📋 Playbooks

Collecting feedback and Google reviews automatically after every order

Updated 2026-09-19 · 5 min read · By the Convo360 team

In short

Review collection on WhatsApp works by sending a 1-tap rating request shortly after an order or visit is complete, then branching automatically: a happy rating routes straight to your Google review link while a low rating routes to a human before it ever becomes a public complaint, with every response logged to Google Sheets or the CRM for tracking over time.

Key takeaways

Time the feedback ask right after the experience, not days later when memory has faded.
Use a simple 1-5 or 1-tap star rating before asking for detailed written feedback.
Route happy ratings straight to your Google review link; route unhappy ones to a human first.
Never ask an unhappy customer to leave a public review before you have tried to fix the issue.
Store every rating in Google Sheets or your CRM so patterns show up over weeks, not just individual replies.

Why timing decides whether you get any feedback at all

A feedback request sent two or three days after a purchase or visit gets a fraction of the responses one sent within a few hours does. The experience is still fresh, the customer remembers specific details, and a quick 1-tap prompt takes seconds to answer while they are still thinking about it. Waiting until a weekly or monthly batch email goes out loses most of this window entirely.

The right delay depends on the business: a restaurant or salon can ask within an hour of the visit ending, a delivered product usually works best a day after delivery once the customer has actually used it, and a service completion (installation, repair) works well immediately after the technician confirms the job is done.

The 1-tap rating step

Ask one simple question first, before anything else: a Buttons block with a 1-5 star scale or a simple 'How was your experience? Great / Okay / Not great' set of options. Keeping this to a single tap removes almost all friction, a customer who would ignore a text box happily taps one button. Save the detailed 'tell us more' question for after this first tap, and only ask it on the branch where it matters most.

Most Indian small businesses benefit most from a Google review link, since it directly affects local search visibility and is what new customers check first. Depending on your business, a marketplace rating (for a restaurant on a food delivery app) or an industry-specific platform may matter just as much. Where more than one platform matters, consider rotating which link you ask for, or offering a quick choice of platform via Buttons, rather than only ever directing happy customers to one place.

Building the branching flow on Convo360

You can build this entire branching flow on Convo360 without writing any code.

Trigger the flow with a Wait / Delay block timed after order completion, delivery confirmation, or appointment end.
Send a Buttons block asking for a quick rating (for example 1-5 stars, or Great / Okay / Not great).
Add a Condition block: high ratings (4-5, or 'Great') branch one way, low ratings (1-2, or 'Not great') branch another, and a middle rating gets its own softer follow-up.
On the happy branch, send your Google review link directly with a short thank-you message.
On the unhappy branch, route immediately to live-chat team inbox or an AI Reply step asking what went wrong, before anything about reviews is mentioned.
Log every rating and any written feedback to Google Sheets or Save CRM Lead for tracking.

Why unhappy customers should never see the review link first

Sending the same review-link message to everyone regardless of rating is a common and costly mistake. A customer who just had a bad experience, asked to leave a public review before you have addressed their issue, either ignores the request or leaves a negative review you could have prevented. Branching on the rating first, and only revealing the review link to genuinely happy customers, protects your public rating while still capturing every piece of feedback internally.

Routing a low rating to a human is only half the job; what happens next decides whether that customer stays. A quick, genuine acknowledgement, a specific fix or explanation, and where appropriate a gesture (a replacement, a partial refund, a discount on the next order) turns a bad experience into a recovered relationship far more often than silence or a generic apology template. Track how many low-rating conversations get a same-day human response, since a slow recovery response undoes most of the goodwill a fast one would have built.

What to do with a middle or lukewarm rating

A middle rating (3 out of 5, or 'Okay') deserves its own branch rather than being folded into either extreme. A short, non-defensive follow-up question, 'What would have made this a great experience?', often surfaces useful, specific feedback that a customer would not have volunteered unprompted. Route the answer to Save CRM Lead tagged for internal review rather than escalating it as urgently as a low rating, but do read these regularly, they often reveal small, fixable friction points before they become bigger complaints.

Asking for feedback after every single interaction, especially for a repeat customer who orders weekly, quickly turns into noise they start ignoring. Set a reasonable frequency cap, for example only asking once per customer per fortnight regardless of how many orders they place, using a Condition check against the date of their last feedback request, so the ask stays meaningful rather than routine.

Turning feedback into something you actually use

Collecting ratings is only useful if someone reviews the pattern, not just individual responses. Export or connect ratings to Google Sheets and look at trends weekly: is a specific outlet, product line or service type consistently rated lower than others? Reviewing Flow Logs alongside the ratings sheet can also show whether low ratings cluster around a particular time (late-night orders, a specific delivery partner) that points to an operational fix rather than a one-off.

Beyond individual recovery, aggregated rating data over a few months is useful input for decisions well beyond support: whether a new product line is landing well, whether a particular delivery partner needs replacing, or whether a specific outlet needs operational attention. Reviewing this pattern alongside your analytics dashboard turns a simple feedback bot into an ongoing, low-effort source of operational insight rather than just a review-generation tool.

Step by step

Decide the right timing for your businessSet the feedback request to fire shortly after the experience is genuinely complete, not on a fixed daily batch.
Build the 1-tap rating questionUse a Buttons block with a simple star or Great/Okay/Not great scale as the first question.
Branch by ratingRoute happy ratings to your Google review link and low ratings straight to a human or AI Reply for recovery.
Add a middle-rating follow-upAsk a specific, non-defensive question for lukewarm ratings to surface fixable friction.
Log every responseConnect the flow to Google Sheets or your CRM so ratings and comments are tracked over time.
Cap request frequencyLimit how often the same customer is asked for feedback to avoid the request becoming routine noise.
Review patterns weeklyCheck the ratings sheet and Flow Logs together to spot recurring issues by outlet, product or time.

Example conversation

Not great (tapped after a delayed delivery)

Sorry to hear that, could you tell me a bit more about what went wrong? I want to make sure we get this right, and I'll flag it to our team right away.

Automate your review collection on Convo360

Start free in minutes — no code, no credit card. Bring your own AI keys and pay no per-message markup.

Start 14-day free trial →See pricing

Frequently asked questions

Should I ask for a star rating or written feedback first?

Ask for a quick 1-tap rating first. It takes almost no effort to answer, and you can then decide, based on that rating, whether to ask a happy customer for a public review or an unhappy one for more detail.

Is it against WhatsApp's rules to ask for a Google review?

No, sharing a review link is allowed within a normal conversation. Keep it factual and optional rather than repeatedly pressuring a customer who has not responded, and only send it as part of a genuine post-purchase interaction, not an unrelated broadcast.

What if a customer gives a low rating but doesn't respond to the follow-up question?

Log the low rating regardless, since the number itself is useful data even without a written explanation, and consider a single gentle human follow-up call or message for genuinely low ratings if the issue seems serious.

Can this work for services with no fixed 'order', like a salon or clinic visit?

Yes, trigger the same flow after an appointment is marked complete using the appointments feature, timed for shortly after the expected visit end time, so a salon, clinic or repair visit gets the same 1-tap rating and branching logic as an order.

How do I stop the bot from asking for feedback too often?

Add a Condition check against the date of the customer's last feedback request before sending a new one, and set a reasonable minimum gap, such as two weeks, regardless of how many orders or visits happen in between.