Case Study

Reviews Automation:
From Feedback to Google & Facebook

How I built a fully automated review collection system across 7 interconnected GoHighLevel workflows — routing customer sentiment, generating AI responses, and handling follow-ups without a single manual step.

Type
CRM Automation · AI Integration
Industry
Mobility Rentals
Tools
GoHighLevel · Make · OpenAI
Workflows Built
7 Interconnected Stages

Reviews were falling through the cracks — and no one had a system to catch them.

After a rental order is completed, there's a window where the customer's experience is still fresh. That window is the best time to ask for a review. Miss it, and the moment is gone. Chase it badly, and you look desperate or disorganized.

The problem wasn't that reviews weren't being requested. It was that the whole process depended on someone remembering to do it at the right time, saying the right thing, routing the right response, and following up if nothing came back. All manually. All on top of everything else a busy operation already demands.

"Negative feedback needs a human-aware response within hours. Positive feedback needs to turn into an actual Google or Facebook review before the customer forgets they even had a good experience. Neither of those things happen reliably when someone has to remember to make them happen."

Beyond timing, the routing itself was a problem. A negative review and a positive review require completely different responses. Getting that wrong — sending a generic follow-up to someone who just had a bad experience — makes a recoverable situation worse. The system needed to think, not just send.

Build a system that listens, decides, and responds — without being told to each time.

The goal was a review pipeline that could handle the full range of post-rental outcomes on its own: ask for feedback, read the sentiment, route accordingly, generate contextually appropriate responses using AI, follow up on silence, and know when to stop. No manual triggers. No one watching the inbox. Just a system doing exactly what a thoughtful person would do, at the right moment, every time.

The architecture ended up spanning 7 GoHighLevel workflows, a Make scenario handling AI message generation and external review posting, and OpenAI powering the response logic for negative feedback. Each stage hands off cleanly to the next. The whole thing runs as one connected pipeline from the moment a rental is marked complete.

The stack that made it possible.

GoHighLevel Make OpenAI (GPT-4) Facebook Reviews Google Reviews SMS Automation
7-Workflow System Overview
#WorkflowWhat it does
1 SMS Evaluation Sends the initial 1–5 rating request, monitors replies, and routes non-responders to drip
2a Negative Review Alerts team, offers callback via AI-analyzed reply, follows up or closes based on response
2b Positive Review Removes from eval flow, sends AI-generated review request, routes to chat loop
3a Review Chats 1 Sends AI response from Make, waits 3 days, loops to Chats 2 or escalates to Follow Ups
3b Review Chats 2 Mirrors Chats 1 logic — processes reply, generates AI response, loops back to Chats 1
4 Follow Ups 3-stage re-engagement sequence for non-responders; adds Conversation Ended after attempt 3
5 Conversation Stopper Clean exit — tags contact and removes from all active review workflows

The full flow, stage by stage.

1
SMS Evaluation: The review request goes out automatically
When an opportunity moves to the Asked for Review stage, the pipeline begins. All previous review fields are cleared first to prevent stale data from interfering. A safeguard checks whether the contact has a Conversation Ended tag — if they do, no message goes out. If not, the customer receives a 1-to-5 rating SMS. The system waits one day for a response. If nothing comes back, a follow-up message goes out automatically. Contacts who don't respond to either message are routed to a separate drip campaign rather than being chased indefinitely.
2
Sentiment Routing: Negative and positive responses go different directions
When a customer replies, Make processes the rating and determines the evaluation sentiment. From there, the pipeline splits into two completely separate tracks — one for negative experiences, one for positive ones. Each track has its own logic, its own messaging, and its own outcome. The same workflow doesn't handle both.
Sentiment Routing Flowchart
Customer Replies with Rating (1–5)
Evaluation
Sentiment?
Negative
Negative Review Track
  • Team alerted immediately
  • Empathetic SMS sent
  • AI determines callback intent
  • Yes → team call / No → follow-up
Positive
Positive Review Track
  • Removed from eval workflow
  • AI-generated review request sent
  • Chats 1 → 2 loop begins
  • FB / Google review guided
Negative Sentiment
  • Internal team alert fires immediately
  • Empathetic follow-up SMS sent to customer
  • AI determines if a callback is requested
  • Response handled with appropriate path
Positive Sentiment
  • Contact removed from evaluation workflow
  • AI-generated review request SMS sent
  • Conversation loop begins for FB/Google review
  • Follow-ups handled across chat stages
3
Negative Review Handling: AI-assisted triage without losing the human element
When sentiment is negative, the internal team is notified right away. A few minutes later, the customer receives a message asking if they'd like someone to call them about their experience. Their reply is analyzed by GPT-4, which outputs a clean Yes or No. If they want a callback, the team is alerted and the conversation closes with a confirmation. If they decline, the system follows up asking what could have been improved, waits one day for a response, and closes with a thank-you. No valid AI output results in a Conversation Ended tag and a clean exit — no customer gets stuck in limbo.
4
Positive Review Conversation Loop: AI keeps the dialogue going until a review lands
Positive responses trigger a multi-stage conversation loop designed to guide the customer toward leaving a public review on Facebook or Google. Make's FB/Google Review Assistant generates personalized responses using the customer's own message as context. The AI reply is stored in a custom GHL field and sent as an SMS. After each exchange, the workflow checks whether the customer has already left a review before sending anything further — preventing unnecessary messages to someone who already took action.
Chat Loop Structure
  • Review Chats 1: Initial AI response sent, conversation monitored for 3 days.
  • Review Chats 2: If the customer replies again, the loop continues. Each reply loops back to Chats 1.
  • Already reviewed check: Runs before every message — no one gets asked twice.
  • Conversation history: Every AI response is timestamped and logged in the CRM record.
5
Follow-Up Sequence: Three structured reminders before the conversation closes
Contacts who don't reply after the initial AI message enter a structured follow-up sequence. Each stage sends a new review request, waits two days, and checks for a response. If they reply at any point, they're moved back into the active chat loop. After three failed attempts, the system adds a Conversation Ended tag, removes the contact from the workflow, and stops. No customer gets a fourth message. No one is contacted indefinitely.
6
Guardrails Throughout: Tags and field checks protect every transition
At every stage, the system checks before it acts. The Conversation Ended tag stops messaging at any point in the pipeline. The already-left-a-review tag prevents duplicate outreach. The ChatGPT custom field is cleared after each use so a new AI response never overwrites one that hasn't been sent yet. Internal notifications fire at multiple points so the team always knows what's happening, even as the automation handles the execution.

What it actually changed.

Before the system, collecting reviews meant someone on the team had to remember to follow up after every completed rental, write a message, send it, check for replies, route the response appropriately, and follow up again if nothing came back. All of that happened manually — or didn't happen at all.

After, the pipeline runs on its own from start to finish. Positive experiences get guided toward public reviews through a conversation that feels personal because the AI is generating contextually appropriate responses, not sending a canned template. Negative experiences get triaged with empathy and routed to the right outcome without anyone having to manually intervene unless a callback is actually requested.

7
Interconnected workflows covering the full review lifecycle from request to resolution
2
Separate tracks — positive and negative — each with their own logic and outcomes
3
Structured follow-up attempts before the system closes a conversation cleanly
0
Manual steps required between rental completion and review request delivery
Owner Involvement: Review Collection Process
Before
Manual end to end
After
Callback alerts only
Manual involvement reduced to exception handling only

A review process that runs like a conversation, not a broadcast.

What made this system worth building wasn't just the automation — it was the decision logic underneath it. A lot of review tools blast the same message to every customer and call it done. This one reads what the customer actually says, decides what that means, and responds accordingly. A negative experience doesn't get a generic "thank you for your feedback." A customer who already left a review doesn't get asked to do it again.

That's the difference between automation that runs and automation that thinks. The goal was always the second one.

🤖
AI that reads context
GPT-4 analyzes negative replies to determine intent. Make generates personalized review request messages based on what the customer actually wrote.
🔀
Sentiment-aware routing
Positive and negative experiences never hit the same workflow. Each track is built for its specific outcome — nothing gets misrouted.
🛡️
Tags that protect the pipeline
Conversation Ended and already-left-a-review tags act as gatekeepers at every transition. No one gets messaged when they shouldn't be.
🔔
Team stays in the loop
Internal notifications fire at key moments — negative reviews, callback requests, unresolved follow-ups. The team knows what's happening without having to check.