Why NPS Is the Metric That Matters Most
Net Promoter Score was invented by Bain in 2003 and is now used by over two-thirds of Fortune 500 companies. For early-stage startups, it's the fastest proxy for product-market fit you can get from your own customers. A rising NPS predicts retention. A falling NPS predicts churn — often before it shows up in revenue. If you're not measuring it, you're flying blind.
Understand the NPS Formula
NPS is deceptively simple. You ask one question: “On a scale of 0–10, how likely are you to recommend us to a friend or colleague?” Then you segment responses into three buckets and run the math.
Promoters
Loyal enthusiasts who will refer others and keep buying. These are your best growth lever.
Passives
Satisfied but unenthusiastic. They won't churn immediately, but they'll leave if a competitor makes a better offer. Not counted in the NPS formula.
Detractors
Unhappy customers who can damage your brand through negative word-of-mouth. Churn risk is high. Address these first.
The Formula
Score ranges from −100 (everyone is a Detractor) to +100 (everyone is a Promoter)
Example:
You survey 100 customers. 55 score 9–10 (Promoters), 30 score 7–8 (Passives), 15 score 0–6 (Detractors).
NPS = 55% − 15% = +40
Key point
Passives are intentionally excluded from the NPS calculation. They improve your absolute percentage numbers but don't reflect true advocacy. This is by design — NPS only cares about your extremes.
Design Your NPS Survey
How you ask is as important as what you ask. The wrong survey design produces garbage data — responses that don't reflect true customer sentiment, collected at the wrong moment, from the wrong segment.
The two-question structure
The NPS question (required)
“On a scale of 0 to 10, how likely are you to recommend [Company] to a friend or colleague?”
Use exactly this wording. Don't paraphrase it — the benchmark data assumes this phrasing.
The follow-up (required)
“What's the main reason for your score?”
Open-ended, optional for respondents. This is where the real insight lives — don't skip it.
When to send it
Good timing
- +30 days after signup (onboarding NPS)
- +After a key milestone (first value moment)
- +60–90 days after renewal
- +Quarterly for your active base
Bad timing
- −Same day as signup (no experience yet)
- −Right after a support ticket
- −During a known outage or incident
- −More than twice per year to the same user
Choose Your Survey Tool
You need a tool that handles NPS templates, automated distribution, response segmentation, and trend tracking over time. Don't use a Google Form — you can't segment, automate, or track trends without a lot of manual work.
SurveyMonkey — Best for NPS
- +Built-in NPS question type with proper 0–10 scale
- +Automated NPS score calculation and dashboards
- +Email distribution with follow-up reminders
- +Segment responses by custom fields
- +Trend tracking across multiple survey waves
- +Industry NPS benchmarks built in
Amplitude — For Product NPS
- +In-app NPS surveys triggered by behavior
- +Correlate NPS scores with product usage events
- +Segment NPS by feature adoption cohorts
- +Identify which features drive Promoters vs Detractors
- +Retention analysis by NPS segment
Early-stage shortcut
If you have under 100 customers, SurveyMonkey via email is all you need. Once you have 500+ active users, add Amplitude to get in-product NPS correlated with feature usage — that's where the real insight lives.
Collect Responses at Scale
Statistical significance matters. An NPS of +40 from 10 responses is meaningless. The same score from 200 responses starts to tell you something. Here's what you need for reliable data.
Minimum response requirements
Segmentation that actually matters
Your overall NPS is a blended average that hides the real signal. Always break it down by at least three dimensions:
- Plan tier: Enterprise customers almost always score higher than SMB. If your enterprise NPS is +60 and SMB is +10, you have very different problems to solve.
- Customer tenure: 0–90 day customers vs 12+ month customers reveal onboarding problems vs long-term value delivery gaps.
- Feature adoption: Customers who use your core feature daily should score dramatically higher. If they don't, your value prop is broken.
Analyze and Benchmark Your Score
A score of +35 could be excellent or terrible depending on your industry. Context is everything. Here are the benchmarks I use when advising portfolio companies.
| Industry | Poor | Average | Good | Excellent |
|---|---|---|---|---|
| B2B SaaS | < 10 | 10–30 | 30–50 | 50+ |
| E-commerce | < 20 | 20–40 | 40–60 | 60+ |
| Financial Services | < 0 | 0–25 | 25–45 | 45+ |
| Healthcare Tech | < 15 | 15–35 | 35–55 | 55+ |
| Consumer Apps | < 25 | 25–45 | 45–65 | 65+ |
| Marketplace | < 10 | 10–30 | 30–50 | 50+ |
What drives Detractor scores
Read every Detractor response. Categorize the open-ended answers into themes. The most common root causes I see across startups:
- Onboarding friction (40% of Detractors): They never got to value. Fix onboarding before anything else.
- Missing features (30%): They use the product but it doesn't solve their full workflow. Roadmap signal.
- Bugs or reliability (20%): They've been burned. Follow up fast — this group is salvageable if you show urgency.
- Pricing perception (10%): They don't see the ROI relative to cost. Packaging or communication problem, not product.
Close the Loop and Move the Number
Running NPS and doing nothing with the results is worse than not running it at all. It signals to customers that you asked for their feedback and ignored it. Closing the loop is what separates companies that use NPS as a vanity metric from those that use it as a growth tool.
Detractors (0–6)
- →Follow up within 48 hours
- →Founder or CSM call (not a form email)
- →Diagnose the root cause
- →Share a specific fix timeline
- →Re-survey in 60 days
Passives (7–8)
- →Invite to product advisory group
- →Share your roadmap and get input
- →Early access to new features
- →Ask what would make them a 9
- →Track usage — are they getting stickier?
Promoters (9–10)
- →Ask for a G2/Capterra review
- →Invite for a case study or testimonial
- →Launch a referral program
- →Ask for warm intros to their network
- →Invite to your customer community
The cadence that works
Run NPS quarterly to your entire active base. Run milestone NPS (30-day, 90-day post-signup) continuously. Review Detractor responses weekly in your product meeting. Track your NPS trend monthly. It should go up every quarter — if it's flat, something is wrong.
The Single Most Important Takeaway
NPS only matters if you close the loop. Sending a survey and not following up with Detractors tells your unhappiest customers that you asked for their opinion and don't care about it — which guarantees they churn. The founders who get the most out of NPS are the ones who treat every Detractor response as a personal customer support ticket that requires a human response within 48 hours.
5 NPS Mistakes That Kill the Signal
Surveying too frequently
Hitting customers with NPS every month creates survey fatigue and artificially suppresses scores. Quarterly is the maximum for your established base. More frequent than that and you're measuring survey annoyance, not product sentiment.
Reporting the overall number without segments
An NPS of +32 company-wide might hide an enterprise NPS of +65 and an SMB NPS of +5. The blended score tells you nothing actionable. Always break it down by cohort before presenting to your board or team.
Skipping the open-ended follow-up
The number tells you the score. The text tells you the reason. Without the open-ended question, you know you have a problem but not what to fix. Never run NPS without the “what's the main reason for your score?” follow-up.
Surveying only satisfied customers
If you send NPS only to customers who logged in this week, you're sampling your most engaged users. Your NPS looks great but misses the customers who are quietly churning. Include customers who haven't logged in for 30+ days — that's where the real signal is.
Not tracking it as a trend
A single NPS reading is a snapshot. What matters is whether it's trending up or down quarter over quarter. A +45 that fell from +60 is a red flag. A +25 that climbed from +5 is a strong signal you're making the right changes. Track the trend, not just the current score.