Net Promoter Score (NPS)

NPS is a single-question customer loyalty metric that measures how many more customers are willing to strongly recommend a brand than to criticise it. Introduced by Frederick Reichheld in Harvard Business Review in 2003 and co-developed with Bain & Company and Satmetrix, it became the most widely adopted CX measurement tool in commerce — used by approximately two-thirds of Fortune 1000 companies (as-of 2026). Despite widespread adoption, NPS is also the subject of persistent academic and practitioner criticism.

---

How it works

Customers are asked a single question on a 0–10 scale:

"How likely are you to recommend [company/product/service] to a friend or colleague?"

Respondents are segmented into three groups:

Segment	Score	Behaviour
Promoters	9–10	Loyal enthusiasts; referral source; "more than 80% of referrals in most businesses" (Bain, via netpromotersystem.com)
Passives	7–8	Satisfied but vulnerable to switching; excluded from the score calculation
Detractors	0–6	Unhappy customers; source of negative word-of-mouth; 50% reported to churn within 90 days of survey (as-of 2026)

Formula:

NPS = % Promoters − % Detractors

Score ranges from −100 (all detractors) to +100 (all promoters). (Qualtrics; NPS System / Bain)

---

Benchmarks

Bain & Company thresholds (as-of 2026)

• 0–30: good — more promoters than detractors
• 50+: excellent
• 70+: world-class (Shopify Blog, 2026-03-24; Ringly.io citing Bain, 2026-05-21)

Cross-industry averages (as-of 2026)

• All-industry pooled average NPS: 32 (mean); median: 44 — median is higher because a handful of very-low scores drag the mean down (Retently 2026 via Ringly.io)
• B2C average: 49 — B2B average: 38 (11-point structural gap reflecting longer B2B decision cycles and multi-stakeholder buying) (Retently 2026 via Ringly.io)
• Top 25% of companies score 72+; bottom 25% score 0 or lower (as-of 2026) (Retently via Ringly.io)

Ecommerce & retail benchmarks (as-of 2026)

Named company examples (as-of 2025–2026, vendor-estimated — low confidence, not independently audited):

• Amazon: 73, Chewy: 68, Flipkart: 70, John Lewis: 68, Coolblue (NL): 66, Zappos: ~58, Costco: 70–80 (Retently/CustomerGauge benchmark database via Opensend & Ringly.io)

---

Relational vs Transactional NPS

Two complementary survey cadences serve different purposes:

Relational NPS

• Sent on a fixed schedule (quarterly, semi-annual, or annual) across the customer base
• Tracks overall brand loyalty over time
• Sample strategy: rolling sample rather than surveying all customers at once (avoids survey fatigue)
• Advice: do not run too frequently; mature programmes often run quarterly rolling samples (Retently; NPS Masterclass YouTube, 2025-08-24)

Transactional NPS (tNPS)

• Fires automatically after a specific touchpoint (post-purchase, post-delivery, post-support)
• Captures touchpoint-specific satisfaction with higher recency relevance
• Response rates are 8–12 points higher than relational surveys (as-of 2026) (Shopify Blog, 2026-03-24)
• Primary ecommerce tNPS triggers: post-purchase (order placed), post-delivery confirmation, post-support resolution (Zonkafeedback)

Ecommerce timing best practices (no authoritative primary source — vendor guidance):

• Trigger on delivery confirmation (not order placement) — 24–48 hours after delivery to allow product use (Zonkafeedback)
• Frequency throttle: send no more than once per customer per 30 days even if multiple transactions occur in that window (Zonkafeedback)
• Combined approach: Retently recommends quarterly relational NPS supplemented by transactional NPS after key moments for both trend data and actionable insight (Retently)

---

Survey response rates

• In-app NPS: 21.71% average (as-of 2026) (Ringly.io)
• SMS NPS: 30–50% (as-of 2026) (Ringly.io)
• WhatsApp NPS: 30–50% (as-of 2026) (Ringly.io)
• Transactional NPS: 25–40% (as-of 2026) (Ringly.io)
• Bain minimum thresholds: B2C 40%+; B2B key accounts 60%+; below these, sample bias contaminates the score (as-of 2026) (CustomerGauge / Bain via Ringly.io)

In Retently's 2025 data, CSAT surveys averaged 9.76% and CES surveys averaged 22.54% in ecommerce — both substantially higher than NPS email surveys (Retently, 2025–2026)

---

Closing the loop

NPS is only operationally useful if feedback triggers action. Practitioners distinguish two levels (Nienke Bloem, YouTube series, Oct 2025):

Small loop (individual recovery)

Individual follow-up after negative feedback:

• Who acts: webcare team, customer contact centre
• What they do: apologise, fix the problem, "go the extra mile"
• When: target 90% of detractor tickets closed and communicated back to customer within 24–48 hours (Nienke Bloem, 2025-10-27)
• Routing: Detractors → private recovery workflow (to prevent public escalation); Promoters → request for public review (Google/Trustpilot) (Retently; Zonkafeedback)
• A detractor who receives follow-up is statistically more likely to remain a customer and become a promoter — making loop closure a retention mechanism, not just goodwill (Bloem, 2025-10-20)

Big loop (structural change)

Aggregate NPS insight drives operational change — "where scale impact lives":

• Analyse open-text feedback (comments are where actionability lives, not the numeric score)
• Run "Friction Fridays" — regular team sessions addressing a specific customer friction surfaced by NPS open text
• Co-creation workshops to generate structural fixes (Bloem, 2025-10-27)
• Companies with successful NPS programmes are 3x more likely to share results internally (as-of 2026) (Shopify, 2026-03-24)

Van der Aa's structural point: at a 20% survey response rate, only 5% of the total customer base are unhappy respondents giving direct feedback. The big loop — fixing the underlying process — matters far more at scale than the small loop of individual follow-up (van der Aa, YouTube, 2022-12-02)

---

NPS and business growth

Associated metrics where there is broader agreement (as-of 2026):

• Promoters account for more than 80% of referrals; 92% of consumers trust personal recommendations; promoters make buyers 4x more likely to purchase (Bain / Ringly.io)
• Promoters spend 20–30% more over their lifetime than average customers; companies with NPS 75+ show at least 13% higher CLV (as-of 2026) (Ringly.io citing xebo.ai analysis)
• 50% of detractors churn within 90 days of survey (as-of 2026) (Ringly.io citing xebo.ai)

---

Limitations and criticisms

Gaming and sampling bias

• Call centre staff may only mention post-call surveys after positive experiences, creating selection bias
• Customers can be bribed with coupons into giving high scores ("rating racketeering") (NNg, 2024-06-21)
• Reichheld himself has publicly criticised the most damaging misuse: linking scores to employee compensation, which leads to "pleading, bribery and manipulation" (CustomerGauge, via web source)

Information loss from the 3-bucket model

• Passives (7–8) are discarded; a customer moving from score 2 to score 5 — a significant UX improvement — is still counted as a detractor
• Requires larger sample sizes than other metrics because passives are excluded from the formula (NNg, 2024-06-21)
• NPS gives the "what" but not the "why" — the open-text comment is essential for actionability

International variation

The Gartner prediction and its aftermath

Non-customer exclusion

Bähre (NIM, 2024) argues that standard NPS programmes measure only existing customers, missing former customers and brand-aware non-buyers — a significant source of potential sales growth. Measuring NPS beyond the current customer base would give a fuller picture of brand health. (NIM YouTube, 2025-02-03)

---

NPS vs CSAT vs CES

Three complementary satisfaction metrics serve distinct purposes:

Metric	Measures	Best for
NPS	Long-term loyalty to the brand	Quarterly brand health tracking; input to CLV and referral models
CSAT (Customer Satisfaction Score)	Immediate satisfaction after a specific touchpoint	Steering CX at journey-level; high-frequency transactional feedback
Customer Effort Score (CES)	Ease / effort required at a specific interaction	Identifying and reducing friction points that drive churn

(Retently, 2026)

Van der Aa's operational recommendation: use CSAT for steering at the touchpoint level (highest actionability for individual experience decisions), and NPS for brand-level annual or quarterly tracking. (van der Aa, YouTube, 2022-12-02)

Metric benchmarks (Retently platform data, 2025):

• CSAT email response rate in ecommerce: 9.76% vs NPS email 3.24%–4.5%
• CES email response rate in ecommerce: 22.54% — highest of the three (as-of 2025)

---

NPS in post-purchase ecommerce context

Ringly.io's 2026 editorial analysis of top-scoring ecommerce NPS brands concludes: "NPS in 2026 has become a lagging indicator of one thing: how fast and how well you handle post-purchase friction. The top scoring e-commerce brands (Amazon, Costco, Chewy) are winning on resolution — same-day delivery, easy returns, and a human voice when something goes wrong." (Ringly.io, 2026-05-21)

Top NPS drivers in ecommerce (as-of 2026, Retently vendor data):

• Shipping speed and reliability
• Product quality matching expectations
• Ease of Returns Management
• Customer service responsiveness
• Website/app user experience (Retently via web source)

The post-purchase stage is described as "the most emotionally charged part of the customer journey" — proactive, personalised delivery updates (see Branded Tracking Pages, WISMO) build trust, manage expectations, and are a direct driver of improved NPS scores and repeat purchases. (ParcelLab Blog, 2026-04-09)

Connection to WISMO: only 57.9% of retailers were using personalised recommendations in post-purchase communications as-of 2025 — a significant gap given NPS's link to Post-Purchase Experience quality (as-of 2025, ParcelLab Shipping Experience Study).

---

Earned Growth Rate (complement to NPS)

Reichheld introduced Earned Growth Rate (EGR) as a hard financial complement to NPS, comprising two components (HBR, 2021-11; Bain & Company, 2021):

• Net Revenue Retention (NRR): % of revenue retained from prior-year customers (captures loyalty-driven retention)
• Earned New Customers (ENC): revenue growth from referrals as a share of total new revenue (captures word-of-mouth growth vs. paid acquisition)

EGR is designed for auditor verification (hard financial data), unlike NPS which is survey-based and self-reported. No public data on how many ecommerce or retail companies have adopted EGR as a reported metric is available in current sources (gap).

See Earned Growth Rate for dedicated coverage (frontier topic).

---

Key terms

Term	Meaning
Promoters	Customers scoring 9–10 on the recommend question; loyal advocates
Passives	Customers scoring 7–8; satisfied but excluded from NPS calculation
Detractors	Customers scoring 0–6; source of churn and negative word-of-mouth
Relational NPS	Fixed-schedule brand-level survey (quarterly/annual)
Transactional NPS (tNPS)	Event-triggered survey fired after a specific touchpoint
Closing the loop	Acting on NPS feedback — individually (small loop) or structurally (big loop)
Earned Growth Rate (EGR)	Hard financial complement to NPS = NRR + ENC
CSAT (Customer Satisfaction Score)	Touchpoint-level satisfaction metric
Customer Effort Score (CES)	Ease-of-interaction metric; CEB found predictive of loyalty; predictive power context-dependent

---

What practitioners report

• Bol.com (NL, 40M+ customers): Head of CX Jeroen Velthof describes NPS not as a score to chase but as "a dynamic, actionable tool that drives change" — focus on what the score reveals about broader customer trends, not fluctuations in the number. They make NPS scores and open feedback visible across the organisation and run sessions where employees listen to live customer calls. (van der Aa / CX Quantum Leap podcast, Episode 37, 2024-11-12 — Spotify, no YouTube URL confirmed)

---

Gaps in current vault coverage

• No fashion-specific NPS data for European retailers (UNIQLO, H&M, Zara) — not publicly available from independent sources
• No data on NPS differences between omnichannel (in-store + online) vs. pure-play ecommerce journeys for the same brand
• No 2024–2026 peer-reviewed academic study on NPS vs. alternative metrics in retail/ecommerce (ScienceDirect 2022 paper is the most recent found; paywalled)
• No data on Earned Growth Rate adoption in practice among ecommerce/retail companies
• Closing-the-loop SLA benchmarks lack a neutral quantitative source — Retently/Zonkafeedback are vendors
• Reddit gap: MCP unavailable in Cowork cloud (known recurring issue) — practitioner sentiment on NPS not retrieved

---

Dangling links (frontier topics)

• CSAT (Customer Satisfaction Score) — referenced 6+ times across vault; not yet written
• Customer Effort Score (CES) — referenced 5+ times across vault; not yet written
• Earned Growth Rate — new dangling from this page; Reichheld's NPS complement
• Voice of Customer (VoC) — referenced tangentially; not yet written