Three hundred reviews in a month

Situation. A company whose customer acquisition depended on public ratings and whose hiring depended on employer-review platforms.

The attack. More than three hundred reviews across three platforms in a single month. Genuine dissatisfaction does not arrive on a schedule, and it does not arrive across unrelated platforms at once.

The rating fell by forty per cent. Sales conversations began opening with questions about it. Candidates went quiet.

What most advisors would have done. Report the reviews and wait. The platforms did eventually remove them in bulk — after three months. Had we done nothing else, the damage in that window would have been far greater than the damage that actually occurred.

What we did — three tracks at once, because the slow one alone loses.

• Built the platforms a case they could act on: arrival patterns, account characteristics, linguistic fingerprints, the absence of any matching transaction or employment record.
• Answered every single review, individually, within hours. This is the part that mattered most and cost the least. A reader deciding whether to believe a wall of one-stars looks at the responses.
• Let the real customers and employees speak — a systematic, compliant programme of asking everyone, with no incentives tied to sentiment and no filtering for people expected to be positive.

Outcome. The rating recovered in one month — two months before the platforms acted. The coordinated reviews were ultimately removed. The genuine review base ended larger than before the attack.

The source. A departing employee, acting with the competitor he had joined.

The lesson. The platform's timeline is not your timeline. The evidence case is necessary and slow. What protects the business in the meantime is the reply layer and a real review base. Run all three from day one, or you spend three months winning an argument nobody was watching.