Bad CX metrics practices are like out-of-date GPS or a broken compass: you think you’re on the right track, only to find you’re going in circles or way off target.
Are you above or below norms? Typical CX benchmarks check your “level” in your industry.
CX in every industry1 is like an old house with uneven floors or walls or ceilings.
“In most US industries, CX quality declined across all three dimensions: effectiveness, ease, and emotion. … The disparity between the customer experience that brands intend to deliver and what customers actually experience is widening,” according to 275,000 customers’ perceptions of 469 brands across 12 industries and 13 countries in 2025.2
“Customer experience continues to erode worldwide, reflecting a concerning multiyear downward trend and a shift in sentiment from positive to neutral,” said Pete Jacques, principal analyst at Forrester.2
Benchmarking your industry is GIGO CX metrics. GIGO is an acronym for “garbage-in, garbage-out”: in any system, the quality of output is determined by the quality of the input.3
Let’s explore the catastrophic consequences of GIGO CX benchmarking — and what to do instead.
This is the third of a 7-part series:
A) Insults repel customers: the opposite of your goals.
1) Requesting a Score: disrespects customers’ truth.
2) Automating Interactions: makes CX hard in unique situations.
3) Upselling After CX Fail: is self-serving, not good CX.
B) GIGO VoC misleads managers.
4) Bias: shows your self-centicity.
5) Inconsistent Interpretation: leads to shaming customers.
6) Population Misrepresentation: makes CX metrics irrelevant.
C) GiGO Benchmarking is a leveling race. Make it a growing opportunity.
7) External Benchmarks are a deceptive slippery slope.
8) Internal Benchmarks must enhance customers’ prosperity.
7) External Benchmarks
It’s literally a slippery slope to compare NPS across industries:
- No industry has ever scored Good or Excellent in all the years of Forrester’s CX Index.1
- Only 14% of companies have mature CX practices.6
- Relational and Transactional NPS mean very different things.8
- Popular VoC practices make most VoC data biased4 and misrepresentative5 of what’s really happening in the market.
- Cultural differences are vast.5
- Timing and contextual relevancy are crucial to true vs. fake metrics.8
- Promoters are a mix of high-value and low-value customers: what % is the mix you’re comparing your brand to?8
CX practitioners ratings of their organization’s CX competency maturity, reported by Qualtrics’ 2025 State of Customer Experience Management:6
Embed: 3%
Stage 4: Scale 11%
Stage 2: Initiate 36%
Stage 3: Mobilize 21%
Stage 1: Investigate 29%
These percentages have barely changed since 2016.7 Note that it has been 17 years now since the 2008 global financial crisis when companies began investing heavily with CX tech/advisory firms.7 Meanwhile, CX/CS/Marketing practitioners are rampantly and hastily declaring themsleves CX Experts. (?!?)
Recovery after CX failure (need to contact Support) is typical transactional NPS (tNPS), not to be confused with relationship NPS (rNPS): customers’ assessment of their full experience with your brand. “tNPS: ‘Based on your experience with customer service, how likely will you recommend…’, while rNPS: “How likely will you recommend… [based on total relationship with the brand]“. Talk about internal focus! — Wai Au8
“Timing and contextual relevancy is so critical; being handed a card by a nurse after a quick op that asks ‘on a scale of 1-5 [erc.] would you recommend this ward?’ Er… you mean the NHS gives a still lightly anaesthetised patient a choice where to recover? Well. Had I known, I’d have asked to have been brought round in a Krispy Kreme shop, surrounded by choral singers wearing t-shirts saying ‘welcome, legend’, not this super clean designated recovery ward which is way better …. Oh wait. I’ve never had an op and I can’t compare … and I’m still high.” — Charles M8
“We keep optimizing for the score while ignoring the why behind it. I’ve seen teams celebrate a +10 jump without realizing it came from a cohort shift, not improved service. Until we normalize for cultural context and respondent behavior, we’re not running CX — we’re running a popularity contest.” — Jonathan Mingorance Fernandes8
“I have conducted an A/B testing and learned that a gift for answering an NPS can swing the NPS to the positive by up to 8 points.” — Wai Au9
“Publicly shared NPS is often more marketing than measurement — and the deception can be subtle but dangerous: Incentivized Feedback — ‘Tell us you’re a 9 or 10 and win a gift card!’
Spoiler: That’s not insight, it’s bribery. Cherry-Picked Segments — Only reporting NPS from their happiest, most profitable customers (while ignoring the rest). Survey Timing Games — Sending NPS surveys right after a great experience but never after a service issue. Score Without Context — No volume, no segment detail, no trend — just a shiny number without the story behind it. Don’t benchmark your internal truth against someone else’s polished fiction.” — Wai Au9
As I see it, the biggest issue is starting with Promoter-Passive-Detractor segmentation:
What % of Promoters are low-value customers?
What % are typically recommending low-value buyers?
What % are not inclined to actually recommend?
Same questions for Detractors for high-value.
Keys to Effective External Benchmarking:
1) Choose your benchmark partners carefully.
Ask how they guard against bias.
Ask how they select samples.
Ask how they pre-test questions and scales.
Ask how they analyze the data.
If you’re unsure of any of the above, get skilled.
2) Focus on practices, not numbers.
What’s their approach to actioning?
How do they close the loop with customers?
etc.
3) Compare customers’ prosperity as path to your prosperity.
What savings have they generated for customers?
What gains have they achieved for customers?
How did these translate to business growth?
8) Internal Benchmarks
It’s problematic to compare internal products or internal geographies.
They have different adoption curves.
Different cultures.
Do not compare satisfaction vs. importance.
All survey questions should be important (pre-test).
Stated importance is rarely accurate.
Customers don’t want your products/geographies competing with one another.
They want consistent performance.
More value for price, over time.
More compatibility, less obsolescence.
Ease of doing business, issue-free journeys.
“The best benchmark is your customers’ expectations. The best purpose for NPS and VoC of any type is to make your managers smarter than any of your competitors. You want managers of all kinds across your company hitting the nail on the head in their efficiency and effectiveness, as viewed by customers, who pay for salaries, budgets, and profit sharing.” — Lynn Hunsaker9
“Having deployed and rolled out the NPS system across over 30 countries at Allianz and also across other countries at both E.ON and Bupa we never compared countries and markets versus one another. I am less interested in the score and would rather focus on what is being done to act on the feedback.” — Andrew Clayton8
“But what does NPS do? It lumps all of them into three neat buckets — Promoters, Passives, Detractors — and treats a 6 the same as a 0. That’s not customer insight. That’s statistical malpractice. — Wai Au9
“Who really cares whether it’s a 9 or 7 in real life? What really matters is what you do about their comments. Instead of obsessing over ratings, look for patterns across customers. Tie survey ratings to operational data and customers’ digital body language, etc. Use AI to data-mine what’s relevant for each department across your company. Conduct workshops to help each department self-manage the top 2 gaps they contribute to. I do not hate the NPS question — except it’s actually about the company, not the customer, so it’s not customer-centric. It’s a calamity. Hurting jobs. Perpetual CX immaturity. Rising prices continually this decade without commensurate rise in quality/volume for price paid. That’s bad CX.” — Lynn Hunsaker8
Keys to Inspirational Internal Benchmarking:
1) Coordinate CX+EX+PX: Customer + Employee + Partner Experience.
Cross-pollinate each with best practices.
Find the intersection of stakeholders’ aims.
Embed the intersection in how you run the business.
2) Segment your customers the way Marketing does — initially.
How are you excelling for the top target market?
What can lessons learned bless every line of business?
How can functional areas use best practice findings?
3) Next, segment your customers by core-growth, then expectations.
Core-growth = intersection of spend expansion x low cost to serve.
These customers are best-fit for operational strengths.
Highlight performance vs. their expectations.
Make what you’re doing to close this gap the basis of recognition and compensation.
Conclusion: Increase Your Expertise
Immaturity of our customer experience profession is why bad CX metrics practices are prevalent.
Invest adequately to drive up maturity.
As I led our 1st year of formal CXM company-wide, we achieved Embed and Scale!
Baby steps and bold steps underway with obvious gains.
You get what you aim for.
We invested 40 hours/year in professional development for every employee.
Formal training is vital, beyond conferences and freemium sources.
Forrester just published realities of CXM expertise:
Very few track customers’ emotions.10
Root cause analysis is rare.10
Even fewer can identify behaviors that improve CX.10
Few can rally internal champions to act on feedback.10
Only 27% effectively communicate insights in a timely way.10
Communication about organizational actions in response to feedback is rare.10
$3.8 trillion revenue is lost annually due to bad CX.11 This is the tip of the iceberg! Spiraling costs curtail your resources available for revenue generation.
Cost spiral: Negative word-of-mouth slows sales velocity.
Lower sales velocity increases the cost of customer acquisition.
Self-service and digitalization make up for non-intuitive scenarios.
Customer Success shepherds customers’ product value and MVP improvements.
Customer Support remedies, returns, refunds, and escalations.
Loyalty incentives, discounts, rebates, product recalls, lawsuits.
Churn expands monthly quotas for Marketing leads and Sales conversion.
Technologies, facilities, staffing, and consultants to manage all the above.
To avoid these massive costs, stop “checking your level” in your industry!
It’s not putting you on the right track, and you’re going in circles or way off target with customers and investors alike.
Insead, start benchmarking practices and customers’ prosperity. Do it to bring your internal organizations up to par, getting in-sync with customers and with one another. This builds trust through consistent performance that stops bad costs and grows ongoing mutual value for every party: customers, employees, partners, and investors.
1Rethinking Customer Experience Strategy: Forrester’s 2022 CX Index Dropped to 2020 Performance, by Lynn Hunsaker, CustomerThink, July 12, 2022.
2Forrester’s 2025 Global Customer Experience Index Rankings, Forrester, June 24, 2025.
3What is garbage in, garbage out (GIGO)?, TechTarget, June 14, 2023.
425 Bad CX Practices to Stop in 2025: Insults (Part 1), by Lynn Hunsaker, CustomerThink, January 30, 2025.
5GIGO Metrics: 26 Bad CX Practices to Stop in 2026 Part 2, by Lynn Hunsaker, CustomerThink, September 7, 2025.
6The State of Customer Experience Management, 2025, by Topher Mitchell and Talia Quaadgras, Qualtrics, 2025.
7Why Only 15% of Voice of Customer Programs are “Very Successful”, by Lynn Hunsaker, ClearAction, 2016.
8Hot Take: NPS is Broken by Design, by Wai Au, LinkedIn, 2025.
9That NPS You’re Chasing? It’s Probably a Lie, by Wai Au, LinkedIn, 2025.
10Six Gaps Hold Feedback Management And CX Measurement Programs Back, by Colleen Fazio and Maxie Schmidt, Forrester, August 8, 2025.
11Bad Customer Experiences Put Nearly $4 Trillion at Risk in Global Sales, Qualtrics, November 19, 2024.
This is the third of a 7-part series: 26 Bad CX Metrics Practices to Stop in 2026.
Part 1: Insults repel customers: the opposite of your goals.
1) Requesting a Score: disrespects customers’ truth.
2) Automating Interactions: makes CX hard in unique situations.
3) Upselling After CX Fail: is self-serving, not good CX.
Part 2: GIGO Metrics: misleads managers.
4) Bias: shows your self-centicity.
5) Inconsistent Interpretations: lead to shaming customers.
6) Population Misrepresentation: makes CX data irrelevant.
Part 3) Benchmarking: is a growing opportunity, not a leveling race.
7) External Benchmarks are a deceptive slippery slope.
8) Internal Benchmarks must enhance customers’ prosperity.
Coming Soon
Part 4) Prioritization
Part 5) Leading Indicators
Part 6) Goal Silos
Part 7) Metric Silos
Image licensed to ClearAction Continuum by Shutterstock.
Lynn, this is spot on: benchmarking NPS across industries is theater unless it’s tied to value created and cost to serve.
We once posted +12 NPS QoQ while churn rose—cohort mix shifted and tickets spiked; only when we tracked time-to-first-value, repeat-issue rate, and recovery time did expansion bounce back.
Benchmarks are useful only as reference classes for practices (e.g., onboarding lead time, defect escape rate, closed-loop velocity), not as scoreboard numbers.
If you could keep just three cross-company metrics that actually predict retention, which would you pick?
Amazing, Jonathan. I agree that benchmarking is best for learning what to do better.
When benchmarking is used to justify doing nothing, or to game the system, then it’s a travesty of wasted opportunities and under-harvested investment.
What an expensive way to feel good about the brand, or to reward subpar performance. It’s ludicrous.
Especially by involving customers in the process: wasting their time when nothing is done to give them an ROI for their generous participation.
Best predictors of retention are:
1) root cause resolution of prevalent pebbles in customers’ shoes.
2) value increasing for customers.
3) pain decreasing for customers.
How I look at these metrics is shown here:
https://customerthink.com/executives-guide-to-customer-experience-value-metrics
Love this, Lynn — especially the callout on wasted opportunities when benchmarking is used as cover for inaction.
The irony is companies run customer surveys that end up serving internal vanity, not customer outcomes. Your three predictors nail it: reduce pain, increase value, fix root causes.
The real question: how do we convince exec teams to fund these fundamentals instead of chasing glossy “best-in-class” NPS slides?
Great question, Jonathan. 2 ways to convince execs to get off the vanity wagon — and instead embrace reducing pain, increasing value, and fixing root causes:
1) Show them the *revenue* power of VoC use:
(a) 8X higher revenue . . .
. . . when innovations are guided on customer intelligence.
(b) 8X return to shareholders when voice of customers + employees + partners is synchronized for aligning how you run your business to your high-potential stakeholders.
(see diagrams at https://customerthink.com/executives-guide-to-customer-experience-value-high-potential)
2) Show them the *budget expansion* power (including margins, profit-sharing) of VoC use:
(see diagrams at https://customerthink.com/executives-guide-to-customer-experience-value-revenue-risks-part-1)
(a) 3X higher revenue when cost containment is guided by customer intelligence.
(b) 9700% ROI on stopping root causes of a prevalent issue.
(see diagrams at https://clearaction.com/quantify-cx-roi)
All of these estimates are vastly understated, I believe, because they do not for the spiraling effect of reducing quotas for Marketing/Sales to meet month-over-month increases in investor expectations.
(see diagrams at https://clearaction.com/acquisition-addictions-impact-on-customer-experience-roi