Reputation Audit

Reputation & Review Management Audit — Phase 15

Executive Summary

Online reputation is a direct local pack ranking signal and an AI visibility gatekeeper. Businesses with 3 days | | % reviews with photos | ≥30% | 20–29% | 10–19% | <10% | | % reviews mentioning services | ≥40% | 25–39% | 10–24% | <10% | | Negative review recovery rate | ≥80% | 60–79% | 40–59% | <40% |

Review benchmarks by market size and niche: | Niche | Small Market (<100K) | Mid Market (100K–1M) | Major Metro (1M+) | |-------|---------------------|---------------------|-------------------| | Home services (plumbing, HVAC) | 2–4/mo, 25+ total | 4–8/mo, 50+ total | 8–15/mo, 100+ total | | Legal / professional | 1–2/mo, 20+ total | 2–4/mo, 40+ total | 4–8/mo, 75+ total | | Healthcare / dental | 2–4/mo, 30+ total | 5–10/mo, 75+ total | 10–20/mo, 200+ total | | Restaurant / food | 5–10/mo, 50+ total | 15–30/mo, 150+ total | 30–50/mo, 500+ total | | Automotive | 2–5/mo, 40+ total | 5–10/mo, 100+ total | 10–20/mo, 200+ total |

Veto: Rating <3.5 → maximum reputation score 40/100; effectively disqualified from local pack. Veto: Rating <4.0 → excluded from Google AIO recommendations for most service categories.

Review Response Analysis

• Owner response rate: [X%] of reviews responded to
• Average response time: [days]
• Response quality: Generic template / Personalized / Service-specific
• Negative reviews responded to: [X%]
• Tone of responses: Professional / Defensive / Empathetic

---

Step 3: Multi-Platform Review Inventory

| Platform | Review Count | Rating | Response Rate | Profile Complete? | Link | |----------|-------------|--------|---------------|-------------------|------| | Google | | | | | | | Yelp | | | | | | | Facebook | | | | | | | BBB | | | | | | | Trustpilot | | | | | | | [Industry-specific] | | | | | | | [Industry-specific] | | | | | |

Industry-specific platforms by niche:

• Healthcare: Healthgrades, Zocdoc, WebMD, RateMDs
• Legal: Avvo, Martindale, Lawyers.com
• Home services: Angi, HomeAdvisor, Thumbtack, Houzz
• Restaurants: Tripadvisor, OpenTable, Grubhub
• Automotive: Cars.com, DealerRater, Carfax
• Hospitality: Booking.com, Hotels.com, Expedia
• Beauty/Wellness: Vagaro, StyleSeat, Mindbody

---

Step 4: Competitor Review Benchmarking

| Metric | Client | Comp 1 | Comp 2 | Comp 3 | Gap | |--------|--------|--------|--------|--------|-----| | Google review count | | | | | | | Google rating | | | | | | | Review velocity/month | | | | | | | % 5-star | | | | | | | Response rate | | | | | |

Findings:

• Is client above/below competitor average?
• What is the review count gap to close?
• Which competitor has the strongest review velocity?

---

Step 5: Sentiment & Content Analysis

Positive Review Themes

What do customers praise most? (Extract from actual reviews)

• [theme 1]: mentioned in X reviews
• [theme 2]: mentioned in X reviews
• [theme 3]: mentioned in X reviews

These are SEO opportunities — build content around what customers love.

Negative Review Themes

What complaints recur?

• [complaint 1]: mentioned in X reviews
• [complaint 2]: mentioned in X reviews

These are operational problems AND reputation risks. Flag for business improvement.

Keyword Presence in Reviews

Do reviews contain service keywords?

• "[primary service]": mentioned in X% of reviews
• "[location]": mentioned in X% of reviews

Service keywords in reviews help local pack rankings.

---

Step 6: Review Generation Assessment

Does the business have a systematic review generation process?

| Question | Yes/No | |----------|--------| | Review request sent after every job/purchase? | | | Review request via SMS? | | | Review request via email? | | | QR code at point of sale/service? | | | Staff trained to verbally ask for reviews? | | | Review link easily accessible on website? | | | Review link on GBP? | | | Follow-up system for non-responders? | |

Assessment: Active system / Passive (sporadic) / None

---

Step 7: Negative Review Analysis

For every 1-star and 2-star review:

• Is there a response? Professional and empathetic?
• Is the complaint legitimate or fake/competitor-placed?
• Is the issue recurring (operational problem)?
• Has the issue been resolved?

Fake Review Detection

Signs of fake reviews:

• Posted in cluster (multiple on same day from accounts with no history)
• Reviewer has reviewed only this business (1-review accounts)
• Generic text ("Great service!" with no specifics)
• Reviewer located in different city

Recommendation if fake reviews found:

• Flag for removal via Google Business Profile reporting
• Respond professionally (do NOT engage aggressively)
• Document pattern for potential legal action if coordinated

---

Step 8: Review Marketing Assessment

Reviews as a marketing asset:

• Are top reviews displayed on the website (testimonials section)?
• Are review stars in Google Ads (seller ratings)?
• Are reviews used in social media content?
• Is review count mentioned in ad copy / GMB description?
• Aggregate rating schema on homepage and service pages?

---

Step 9: AI Review Impact Assessment (2025)

Reviews directly influence AI recommendation engines — not just traditional search.

Test protocol:

1. Search best [service] in [city] in Google AI Overviews → Does business appear? What rating/review count is displayed?
2. Ask ChatGPT: Who are the top [service] providers in [city]? → Is business mentioned?
3. Ask Perplexity: Best reviewed [service] in [city] → What review thresholds does it cite?
4. Check Google Maps AI summary (2025) — is business featured in AI-generated city/category overviews?

2025 AI Review Thresholds Observed:

• Google AI Overviews: typically features businesses with 4.3+ stars and 50+ reviews
• ChatGPT/Perplexity: cite businesses with established web presence + review mentions on trusted sources (Yelp, BBB, industry directories)
• Siri (Apple Maps): surfaces highest-rated options in category — requires Apple Maps verification

---

Step 9b: Brand Mention Scan for AI Visibility

Critical insight: Brand mentions correlate 3× more strongly with AI visibility than backlinks (Ahrefs December 2025 study of 75,000 brands). AI platforms cite businesses they "know" from mentions across the web — not just businesses with strong link profiles.

Platform Mention Correlation with AI Citations

| Platform | AI Citation Correlation | Weight | Why It Matters | |----------|----------------------|--------|---------------| | YouTube | ~0.737 (strongest) | 25% | AI systems (especially Gemini) heavily index YouTube. Videos, reviews, and tutorials mentioning the brand = high AI visibility. | | Reddit | High | 25% | Perplexity sources 46.7% of citations from Reddit. ChatGPT also weights Reddit discussions. Authentic brand mentions in subreddit discussions = strong signal. | | Wikipedia / Wikidata | High | 20% | ChatGPT sources 47.9% from Wikipedia. Wikidata entity = 3× more AI citations. The #1 entity signal for AI. | | LinkedIn | Moderate | 15% | Copilot (Bing) weights Microsoft ecosystem. Thought leadership posts and company page completeness improve Copilot citations. | | Domain Rating / Backlinks | ~0.266 (weak!) | 15% | Traditional backlinks still matter for organic SEO but are a weak predictor of AI citation. Brand mentions outperform links 3:1. |

Key takeaway: A business with 50 genuine brand mentions across YouTube, Reddit, and industry forums will likely have better AI visibility than a business with 500 backlinks but no platform presence.

Brand Mention Audit Protocol

For each platform, search "[Business Name]" and document:

| Platform | Search Method | Mentions Found? | Sentiment | Recency | |----------|-------------|----------------|-----------|---------| | YouTube | Search [Business Name] on youtube.com | Yes/No — [count] videos | Positive/Neutral/Negative | Last 6 months? | | Reddit | Search [Business Name] on reddit.com | Yes/No — [count] threads | Positive/Neutral/Negative | Last 6 months? | | Wikipedia | Search [Business Name] on en.wikipedia.org | Article / Mentioned / Absent | N/A | N/A | | Wikidata | Search [Business Name] on wikidata.org | Entity exists? Q-number? | N/A | N/A | | LinkedIn | Search [Business Name] on linkedin.com | Company page? Posts? | Positive/Neutral/Negative | Active? | | Quora | Search [Business Name] on quora.com | Yes/No — [count] answers | Positive/Neutral/Negative | Last year? | | Industry forums | Search niche-specific communities | Yes/No | Positive/Neutral/Negative | Recent? |

Brand Authority Score for AI (0–100)

| Component | Points | How to Score | |-----------|--------|-------------| | YouTube presence (channel exists + brand mentioned in videos) | 25 | 25 = active channel + external mentions; 15 = channel only; 5 = mentioned by others; 0 = absent | | Reddit presence (genuine discussions, not spam) | 25 | 25 = active contributor in relevant subreddits; 15 = mentioned positively; 5 = minimal mentions; 0 = absent | | Wikipedia/Wikidata entity | 20 | 20 = Wikipedia article; 15 = Wikidata entity; 10 = mentioned in other articles; 0 = absent | | LinkedIn company page (complete + active) | 15 | 15 = complete + regular posts + employee engagement; 10 = complete; 5 = basic; 0 = absent | | Cross-platform mention consistency | 15 | 15 = consistent NAP + brand description across all platforms; 10 = mostly consistent; 5 = some conflicts; 0 = major inconsistencies |

Brand Mention Action Plan

| Action | Impact (1–5) | Feasibility (1–5) | Priority | Effort | |--------|-------------|-------------------|---------|--------| | Create YouTube channel + publish 3 educational videos | 5 | 3 | 15 | 8–16 hrs | | Participate authentically in 2–3 relevant subreddits | 5 | 4 | 20 | 2 hrs/week ongoing | | Create Wikidata entity (if business has external coverage) | 4 | 4 | 16 | 2–4 hrs | | Complete + activate LinkedIn company page | 3 | 5 | 15 | 1–2 hrs | | Encourage customers to post YouTube review videos | 4 | 3 | 12 | Ongoing | | Answer Quora questions in business category | 3 | 4 | 12 | 1 hr/week | | Add sameAs schema linking all platform profiles | 4 | 5 | 20 | 30 min | | Set up brand mention monitoring (Google Alerts + Brand24) | 3 | 5 | 15 | 30 min setup |

---

Step 10: Reputation Recovery (If Needed)

If average rating < 4.0 or significant negative content:

Priority Recovery Roadmap: | Step | Action | Effort | Timeline | Impact (1–5) | Feasibility (1–5) | Priority | |------|--------|--------|----------|-------------|-------------------|---------| | 1 | Resolve operational issues causing negative reviews | 2–20 hrs | Immediate | 5 | 3 | 15 | | 2 | Set up SMS review requests via Podium/Birdeye | 2 hrs setup | Week 1 | 5 | 5 | 25 | | 3 | Respond to every existing negative review | 30 min/review | Week 1 | 4 | 5 | 20 | | 4 | Request removal of clearly fake reviews (GBP flag) | 15 min each | Week 1 | 3 | 4 | 12 | | 5 | Create suppression content (FAQs, About page, PR) | 4–8 hrs | Month 1 | 4 | 4 | 16 | | 6 | Implement Birdeye/Podium for systematic management | 4 hrs setup | Month 1 | 5 | 4 | 20 |

---

Step 10: Review Response Templates

Provide 3 customized response templates:

5-Star Response (Personalized): "[Customer name], thank you for taking the time to share your experience with [specific service mentioned]. We're thrilled [specific thing they praised]. [Business name] team loves serving the [city] community. See you next time!"

Negative Review Response (Empathetic): "[Customer name], we sincerely apologize this wasn't the experience you expected. We take feedback very seriously. We'd love to make this right — please contact us at [phone] so we can resolve this personally. — [Owner name], [Business Name]"

Neutral Review Response: "Thank you for your feedback, [Name]. We appreciate you choosing [Business Name]. If there's anything we can do to make your next experience a 5-star one, please let us know."

---

Priority Recommendations

Priority Matrix (Impact × Feasibility)

| Action | Impact (1–5) | Feasibility (1–5) | Priority Score | Effort | |--------|-------------|-------------------|----------------|--------| | Set up SMS review request system (Podium/Birdeye) | 5 | 5 | 25 | 2 hrs setup | | Respond to every unanswered review (positive + negative) | 5 | 5 | 25 | 30 min/batch | | Resolve operational issues driving negative reviews | 5 | 3 | 15 | Varies | | Add AggregateRating schema to homepage + service pages | 4 | 5 | 20 | 30 min | | Flag and report fake/competitor reviews via GBP | 3 | 5 | 15 | 15 min/review | | Train staff on verbal review request after service | 4 | 4 | 16 | 1 hr training | | Display top reviews on website (testimonials section) | 3 | 5 | 15 | 1–2 hrs | | Create review-optimized landing page with schema | 4 | 4 | 16 | 2–3 hrs | | Set up QR code for review requests (print + digital) | 3 | 5 | 15 | 30 min | | Build multi-platform review monitoring dashboard | 4 | 4 | 16 | 2 hrs setup |

Immediate Actions (Week 1)

1. Deploy review request system — Set up Podium or Birdeye SMS flow: trigger = job completed → SMS within 2 hrs → link to GBP review page → automated follow-up if no response in 48 hrs. Expected: 4–8 new reviews/month from month 1.
2. Respond to all unanswered reviews — Prioritize all 1-star and 2-star first (reputation recovery), then 5-star (engagement signal). Use personalized templates (not generic). Expected: response rate 100%.
3. Add AggregateRating schema — Implement JSON-LD on homepage + service pages. Use ratingValue, reviewCount, bestRating:5, worstRating:1. Validate at search.google.com/test/rich-results. Expected: star ratings appear in SERP snippets = +17–25% CTR.
4. Flag fake reviews — For any cluster of reviews from single-review accounts posted on same day → Report via GBP Manager → "Flag as inappropriate" → Document pattern for potential legal action.

Short-Term (Month 1)

1. Fix operational root causes — Identify top 3 recurring negative themes from review content → escalate to operations team → create service delivery improvement SOP.
2. Build suppression content — If damaging content appears in branded SERPs: create positive content (case studies, awards page, testimonials hub, PR mentions) to push negative results below page 1.
3. Expand multi-platform presence — Claim and optimize profiles on 2–3 industry-specific platforms (see niche list in Step 3). Coordinate cross-platform review asks.
4. Create review marketing assets — Export top 5-star reviews → design social media cards → post on Instagram/Facebook weekly. Use as trust signals in Google Ads copy.

Medium-Term (Months 2–3)

1. Run 90-day velocity sprint — Goal: close gap to #1 competitor review count within 90 days. Calculate gap: if competitor has 150 reviews and client has 60 → need 90 reviews → 30/month → intensify SMS campaign + personal outreach from owner.
2. Build review diversity — Aim for reviews that mention: specific services (40%+), location/neighborhood (25%+), staff names (15%+), specific outcomes (20%+). These keyword-rich reviews improve local ranking and AI citation likelihood.

---

Scoring

| Category | Weight | Score | |----------|--------|-------| | Google review count vs. competitors | 15% | /15 | | Average rating (target: ≥4.5) | 20% | /20 | | Review velocity (≥4/month for mid-market) | 15% | /15 | | Response rate (100% = perfect) | 15% | /15 | | Multi-platform presence (3+ platforms complete) | 15% | /15 | | Brand mention authority for AI (YouTube/Reddit/Wikipedia/LinkedIn) | 20% | /20 |

Veto: Average rating <3.5 → maximum score 40/100 regardless of other factors. Veto: Average rating <4.0 → flag as AIO exclusion risk; note in report.

---

Output

Write to {AUDIT_DIR}/reputation-findings.md with YAML frontmatter:

---
skill: local/reputation-audit
phase: 15
date: [YYYY-MM-DD]
business: [B

…

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [mshahiddigital](https://github.com/mshahiddigital)
- **Source:** [mshahiddigital/agentic-local-seo-audit](https://github.com/mshahiddigital/agentic-local-seo-audit)
- **License:** MIT

Install and usage instructions live in the source repository linked above.