2.8 Rating
Two years ago, a hot spring resort in Beitou. Google rating 2.8. 15-20 new negative reviews every month. Response rate: 0%.
Every day, the owner opened his phone to a row of red one-stars. “Terrible service.” “Outdated facilities.” “Not worth the price.” “Rude front desk staff.”
He tried lowering prices. He tried replacing front desk staff. He tried hiring a PR firm. The rating didn’t budge.
Lowering prices cost NT$150,000 in lost revenue per month. Replacing three front desk staff members added up to NT$80,000 in training costs. The PR firm charged NT$120,000, wrote a few blog posts, sent out a few press releases — and brought zero new positive reviews. The rating didn’t budge. Not because he wasn’t trying hard enough, but because every action he took missed the root cause.
He was fighting with intuition, not with data.
The First Thing He Did: Classify
He stopped replying one by one. Instead, he listed all negative reviews from the past three months and categorized them one by one.
Result: out of 60 negative reviews, 38 were related to “service attitude.” 20 were about “outdated facilities.” 2 were about parking.
“Service attitude” accounted for 63%. This is Pareto’s “vital few.”
Italian economist Vilfredo Pareto observed in 1897 that 80% of Italy’s land was owned by 20% of the population. Later, Joseph Juran popularized this observation in 1937 as the principle of the “vital few and useful many”: 80% of problems typically come from 20% of causes. In this resort’s case, 63% of negative reviews pointed to the same problem — service attitude. Solve this one problem, and you eliminate more than 60% of the source of negative reviews.
But “service attitude” was too vague. He used the Five Whys to dig deeper:
- Why do guests say the service attitude is poor? → Because front desk staff are “cold” and “impatient.”
- Why are front desk staff cold? → Because they’re simultaneously handling check-in, phone calls, and guest complaints.
- Why are they handling so many things at once? → Because there isn’t enough staff.
- Why isn’t there enough staff? → Because the owner cut weekend part-timers to save costs.
- Why cut part-timers? → Because revenue was declining.
Root cause: cutting staff → overload → poor attitude → negative reviews → fewer guests → lower revenue → more staff cuts. A vicious cycle.
The Five Whys is a root cause analysis tool developed by Sakichi Toyoda, founder of the Toyota Production System, in the 1930s, and later promoted by Taiichi Ohno within Toyota Motors. Its core principle is simple: ask “why” five times consecutively until you reach the true root cause. Most people stop at the first or second level and start solving. But the answer at the first level is usually a symptom, not the disease. If this resort had stopped at the first level (“staff have bad attitudes” → “replace staff”), the problem wouldn’t have been solved. Digging to the fifth level revealed: the root cause wasn’t the staff. It was the owner’s decision to cut headcount.
The Second Thing He Did: Target the Root Cause
He didn’t lower prices. He did something counterintuitive: he brought back the weekend part-timers — and added one more.
Costs went up. But the front desk was no longer overloaded. Wait times dropped from an average of 12 minutes to 4 minutes. “Bad attitude” complaints dropped 70% within two months.
Next, he addressed “outdated facilities.” Of the 20 reviews, 12 mentioned “peeling tiles in the hot spring pool.” He spent NT$20,000 to repair the tiles — not a renovation, just fixing tiles. “Outdated facilities” complaints nearly dropped to zero.
Repairing tiles cost NT$20,000. Adding part-timers cost NT$30,000 per month extra. Lowering prices cost NT$150,000 per month in lost revenue. He chose to spend NT$50,000 to solve the problem, rather than lose NT$150,000 pretending to solve it.
Here’s an important business insight: the cost of solving a problem is usually far lower than the cost of pretending to solve it. Tile repair: NT$20,000, vs. continuing to bear the booking losses caused by “outdated facilities” reviews — assuming a loss of 5 groups of guests per month at an average spend of NT$6,000, that’s NT$30,000 lost per month. One month. The tile repair lasts three years. The return on investment is tens of times.
But most operators, when they see “spend money to repair tiles,” their first reaction is “more expenses.” What they don’t see: not repairing costs even more — it just costs in invisible places.
The Third Thing He Did: Respond
After classifying, he started responding — but not by apologizing one by one. He designed three response templates, used according to the type of negative review:
Service attitude type: “Thank you for your feedback. We have adjusted our weekend staffing, and front desk wait times have been reduced from 12 minutes to 4 minutes. Please let us know on your next visit, and we will personally ensure your experience.”
Facilities type: “The hot spring pool tile issue you mentioned was repaired last month. We will inspect all facilities regularly. Thank you for letting us know.”
Other types: Individually customized responses.
The key isn’t the template. It’s that each response specifically states the action taken. Not “we will improve” — but “we have done X, and the result is Y.”
The difference between these two is enormous. “We will improve” is an empty promise — when potential guests read it, their reaction is “every place says that.” “We have adjusted our weekend staffing, and wait times have dropped from 12 minutes to 4 minutes” is a verifiable fact — when potential guests read it, their reaction is “this place actually did something.”
BrightLocal’s 2023 consumer trust survey found: 62% of consumers said that “responses with specific management actions” increase their trust in a brand. And responses that “only apologize without action” actually decrease trust — because consumers feel “they’re just doing PR.”
When potential guests read these responses, they don’t see apologies — they see a brand actively solving problems.
NPS Back-Testing
Fred Reichheld published “The One Number You Need to Grow” in Harvard Business Review in 2003, formally introducing NPS (Net Promoter Score). He replaced dozens of pages of satisfaction surveys with a single question: “How likely are you to recommend this company to a friend?” On a 0-10 scale, 9-10 are promoters, 0-6 are detractors. Promoter percentage minus detractor percentage equals the NPS.
Reichheld subsequently partnered with Bain & Company to track 200 companies over three years. For every 10-point increase in NPS, revenue growth rate averaged 2.5 percentage points above industry peers.
But the point isn’t the score itself — it’s the structured analysis of “promoters’ reasons for recommending” and “detractors’ reasons for criticizing.” That’s exactly what this resort did: instead of asking “what’s the score,” they asked “why promote, why not promote,” and then acted on the root causes.
Reichheld later added a key insight in his 2020 book Winning on Purpose: high NPS isn’t achieved by “making everyone happy.” It’s achieved by “first solving detractors’ pain points, turning them into at least neutral.” Because the negative word-of-mouth generated by one detractor is 2-3 times more powerful than the positive word-of-mouth from one promoter. Eliminating detractors is more efficient than creating promoters.
18 Months Later
Rating rose from 2.8 to 4.1. Bookings grew 40%. Response rate went from 0% to 95%.
Not a miracle. A method: classify → Five Whys to find root causes → target the root cause → systematize responses.
But this method takes several days of manual work every month. Classification relies on manual reading. Frequency relies on manual counting. Responses rely on manual writing. If the scale were larger — ten branches, each with 100 negative reviews per month — labor costs would grow exponentially.
Download the 12 Types of Negative Reviews Checklist
Want to classify your negative reviews using the same method? Leave your email to get the 12-type checklist and see which types your negative reviews cluster around — that’s where your root causes lie.
Further Reading
← Classify Before You Reply: Turning Every Negative Review Into Actionable Intelligence
The method this resort used is the 12-type classification + Five Whys. Read this article first to understand the framework.
→ Read
Loading…