By Hassan, Technical Lead at Cellbot
Published: 24 April 2025 · Fully reviewed: 26 August 2026
An automated review request should ask every eligible customer for an honest account of a genuine completed repair, without predicting or influencing the rating. Trigger from a verified completion event, use a neutral message, preserve channel and delivery evidence, stop after a defined response or limit, and route complaints separately without hiding the public-review option.
Do not offer a discount or gift for a Google review, send the link only to customers expected to be positive, ask staff to hit a rating quota or request specific praise. Google classifies those patterns as fake engagement or rating manipulation.
This guide covers the automation control. The broader `get more Google reviews` intent remains on the repair shop review strategy page while Search Console evidence is checked before any consolidation.
The policy-safe route in one view
| Stage | Required evidence | Stop or exception |
| Eligible completion | Genuine customer and completed repair or delivery | Open complaint, active rework, invalid contact or no appropriate channel |
| Neutral request | Same core ask regardless of expected sentiment | No rating condition, incentive or scripted praise |
| Direct destination | Current first-party Google review link or approved platform link | Broken, redirected or wrong-location link |
| Delivery record | Provider event and customer/repair relationship | Failed or undeliverable message |
| Limited follow-up | Explicit maximum and timing rule | Review observed, opt-out, reply, complaint or limit reached |
| Public response | Review content, policy and privacy check | Personal data, active dispute or safety issue needs private handling |
| Measurement | Eligible, sent, delivered, clicked, reviewed and opted-out kept distinct | No invented attribution or conversion claim |
Download the review request control register. Configure one row per location and channel so link, sender, timing and ownership remain reviewable.
1. Define an eligible repair
Start from a completion event the shop can prove. For a walk-in job, that may be collected and paid or otherwise closed under the shop's process. For mail-in work, carrier delivery may be more meaningful than dispatch.
Eligibility should not be a satisfaction score. It can legitimately exclude:
- a test, duplicate or staff record;
- an invalid or undeliverable contact;
- a customer who has opted out of the chosen communication where applicable;
- a repair not yet collected or delivered;
- a request already sent within the shop's frequency limit; or
- an open complaint, refund or active warranty rework where a routine message would be inaccurate or insensitive.
The last exclusion must route the case to customer care, not permanently suppress the customer's ability to review. Do not send a private satisfaction survey first and reveal the public review link only after a positive answer. Google expressly prohibits selectively soliciting positive reviews.
Connect the event to the repair shop automation control loop and preserve the customer, device and job relationship described in the CRM guide.
2. Use a neutral request
A good request identifies the shop and completed service, asks for an honest review, gives the direct link and makes no promise or pressure.
Example:
Hi first name], thank you for choosing shop] for your recent repair. If you would like to share your honest experience, you can leave a Google review here: direct link]. If you need help with the repair, reply to this message or contact support route].
Do not use:
- “Leave us five stars”;
- “Help technician] hit their review target”;
- “Show this review for 10% off”;
- “If you were happy, review us; if not, message privately”; or
- generated wording that tells the customer what to mention.
Google's Maps contribution policy permits asking for content based on a genuine experience without incentives or influence. It prohibits paid or discounted reviews, selective positive solicitation, rating pressure and requested content.
3. Use the current direct link
Google's review link and QR-code instructions explain how a verified business can copy the link or download a QR code from a computer browser. Store the link by real location and test it outside the manager account.
Check that it opens the intended profile rather than a similarly named branch. Retest after a move, rebrand, profile reinstatement or merge. Keep the last test date and owner in the register.
A QR code at the counter can reduce friction, but do not require a customer to review while staff watch or pressure them to include a name or phrase.
4. Choose the channel from evidence and permission
SMS, email and WhatsApp have different costs, delivery evidence and customer expectations. No honest benchmark guarantees one channel will outperform another for every shop.
Use a channel only when the shop has a documented basis and the message complies with current rules and provider terms. The ICO's April 2026 electronic-mail marketing guidance covers email, text and similar messages under PECR and data-protection rules. Whether a particular review request is direct marketing depends on its content and context; obtain appropriate advice rather than relabelling a promotional campaign as service communication.
Keep:
- the customer and repair that created eligibility;
- message purpose and template version;
- channel and sender identity;
- sent, delivered, failed and replied events;
- opt-out or objection state where relevant; and
- the policy and legal-basis review date.
Do not automatically switch a failed SMS to WhatsApp or email unless the second channel is appropriate and configured for that purpose.
5. Set timing and stop rules
The right delay is long enough for the customer to test the repair and short enough for the event to remain clear. It depends on repair type, handover method, promised testing and customer time zone. Do not publish a universal “best” hour without a shop dataset.
Test at least:
- a collection just before closing;
- mail-in delivery on a weekend;
- a repair reopened after collection;
- a refund or complaint opened before the timer fires;
- duplicated completion events;
- an invalid destination;
- a customer who already reviewed; and
- a follow-up limit reached.
Use one request initially. Add a single reminder only if delivery and customer-response evidence justifies it. Stop after review attribution where available, a direct reply, an opt-out, a complaint, a permanent failure or the configured maximum.
The system should never continue merely because it cannot detect the review. Attribution is imperfect, so frequency caps and customer-level history remain necessary.
6. Do not build a negative-review intercept
Review gating usually asks for private feedback first, then gives satisfied customers the public link and keeps dissatisfied customers in a private route. That produces a biased sample and conflicts with Google's prohibition on selectively soliciting positive reviews.
Offer support in the same neutral request, but keep the public review link available regardless of expected sentiment. If a customer reports a problem, respond promptly and route the repair through the shop's dispute-handling process. Do not make assistance conditional on changing or removing a review.
The CMA's fake-reviews guidance collection reflects the UK rules introduced under the Digital Markets, Competition and Consumers Act. Fake reviews and concealed incentivised reviews are banned practices. This operational guide is not legal advice; review the current guidance and obtain advice for your workflow.
7. Respond without exposing the repair record
A public reply is for the reviewer and future readers. It should not reveal device identifiers, contact details, passcodes, payment records or facts that identify another person.
Use a four-part structure:
- thank the reviewer for the feedback;
- acknowledge the experience or concern without admitting an unverified fact;
- state the general resolution route; and
- invite direct contact through a named channel.
For example:
Thank you for the feedback. We are sorry the collection experience did not meet expectations. We would like to review the job record with you; please contact role/channel] and quote your repair number privately.
Do not argue line by line in public or paste internal notes. Report a review only when it appears to breach the platform policy, not because it is critical.
Measure the workflow honestly
Keep each denominator visible:
| Measure | Calculation | Limitation |
| Eligibility rate | Eligible completed repairs ÷ completed repairs reviewed | Depends on a consistent completion definition |
| Send rate | Requests accepted by provider ÷ eligible repairs | Provider acceptance is not delivery |
| Delivery rate | Delivered requests ÷ sent requests | Delivery receipts vary by channel |
| Click rate | Unique tracked link visits ÷ delivered requests | Shared devices and privacy controls affect tracking |
| Observed review rate | Attributable new reviews ÷ delivered requests | Google does not identify every request source |
| Opt-out or objection rate | Relevant objections ÷ delivered requests | Interpretation depends on message purpose and channel |
| Complaint leakage | Complaints first discovered through a public review ÷ reviews | A high figure indicates an aftercare gap |
Do not claim that automation caused every new review. Organic reviews, counter QR codes and delayed customer action can overlap. Keep an “unknown source” state.
A worked measurement example
This is illustrative, not a Cellbot customer result.
Suppose the shop reviews 120 completed repairs in a month. Ten are test, duplicate, undeliverable or already over the frequency limit, leaving 110 eligible. The provider accepts 108 messages and confirms 100 deliveries. Six review-link visits are observed and four reviews can be conservatively attributed.
Report:
eligibility: 110 ÷ 120 = 91.7%;
send acceptance: 108 ÷ 110 = 98.2%;
confirmed delivery: 100 ÷ 108 = 92.6%; and
attributable review rate: 4 ÷ 100 = 4%.
Do not report a 4% result as a universal benchmark or call the other 96 deliveries failures. The request may have influenced an untracked later review or may simply have been ignored.
A 14-day implementation test
Days 1 to 3: policy and data
Confirm the genuine completion event, eligibility rules, approved channels, direct location link, message wording, owner and maximum frequency. Review current Google, CMA, ICO and provider terms.
Days 4 to 6: shadow mode
Generate the proposed eligible list without sending. Inspect duplicates, open rework, wrong locations and channel status. Correct the workflow before messages go live.
Days 7 to 10: capped release
Send to a small, ordinary sample of eligible completed repairs. Do not hand-pick happy customers. Monitor delivery, replies, complaints, opt-outs and wrong-link incidents each day.
Days 11 to 14: reconcile and decide
Compare eligible, sent, delivered, clicked and attributable reviews. Record policy incidents separately. Keep, change, pause or retire the rule with a named reason.
How Cellbot fits
Cellbot can trigger a review request from the completed repair workflow and retain customer, job and message context. Current paid-plan scope is shown on the Cellbot pricing page.
Cellbot publishes this guide and sells the workflow being discussed. Test the direct link, eligibility, suppression, delivery evidence, export and pause control before buying. A shop with very low volume can use a manually sent neutral template and the same register without automation software.
Automated review request FAQs
Can a repair shop offer a discount for a Google review?
No. Google's policy prohibits incentives such as payment, discounts or free goods or services in exchange for a review, revision or removal. UK rules also ban fake reviews and concealed incentivised reviews.
Can I ask only satisfied customers?
No. Google prohibits selectively soliciting positive reviews. Use neutral eligibility based on genuine completed work and operational exclusions, not predicted sentiment.
Should I send SMS, email or WhatsApp first?
Choose from the shop's permissions, customer expectations, delivery evidence and cost. Test one compliant channel on a normal sample. Do not rely on generic open-rate claims.
How soon after collection should the request send?
Use a delay that gives the customer a fair chance to test the repair, respects local time and stops for rework or complaints. Measure the shop's own outcomes; there is no universal best hour.
Can AI write review replies?
It can draft, but a person should verify facts, tone, privacy and remedy before publication, especially for disputes or low ratings. Do not let a draft expose the repair record.
How many reminders should I send?
Start with none. Add at most one only after reviewing delivery and response evidence, and keep a customer-level frequency cap. More reminders can create complaints without improving trustworthy reviews.
Sources, search evidence and update note
This guide was fully rebuilt on 26 August 2026. It removes unsupported timing, channel-open, conversion, growth and Cellbot-performance figures and removes the former negative-review intercept recommendation.
Fresh UK DataForSEO data recorded 10 monthly searches for “automated review requests”, informational intent and £12.31 CPC, with the provider record updated 23 August 2026. The broader “get more Google reviews” term recorded 90 monthly searches and difficulty 3; that broader intent remains assigned to the strategy URL. The live repair-specific automation results did not show this Cellbot article in the sampled organic listings. Exa was used for semantic source and competitor discovery, not rank measurement.
Primary references:
- Google: prohibited and restricted Maps content, checked 26 August 2026
- Google: create a review link or QR code, checked 26 August 2026
- CMA: reviews and endorsements guidance collection, checked 26 August 2026
- ICO: electronic-mail marketing guidance, updated 28 April 2026
Continue with the local SEO evidence plan, the review strategy guide or the repair automation guide.
