ReviewShield

Case Study: Plumbing Company Removes Fake Review Cluster and Recovers Local Pack Ranking

Marcus Reyes, Reputation Specialist · June 20, 2026 · 6 min read

Key Takeaways

  • 6 fake reviews in 36 hours dropped the average from 4.7 to 3.9 stars — equivalent to a visible local pack ranking drop.
  • 5 of 6 reviews were removed; 1 was a genuine complaint left in place.
  • Full removal timeline: 6 weeks. Ranking recovery: 3 months to return to pre-attack position.
  • The company built a generation system concurrently — new positive reviews accelerated rating recovery.

In March 2026, a Chicago-area plumbing company with a strong review profile was targeted by what appeared to be a coordinated fake-review attack. Six one-star reviews appeared over 36 hours, dropping their average from 4.7 to 3.9 stars — a shift visible enough to affect their local pack position within days.

This case study covers the full arc: recognizing the attack, working with ReviewShield to pursue removal, the timeline, and what the company did to protect themselves going forward.

Key Takeaways

  • 6 fake reviews in 36 hours dropped the average from 4.7 to 3.9 stars — enough to impact local pack ranking.
  • ReviewShield removed 5 of 6 reviews; the 6th was a genuine complaint handled by response.
  • Full removal timeline: 6 weeks. Ranking recovered to pre-attack levels within 3 months.
  • Concurrently building a generation system accelerated recovery — new positive reviews diluted the remaining complaint.

Starting situation and the attack

Before the attack (early March 2026):

MetricEarly March 2026
Total reviews87
Average rating4.7 stars
Local pack position#2 for "plumber Chicago [neighborhood]"
Monthly new reviews4–6 (spontaneous, no ask system)

The company had built their 87-review profile organically over 6 years without a systematic ask process — primarily through spontaneous reviews from satisfied customers. A solid profile, but fragile: no ongoing generation meant they were one bad-review cluster away from a visible rating drop.

The attack (March 14–15, 2026):

Over 36 hours, six one-star reviews appeared. Warning signs were immediate:

  • All 6 appeared between Thursday evening and Saturday morning
  • 5 of 6 reviewers had Google accounts created within the past 2 weeks
  • None of the reviewer names matched any customer in the company's ServiceTitan records
  • Review content was generic ("terrible," "no-show," "ripped me off") with no job-specific details
  • One review mentioned "the receptionist named Jennifer" — the company has no receptionist named Jennifer

Impact:

MetricMarch 16, 2026 (post-attack)
Total reviews93 (87 real + 6 fake)
Average rating3.9 stars (down from 4.7)
Local pack positionDropped to #5–7 within 72 hours

The ReviewShield removal process

Week 1: Evidence gathering

The owner contacted ReviewShield immediately. Evidence compiled:

  • Screenshots of all 6 reviews with reviewer profiles
  • ServiceTitan records showing no matching customers for any of the 6 reviewer names, dates, or addresses
  • Account creation dates for each reviewer (all within 2 weeks of posting)
  • Note about the "Jennifer" reference confirming at least one review was fabricated
  • Screenshots of the company's other Google reviews for comparison (to establish the anomalous pattern)

Week 2: Cases submitted

ReviewShield submitted removal requests for all 6 reviews, citing:

  • Fake reviews from non-customers (no customer record, fabricated details)
  • Coordinated inauthentic behavior (cluster of reviews from new accounts within 36 hours)
  • Specific content violations (factual inaccuracies, fabricated employee names)

Weeks 3–4: First removals

The 3 clearest cases — accounts with no review history, created within days of posting, review content with factual errors — were removed within 3 weeks of submission. Average rating improved from 3.9 to 4.3 stars.

Weeks 5–6: Remaining cases

Two remaining reviews required escalation beyond the initial report — Google's initial automated review found no violation. ReviewShield submitted additional documentation including the account creation dates and the pattern evidence (coordinated timing). Both reviews were removed in week 6. One review remained: a review from an account with longer history, where the customer claimed a no-show for an estimate that couldn't be definitively disproven (no ServiceTitan record for an estimate call that may have been informally scheduled). This review was responded to professionally and left in place.

Final removal result: 5 of 6 reviews removed. Average rating after removals: 4.6 stars (down 0.1 from pre-attack due to the remaining legitimate complaint).

What they built concurrently: the generation system

While the removal process was underway, the company built a review generation system from scratch. Previously, their 4–6 monthly reviews were entirely organic. Post-attack, they implemented:

  • ServiceTitan Marketing Pro automation (they were already on ServiceTitan, upgraded to Marketing Pro)
  • In-person ask by technicians on all service calls and installs
  • Response habit: owner responds to every review within 24 hours

Generation results over 3 months post-attack:

MonthNew reviewsRunning totalAverage ratingLocal pack
April 2026131014.4 ⭐#3–4
May 2026151164.5 ⭐#2–3
June 2026161324.6 ⭐#1–2

The combination of removals plus new positive volume accelerated rating recovery significantly. The remaining genuine 1-star complaint, which would have been visible at 87 reviews, became much less impactful at 132 reviews with a consistent generation rate.

What the company changed permanently

1. Review generation system: The attack exposed the fragility of relying on organic reviews alone. A profile built over 6 years with 87 reviews is resilient to normal complaint patterns but vulnerable to coordinated attacks. An ongoing generation system (15+ reviews/month) builds the volume that makes any single attack less impactful.

2. Weekly review monitoring: The attack went unnoticed for 12 hours before a customer called to mention it. The company now has GBP notifications enabled and checks for new reviews every morning.

3. Competitor monitoring: They suspect the attack came from a direct competitor based on timing (immediately after a bid loss) and the geographic pattern of the fake accounts. They now track their top 3 local competitors' review activity monthly.

4. ReviewShield on retainer understanding: The company knows that if a future attack occurs, they have an established relationship and process for removal. The evidence-gathering process (ServiceTitan records, account screenshots) is now documented as a playbook.

Cost summary

ItemCost
ReviewShield removal (5 reviews)$2,495 (pay-on-removal)
ServiceTitan Marketing Pro (ongoing)$200/month
Lost leads during 72-hour ranking dropEstimated $1,500–$3,000 in call volume
Total impact of attack + recovery~$4,000–$5,500

The owner's view: "The removal cost was the right call. We had a 4.7-star profile we'd built over 6 years. Letting fake reviews sit at 3.9 indefinitely would have cost us far more in ongoing lead loss than the removal fees. And now we have a system that makes future attacks less damaging."

Got a fake review killing your jobs?

Paste it into our free checker: find out in seconds if it violates Google policy and can be removed. No account needed.

Check a review free

Fake review attacks happen to plumbing companies, HVAC companies, roofers — any contractor who is visible enough to be a competitive threat in their market. The defense is a two-part system: a generation engine that keeps building volume (making any attack less impactful per review) and a removal process for when fake reviews actually violate Google's policy. ReviewShield handles the removal side on a pay-on-removal basis. The generation side — ask every customer, every job — is yours to build.

FAQ

How did the company identify that the reviews were fake?

Several signals indicated coordinated fake reviews: 6 one-star reviews appeared within 36 hours (well outside normal complaint patterns), all reviewers had accounts created within 2 weeks of posting, none of the reviewer names or phone numbers matched any record in the company's ServiceTitan CRM, and review content was generic with no job-specific details. One review mentioned a service the company didn't even offer.

How long did the removal process take?

The full removal process took approximately 6 weeks from the initial ReviewShield case submission to all 5 removed reviews being taken down (1 review was found to be a genuine complaint and was left in place). Individual reviews came down at different times — the clearest cases (fake accounts, no customer record) were removed in 2–3 weeks; a more ambiguous case took the full 6 weeks with a successful escalation.

Did the company's ranking recover after the reviews were removed?

Yes — average rating recovered from 3.9 to 4.6 stars within a week of the final removals. Local pack ranking, which had dropped from #2 to #5–7 during the attack, recovered to #2 within 6 weeks and to #1 within 3 months. The recovery was accelerated by the generation system they built concurrently, which added 12–15 new positive reviews per month to both improve the average and signal continued engagement.

Reputation Specialist

Marcus Reyes

Marcus has spent over a decade helping home-services businesses protect their online reputation and navigate Google review policy. He leads dispute strategy at ReviewShield and has personally managed review campaigns for hundreds of contractors across the US.

  • 10+ years in local reputation management
  • Google Business Profile specialist
  • Managed 500+ contractor review campaigns

Related articles