Review Strategy
RecentSOP: Entity-Rich Reviews, Velocity Management, and NLP Analysis
Reviews are not just social proof -- they are entity-rich content that Google reads, analyzes with Natural Language Processing, and uses as direct ranking signals. This SOP treats reviews as one of the most powerful SEO assets in the Merlino system. A well-executed review strategy can move the needle faster than almost any other tactic.
"30 reviews in one day on a bare profile and the phone rang with a paid call 36 hours later. Reviews are rocket fuel." -- Mike Merlino
Why Reviews Matter for Brand Signals
Google extracts entities from review text using NLP. When a review mentions your business name, a specific service, a staff member's name, and a location -- Google builds those associations into your entity profile. Reviews are one of the few places where real customers generate entity-rich content about your business without you writing it.
What Google Extracts from Reviews
| Signal | Example | SEO Impact |
|---|---|---|
| Business Name | "Joe's Plumbing was great" | Entity recognition + brand signal |
| Service Keywords | "emergency drain cleaning" | Topic/category association |
| Staff Names | "Mike showed up on time" | Person entity linked to business entity |
| Location Mentions | "our home in South Tampa" | Geographic entity association |
| Positive Sentiment | "best plumber," "highly recommend" | Sentiment scoring for ranking |
| Related Services | "also fixed our water heater" | Service breadth signals |
Entity-Rich Review Content
The goal is reviews that naturally contain service keywords, location names, staff names, and positive sentiment -- without sounding scripted or fake.
What a Great Review Looks Like
"We hired Joe's Plumbing for an emergency drain cleaning at our home in South Tampa. Mike showed up within an hour, diagnosed the problem, and had it fixed the same day. Best plumber in Tampa -- highly recommend their drain cleaning and water heater services."
Why this review is powerful:
- Business name mentioned: "Joe's Plumbing"
- Primary service keyword: "emergency drain cleaning"
- Location mentioned: "South Tampa" and "Tampa"
- Staff name: "Mike"
- Additional services: "water heater services"
- Positive sentiment phrases: "best plumber," "highly recommend"
- Natural language -- not scripted
How to Guide Customers (Without Scripting)
Never Script Reviews
Never write reviews for customers, provide pre-written text, or use a review template. This violates Google guidelines and risks account suspension. Instead, guide the conversation naturally so customers include the right elements on their own.
Ask these questions when requesting a review to naturally prompt entity-rich content:
- "What service did we perform for you?" -- prompts them to mention the service keyword
- "Which of our team members helped you?" -- prompts them to mention staff names
- "Where is your home or business located?" -- prompts them to mention the area or neighborhood
- "Would you recommend us? What would you tell a friend?" -- prompts positive sentiment and brand mention
Delivery methods for review requests:
- In-person conversation at the end of a service call
- Follow-up text message with a direct review link
- Email follow-up within 24 hours of service completion
- Review card handed to the customer on-site
Running Reviews Through Google NLP API
Use the Google Natural Language API to analyze review text and verify that Google is extracting the right signals.
Step-by-Step NLP Analysis
- Go to the Google Cloud Natural Language API demo
- Paste the review text into the analyzer
- Check these three outputs:
Entity Salience
- Are your target keywords recognized as entities?
- Is the brand name the highest-salience entity?
- Are service keywords present as entities?
- Are location names detected as geographic entities?
Sentiment Score
- Overall sentiment should be strongly positive (0.7+ on a 0-1 scale)
- No negative sentiment phrases that could dilute the signal
- Mixed sentiment reviews still provide entity data but weaker ranking signal
Category Classification
- Does Google categorize the review text into the correct industry?
- Example: a plumbing review should classify under "Home & Garden > Home Improvement" or similar
Using NLP for Review Coaching
Run 5-10 existing reviews through the NLP API. Identify which reviews have the strongest entity extraction and sentiment scores. Use those as benchmarks -- share them with your team as examples of what great reviews look like (without scripting future ones).
Review Velocity and Consistency
The Velocity Principle
Google tracks review velocity -- how many reviews you receive over time. Sudden spikes followed by silence look unnatural. Consistent, steady review flow looks organic.
Velocity Targets
| Business Stage | Monthly Target | Weekly Pace |
|---|---|---|
| New listing (0-6 months) | 8-15 reviews/month | 2-4 per week |
| Established (6-24 months) | 15-25 reviews/month | 4-6 per week |
| Mature (2+ years) | 10-20 reviews/month | 3-5 per week |
The 30-Review Velocity Test
Real Results from the Mastermind
A member tested pushing 30 reviews in a single day on a brand-new bare GMB profile. Within 36 hours, the listing received its first paid call from a Google Maps search. While aggressive velocity like this carries risk, it demonstrates the raw power of review signals on fresh listings. For ongoing strategy, controlled steady velocity is safer.
Competitor Velocity Analysis
- Check the top 3 competitors in the map pack for your target keyword
- Count their total reviews and note the date range
- Calculate their monthly review velocity:
(Total reviews) / (Months since first review) = Average monthly velocity
- Set your target at 1.5x the highest competitor velocity
- Maintain this pace consistently -- do not burst and stop
Review Response SOP
Respond to every single review within 24 hours. Your response is another piece of content Google reads and extracts entities from.
Responding to Positive Reviews
Your response should naturally reinforce WHO/WHAT/WHERE:
- Thank the reviewer by name
- Mention the specific service performed (reinforces the keyword)
- Reference the location or service area
- Mention a related service (cross-sell signal)
- Invite them back or reference future needs
Example response:
"Thank you, Sarah! We're glad our drain cleaning team could get your kitchen sink flowing again quickly. South Tampa is one of our busiest service areas, and we're always happy to help neighbors in the community. If you ever need water heater service or any other plumbing work, Joe's Plumbing is just a call away!"
Responding to Negative Reviews
- Acknowledge the concern professionally and empathetically
- Do NOT argue, blame, or get defensive in public
- Take the conversation offline: "Please contact us at [phone] so we can make this right"
- Still include brand name and service reference naturally
- Follow up privately and resolve the issue
- If resolved, politely ask if they would consider updating their review
Never Ignore Negative Reviews
An unanswered negative review tells Google (and potential customers) that you do not care. A professional response shows accountability and keeps the entity signals flowing. Even negative reviews with professional responses contribute to review count and engagement metrics.
Varied Review Links to Enhance Engagement
Do not send every customer to the same review link. Vary the review platforms to build a diverse review profile:
| Priority | Platform | Link Type | When to Use |
|---|---|---|---|
| 1 | Google (GMB) | Direct GMB review link | 70% of review requests |
| 2 | Yelp | Yelp business page link | 10% of requests -- high DA, often ranks for brand name |
| 3 | Facebook page recommendations | 10% of requests -- social proof signal | |
| 4 | Industry-specific | HomeAdvisor, Avvo, Healthgrades, etc. | 10% of requests -- niche authority signals |
Why Varied Links Matter
- Google cross-references reviews across platforms to validate entity signals
- A business with reviews only on Google looks less established than one with reviews everywhere
- Yelp and Facebook reviews often rank on page 1 for brand name searches
- Industry-specific reviews build niche authority
Review Platform Priority by Industry
| Industry | Primary | Secondary | Tertiary |
|---|---|---|---|
| Home Services | Google, Yelp | HomeAdvisor, Angi | BBB, Nextdoor |
| Legal | Google, Yelp | Avvo, Justia | Martindale, FindLaw |
| Medical | Google, Yelp | Healthgrades, Zocdoc | Vitals, WebMD |
| Restaurant | Google, Yelp | TripAdvisor, OpenTable | Zomato, Foursquare |
| Automotive | Google, Yelp | CarFax, DealerRater | Cars.com, Edmunds |
Review Strategy Checklist
- [ ] Review request process established (text, email, or in-person)
- [ ] Direct Google review link generated and tested
- [ ] Varied review links set up for Yelp, Facebook, and industry-specific platforms
- [ ] Team trained on how to guide (not script) entity-rich review conversations
- [ ] Competitor velocity analyzed and target set at 1.5x
- [ ] Weekly review pace goal documented and tracked
- [ ] Response templates created for positive and negative reviews
- [ ] All reviews responded to within 24 hours (verify weekly)
- [ ] Monthly NLP API analysis of new reviews for entity extraction quality
- [ ] Review count and velocity tracked in a spreadsheet monthly
- [ ] No purchased or incentivized reviews (compliance check)
The Risks of Buying Reviews
Do Not Buy Reviews
Purchased reviews carry severe risks:
- Google can detect fake review patterns (same IP ranges, similar language, bulk timing)
- A single report from a competitor can trigger a Google review audit
- Review removal can happen in bulk -- losing 50+ reviews overnight
- GMB suspension is possible for repeated violations
- FTC guidelines prohibit fake or incentivized reviews without disclosure
Build reviews the right way: great service, consistent requests, natural language. It takes longer but the results are permanent and risk-free.
See Also
- GMB Optimization -- Reviews are a core component of GMB completeness
- Brand Identity -- Reviews reinforce the WHO/WHAT/WHERE framework
- Magic Blog Post -- Blog content creates topics customers can reference in reviews
- Press Releases -- PR mentions complement review-based entity signals