Skip to main content
Collecting feedback after a purchase is one of the highest-leverage automations you can run. It tells you which products customers love, surfaces fulfillment problems before they snowball into refund requests, and feeds a steady stream of social proof into your product pages and third-party review sites. This guide walks through a complete post-purchase feedback flow in Maestra Platform. The flow waits until the customer has actually had time to use the order, confirms they are still subscribed and reachable, and then sends a single, well-timed message asking them to rate the order or leave a review on a third-party resource (Google, Trustpilot, Yelp, a category-specific site, or your own product reviews).

Goal

Send a communication after delivery that asks the customer to rate their order or leave a review on a third-party resource. The flow is channel-agnostic — you can deliver the ask over email, SMS, Viber, mobile push, or web push. Pick the channel where your customer is most likely to respond. For most brands, email is the workhorse here because it gives you room for a star-rating widget, a written review prompt, and links out to external review sites.

Before you build the flow

Set up the automated campaign you plan to send from inside the flow. The campaign lives outside the flow itself — the flow just triggers it.
1

Create an automated campaign in your chosen channel

Build one campaign per channel you want to use: email, SMS, Viber, mobile push, or web push.
2

Add the Order parameter (optional)

If you want the message to reference specific order details — item names, images, the order number, the delivery date — pass the Order parameter into the campaign. The campaign template can then pull fields directly from that order.Alternatively, use the new constructor, which renders order content without requiring you to wire up the parameter manually.
3

Test the campaign once on its own

Send a preview to yourself before plugging the campaign into the flow. It is much easier to debug a broken template in isolation than inside a multi-step flow.
If you want to A/B test the channel or the ask itself (star rating vs. open-ended review, internal vs. external review site), build the variants as separate campaigns now. You can branch between them inside the flow later.

Flow structure

The flow has six pieces, in order:
  1. Trigger — Order status changed
  2. Delay — Give the customer time to actually use the product
  3. Subscription check — Confirm the customer is still reachable in the channel
  4. Send campaign — Deliver the feedback ask
  5. Frequency cap — Limit how often this flow runs per customer
  6. Activate — Turn the flow on
The sections below cover each piece in detail.

Step 1. Trigger: Order status changed

Use the Order status changed event as the flow trigger. Configure it to fire when an order transitions to the status that means “the customer now has the product in hand” — typically Delivered, Completed, or Fulfilled, depending on how your order pipeline is modeled.

How the trigger behaves

A few details about this event are worth knowing up front, because they shape how the flow behaves in edge cases:
  • It fires on entering the selected status, including on order creation. If an order is created already in the target status (for example, an in-store pickup recorded as Completed immediately), the flow still triggers.
  • Backdated orders trigger the flow if the order falls inside the relevance window of the trigger step and there are no later status changes that would supersede it. This matters if you import historical orders or correct fulfillment data after the fact.
  • All line items on the order must reach the target status. If a single line item is canceled, the flow will not trigger — unless you switch the trigger into “Any position transitioned” mode, which fires as soon as one item hits the status.
  • Line items do not have to transition at the same time. The flow waits until the last remaining item reaches the target status, then fires once.
  • You can cap how often the trigger fires per order. This prevents weird situations like an order bouncing between statuses and re-triggering the flow.
  • Status transitions only count from the moment the flow goes live. Transitions that happened before you activated the flow are not retroactively counted.
Choose the status that most reliably means “the customer has the product.” If your carrier integration writes back a Delivered status, use that. If you only have Shipped, you will need a longer delay in Step 2 to compensate for shipping time.

Step 2. Delay: let the customer use the product

Add a Delay block immediately after the trigger. The point of the delay is simple: a customer who got their order ten minutes ago cannot meaningfully review it yet. How long to wait depends on the product:
  • Consumables, fast-moving goods, food, beauty samples — 3 to 5 days is usually enough.
  • Apparel and accessories — 5 to 7 days, long enough for the customer to wear the item at least once.
  • Electronics, home goods, anything with a setup curve — 7 to 14 days.
  • Furniture, appliances, big-ticket items — 14 to 30 days.
Inside the Delay block, set send-time windows so the message cannot go out in the middle of the night. A safe default is to allow sends only between 9:00 AM and 8:00 PM in the customer’s local timezone. If the delay expires at 3:00 AM, the flow will hold the message until the window opens.
If you have multiple product categories with very different “time to first use,” consider splitting the flow into branches by category and using a different delay on each branch.

Step 3. Subscription check

Before sending, add a condition that checks the customer’s status in the channel you are about to use:
  • The customer has a valid subscription in the channel (email, SMS, Viber, mobile push, or web push).
  • The customer has a valid contact in that channel — a deliverable email address, a phone number that can receive SMS, an active push token, and so on.
If either check fails, exit the flow. Do not try to fall back to a different channel here unless you have explicit consent for that channel as well — burning subscription health for a review request is not worth it.
Skipping this check is the most common cause of bounce-rate spikes on post-purchase flows. Customers who unsubscribed between the time they placed the order and the time the delay expired must not receive the message.

Step 4. Send the campaign

Send the automated campaign you built in the “Before you build the flow” section. A few content patterns that work well here:
  • Lead with the ask, not the brand. “How are you liking your [product]?” outperforms “A message from [brand].”
  • Make rating one-click. Embed a star widget directly in the email so the customer can rate without opening a separate page. The click should deep-link them into the full review form with their rating pre-filled.
  • Split positive and negative responses. If the customer picks 4 or 5 stars, route them to your public review site (Google, Trustpilot, the App Store, your product page). If they pick 1, 2, or 3 stars, route them to a private feedback form that goes to your support team. This is the single biggest lever for both your review scores and your churn rate.
  • Include the order context. Showing the actual item the customer bought — image, name, order number — lifts response rates significantly over generic “rate your recent order” copy.
  • Don’t offer an incentive for a positive review. Incentives for any review (positive or negative) are fine on most platforms; incentives conditional on a 5-star rating violate the policies of most review sites and will get your reviews removed.

Step 5. Frequency cap

In the flow’s start block, set a frequency cap so a customer who orders three times in a month does not get three feedback asks in quick succession. A reasonable default is one feedback request per customer every 30 days, but tune this to your typical purchase cycle. If you sell something customers buy weekly (coffee, supplements, pet food), once every 30 days is right. If you sell something customers buy twice a year (outerwear, electronics), every 60 to 90 days is more appropriate so you are not asking on every single order.

Step 6. Activate

Turn the flow on. From this point forward:
  • Every order that transitions into the target status will enter the flow.
  • After the delay, eligible customers will receive the campaign.
  • The frequency cap will suppress repeat asks within the configured window.
Status transitions that happened before you activated the flow will not trigger it retroactively. The flow only sees events that occur from the activation moment onward.

What to measure

Once the flow has been running for a few weeks, look at:
  • Response rate — percentage of recipients who submitted a rating or review. Healthy benchmarks: 5–15% for email, 1–5% for SMS.
  • Average rating — both the trend over time and the distribution. A creeping average is the earliest signal of a quality or fulfillment problem.
  • Negative-feedback recovery rate — of customers who left a 1–3 star private response, how many did your support team reach and resolve. This is where feedback flows pay for themselves.
  • Review volume on external sites — if you are routing positive responses to Google or Trustpilot, your public review count should be climbing.
  • Unsubscribe rate on this flow vs. your marketing campaigns — if it is materially higher, your delay is probably too short or your frequency cap too loose.

Variations worth trying

  • Two-touch flow. Send a short “did it arrive okay?” message a day or two after delivery, then the full review ask a week later. The first touch catches fulfillment problems before they become bad reviews.
  • Repeat-buyer branch. Customers on their third or fourth order behave differently than first-time buyers. Consider a separate branch with a more familiar tone and a higher-trust ask (referral, UGC photo, video testimonial).
  • Category-specific copy. If you sell across categories, split the flow by product category at the trigger and use category-specific subject lines and questions.
  • Win-back loop. Pipe customers who leave a low rating directly into a support workflow with an SLA. Customers whose complaint is resolved within 48 hours have meaningfully higher retention than customers who are never contacted again.