sent in the same period; the flow is built per reward, so the load on the person has to be watched on its own.
• Push opt-outs and unsubscribesMembers who turn off push or unsubscribe from email after the reminders are the cost side; a handful of redeemed rewards may not cover that loss.

• Trigger and entryAny valid loyalty reward sitting unused opens the journey. Rewards are assigned at random at entry to the journey group or to a control group that receives neither the push nor the email from this flow.
• Who must not enterRewards already used, expired, or not actually redeemable despite a sufficient-looking balance stay out. One reward per instance; while Reward Confirmation handles a freshly earned reward on the same membership this flow is suppressed, a brand-new membership is explained by Loyalty Program Activation, and this flow offers nothing new.
• Timing and channelsDay 3: a push goes out if the reward is unused. A second check 2 days later; if unused, an email with the reward, how to redeem it and its real expiry date, then a 3 to 7 day wait tied to that expiry and a final check. Never invent an expiry the reward lacks.
• Exit and handoffA redemption at any of the three checks ends the journey at once and cancels the pending message. No use at the final check closes the reward as unused; an expired reward is never carried on. No journey takes over; this one ends in its own two exits.
• Frequency and consentOne push and one email at most per reward, each channel only where its consent exists. When a member holds several rewards, keep the messages apart by reward identifier and cap the person's total contacts.

1. 1Redemption that does not beat the control group means members were going to use the reward anyway; higher clicks or opens do not change that. Decide only on the gap in redemptions matched to the reward record.
2. 2These numbers hold for these two touches, the 3 and 2 day waits and the final wait tied to the expiry date. Where Reward Confirmation is running on the same membership this flow is suppressed; credit a redemption in that period there, not to both. A sale of a new product belongs to Cross-Sell / Complementary Offer; this flow only follows a reward the member already holds.
3. 3With a small or unclear gap, enter each group's rewards and redemptions into the A/B test significance calculator; if it is not significant, do not count the journey as a win. Use the same follow-up window for both groups and bound it by the reward's validity; an expired reward cannot be redeemed.
4. 4If redemptions rise in both groups in the days before expiry, credit the date, not the message. When the date in the email differs from the real one on the reward record, fix the template before reading the result.