Feedback Request

You need this journey when you only learn how a purchase, delivery, onboarding or service experience went once a complaint arrives, or when surveys go to everyone at the same time through the same channel. Confirm that the completion event, open complaint and support case status, the date of the last feedback request and whether the user is in the app right now can all be read on one identity. If responses are low while support cases are rising, the problem is the experience, not the survey; fix the resolution process first and measure afterwards.

Incremental scored-feedback rate vs control

Asking is worth it only for the scored answers that would not have arrived on their own; an opened or clicked request earns nothing until a score comes back.

How it is counted: Among all completed experiences assigned to both groups at entry, the share that return scored feedback within the same follow-up window; feedback the control group sends unprompted counts too, so the gap is what asking adds. Whether the request was delivered or clicked does not change the denominator.

Also track

• Request sent rateEntrants found eligible and actually sent a request, as a share of all entrants; the intermediate step. The split between deferrals, recovery handoffs, waits for completion and ineligible segments shows whether the trigger fires too early or on the wrong audience.
• Response by channel path and reminderRead response rates for the in-app, push and email paths, and what the reminder adds, separately; the user's state decides the path, so the gaps are not a channel comparison.
• Score split and handoffsShares of high, middle and low scores among responses, and how many records reach each handoff's destination; use it to confirm low scorers actually reach human support.
• Permission loss and complaintsWatch push opt-outs, email unsubscribes and support contacts that answer the question with a complaint after the request; if they rise, the moment or the channel is wrong.

• Trigger and entryCompletion of a critical experience starts the journey, which first checks whether asking for feedback is appropriate. Those found eligible are randomly split into a journey group that gets the request and a control group that gets none from this flow.
• Who must not enterA customer with an open complaint or unresolved support case is not asked; the record moves to service recovery / issue resolution. A recent request defers a similar one; an ineligible segment does not enter. Each instance covers one completed experience; someone whose feedback has already arrived belongs to the Feedback Follow-Up journey.
• Timing and channelsWhen the experience is unfinished, it waits up to the maximum set per experience, then rechecks eligibility. A user in the app gets a short in-app rating; otherwise a push if permitted, otherwise email if available. After 2 to 3 days without a response, one reminder, then a final check 3 to 5 days later.
• Exit and handoffScore handoffs: high to public review / referral when appropriate, middle to one short improvement question, low to recovery / human support. An open complaint at entry or recheck goes to service recovery / issue resolution. Silence after the reminder, no channel, lost eligibility or an unfinished experience ends the journey; nothing more is sent.
• Frequency and consentAt most two touches per experience: one request and one reminder, which may use a different permitted channel when appropriate. Push only with permission, email only where it is available, in-app only during an active session, and a single channel path at any one time.

1. 1If the scored-feedback gap does not beat the control group, you asked for answers that would have come anyway; an opened request or a clicked reminder does not change that. A high response rate alone is not success either: customers handed to recovery over an open complaint never answered, yet they stay in the dissatisfaction record.
2. 2What you measured holds for this channel priority (in-app when present, else push, then email), one reminder and these waits. The review, the improvement question or the support resolution after the score handoff is the receiving journey's outcome; do not credit a record the Feedback Follow-Up journey closes to this one.
3. 3Enter each group's experiences and scored responses into the A/B test significance calculator when the gap stays unclear; if it is not significant, collect more completed experiences before you change the reminder channel.
4. 4Never ask the same question on two channels at once; the reminder may switch channels, but it is not a second copy. Do not build a filter that opens public review invitations only to high scorers; private feedback follow-up and public review collection serve different purposes.