What is Survey microtexts (ready phrases)?

Q: What is Survey microtexts (ready phrases)?

Main question: "How likely are you to recommend [company/product name] to a friend or colleague? (0 — not at all likely, 10 — extremely likely)". Follow-up (open): "What was the main reason for your rating?" / "What should we change for you to rate 2 points higher?"

NPS (Net Promoter Score): Complete Business Guide — Formula, Interpretation, Error, Actions

What is Net Promoter Score (loyalty index)

Net Promoter Score (NPS) — "loyalty index" that measures customer willingness to recommend a product/service to friends or colleagues (0–10 scale). The concept was published by Frederick Reichheld in Harvard Business Review (2003); methodology was further developed by Bain & Company and Satmetrix (Net Promoter System).

After answering, customers are divided into 3 groups: detractors (0–6), passives (7–8), promoters (9–10). This is the classic system described in Ukrainian guides.

How to calculate NPS (with examples)

Formula: NPS = % promoters − % detractors (in percentage points, range from –100 to +100).

Example 1. 200 responses: 110 promoters (55%), 60 passives (30%), 30 detractors (15%).

NPS = 55 − 15 = +40.

95% confidence interval (CI):

SE = sqrt((0.55 + 0.15 − (0.55 − 0.15)^2) / 200) ≈ 0.052

CI ≈ 40 ± (1.96 × 0.052 × 100) ≈ 40 ± 10.2 p.p. (i.e., ~29.8…50.2).

Types of NPS and when to ask

Relational NPS (relationship) — periodic surveys (e.g., quarterly/semi-annually) to assess brand "health" and benchmarking.

Transactional NPS (transactional) — immediately after key interaction: purchase, delivery, support contact, onboarding completion. Provides tactical insights at touchpoint level.

Timing. For digital products, often works cycle: 7–15 days after installation/first use; then repeat after 30–90 days; for continuous services — every ~90 days.

Best NPS practices (short checklist)

Simplicity and consistency of methodology. Same wording, one scale, transparent audience selection logic.

Correct touchpoints (transactional) + periodic waves (relationship).

Follow-up. Ask for rating reason and close the loop (contact/decision fixation in CRM).

Measure dynamics, not just absolute. Look at trend by channels/segments/functions.

Data ethics. Don't incentivize "9–10" with bonuses — this distorts the metric.

NPS statistics without pain: error, CI, sample size

Standard error (SE):

SE = sqrt((p_prom + p_det − (p_prom − p_det)^2) / n)

95% confidence interval (in p.p.):

NPS ± 1.96 × SE × 100

Quick sample size guidelines.

Worst case (MOE): ≈ 196 / √n (in p.p.). For ±5 p.p. need ≈ 1,537 responses (round up).

Realistic mix: for p_prom=45%, p_det=20%, n=500 → NPS=+25 ± 6.7 p.p. (95% CI).

How to convert NPS into business results

Cohort & journey analysis. Calculate NPS by segments (new vs repeat, regions, products, channels).

Driver analysis. Code open comments → topics → impact on NPS/churn → backlog.

Connection with financial results. Compare churn and LTV between groups; set OKRs for "detractor containment" (detractor processing within X days) and promoter activation (how many referrals/reviews activated).

Limitations and common mistakes

NPS ≠ "holy grail". Research (Keiningham et al., Journal of Marketing, 2007) showed that NPS is not always a better predictor of growth than alternatives (CSAT etc.). Keep a set of metrics (NPS + CSAT + CES + behavioral KPIs).

"Gaming" the metric. KPIs sometimes provoke "squeezing out" 9–10 — this kills diagnostic value (see modern discussion around NPS 3.0).

Survey microtexts (ready phrases)

Main question: "How likely are you to recommend [company/product name] to a friend or colleague? (0 — not at all likely, 10 — extremely likely)".

Follow-up (open): "What was the main reason for your rating?" / "What should we change for you to rate 2 points higher?"

Frequently asked questions (FAQ)

What is considered a "good" NPS?

Depends on niche and country. Look at external benchmarks (NICE Satmetrix), but make decisions based on your own trend and CI.

How does NPS differ from CSAT/CES?

NPS — about willingness to recommend, CSAT — about satisfaction with specific interaction, CES — about effort to solve a task. Combine them.

How much and how often to survey?

For product services — always-on transactional NPS at key points + periodic relational (quarterly/semi-annually). For mobile applications — cycle 7–15 days → 30–90 days.

Blog