Full score distribution
Where are the customers who are almost ready to recommend us?
Beyond the NPS itself, you see the share at each score. Here, 18% gave an 8: they’re one step away from becoming promoters.
NPS research that takes you from measurement to analysis to an action plan, based on the score and the reasons behind it.
I ordered on Monday and it didn’t arrive until the following week. Nobody let me know.
I almost gave a 10. They just should have told me about the delay.
They sorted everything out on the first call.
“Our NPS dropped 8 points and nobody can tell me why.”
“We’re sitting on thousands of comments nobody reads.”
“The score becomes a target, but it never turns into action.”
On its own, the number tells you how big the problem is. The real value is in the why, and in the analysis that connects the two.
Net Promoter Score (NPS) measures customer loyalty with a single question, answered on a 0–10 scale.
“On a scale of 0 to 10, how likely are you to recommend [company] to a friend or colleague?”
% promoters − % detractors
52% − 20% = 32
The score ranges from −100 (all detractors) to +100 (all promoters). Passives count toward the total number of responses but neither add nor subtract.
Scores are often grouped into four bands that show where you stand and what the next step up looks like.
As many detractors as promoters, or more: a real risk of losing customers and reputation.
Promoters outnumber detractors, but there are gaps holding recommendations back.
A good experience, with a few specific things to fix on the way to excellence.
Customers who have become brand advocates: the challenge is to keep it that way and put your promoters to work for the business.
Knowing your zone is a start. The analysis shows what it takes to reach the next one.
These zones are a market convention widely used in Brazil, not part of the original NPS methodology. Benchmarking against your industry and your own history is more accurate.
Two identical scores can hide very different realities.
Plenty of fans and plenty of unhappy customers: an uneven experience, with the risk sitting in the detractors.
Few fans and few unhappy customers: a lukewarm experience, where the challenge is to genuinely delight.
Right after the score, customers tell you why. That’s where most of the value in NPS lies. Tap a score to see the follow-up question.
On a scale of 0 to 10, how likely are you to recommend Company X to a friend or colleague?
What would we need to change for you to recommend us?
What would it take to get a 10 from you?
What do you value most about us?
Each analysis answers a business question, and they fall into three groups: what the score tells you, what the comments tell you, and what you learn by putting the two together. All examples use illustrative data.
Distribution, segments, trends, competitors, drivers and business outcomes.
Where are the customers who are almost ready to recommend us?
Beyond the NPS itself, you see the share at each score. Here, 18% gave an 8: they’re one step away from becoming promoters.
Which channel, location or profile is the problem?
By region, location, channel, product, profile and customer tenure. Here, the call center is dragging the total down.
Is this quarter’s drop real or just chance?
The test shows whether a change between surveys is real, so you avoid false alarms and celebrating too early.
Is our NPS good for our market?
Your NPS compared with others in your market, using the same question and the same yardstick.
Delivery time, service or price: which one matters most for recommendation?
We cross the score with attribute ratings to show which ones have the biggest impact on recommendation.
What is a promoter worth compared with a detractor?
When you share customer data, we link NPS to repurchase, churn and spend.
Themes, reasons by group, the cost of each theme, tone, emerging themes and real quotes.
What do customers talk about most?
Comments sorted into categories, with how often each one comes up.
Are the things that drive customers away the same as the things that win them over?
They’re usually different themes: fixing what drives people away isn’t the same as doubling down on what wins them over.
Which themes take the most points off your NPS?
We compare the NPS of customers who mention a theme with the overall NPS. The bigger the gap, the more that theme is costing you.
Are the comments praise, mild criticism or serious complaints?
Intensity separates a minor annoyance from a problem that could lead to churn.
What topic is starting to come up?
Mentions of in-app billing rose from 1% to 9% in six quarters: a problem growing before it shows up in the score.
How do you bring the numbers to life in a presentation?
Real quotes that help leadership see the customer behind every number. Illustrative examples.
Priorities, estimated gains, detractor profiles and the action plan.
Where do we start?
How often each theme is mentioned versus how many NPS points it costs. Circle size shows how many customers raised it.
How much could NPS go up if we fix each problem?
Select the problems you’d fix. The estimate is based on the NPS of customers who mentioned each theme. Illustrative data.
Who are they, what do they complain about and which channel do they use?
Groups of detractors with a similar profile, so each can be addressed directly.
Who does what, and how will we know it worked?
| Team | Action | Metric |
|---|---|---|
| Logistics | Proactive delay alerts and revised delivery times by region | On-time deliveries and NPS among customers who mentioned delivery |
| Financial | A simpler invoice that explains each item | Billing mentions in comments |
| Customer service | Empower agents to resolve issues on first contact | First-contact resolution rate |
| Digital product | Order tracking inside the app | App channel NPS |
Recommendations by responsible team, each with a metric to track.
Measures the overall relationship with your brand, in quarterly, twice-yearly or annual waves.
Measures a specific interaction, right after a purchase, a service call or a delivery.
Market research that includes your competitors’ customers, using the same yardstick.
The same approach applied to employees: how likely they are to recommend your company as a place to work.
The survey doesn’t end with the score. The customer who gave it deserves a response, too.
Every detractor triggers an automatic alert, so your team can follow up fast.
Customers hear back about what changed because of their feedback.
We track how many contacted detractors moved to a different group in the next survey.
120 detractors contacted after wave 1
The six steps of the first wave. In later waves, the cycle starts again at data collection.
Goals, touchpoints and, where available, a review of the NPS you already track.
Questionnaire, the why question, supporting questions, sample and frequency.
Data collection through the most suitable channel, with quality control.
Analysis of the score, the comments and the two together.
Dashboard, report and presentation with the action plan.
Following up on the actions and preparing the next wave.
The numbers on the dashboard match the deliverables listed here.
Done! The dashboard has been sent to . Demo only: no email is actually sent.
On a scale of 0 to 10, how likely are you to recommend Company X to a friend or colleague?
Honestly, I’d give it a 4.
And what would need to change for you to recommend ABC?
I ordered on Monday and it didn’t show up until the following week. And nobody told me it was running late.
If they’d just sent me a quick message, I wouldn’t have been this annoyed.
On a scale of 0 to 10, how likely are you to recommend Company X to a friend or colleague?
I’d say an 8.
What would it take to get a 10 from you?
The staff at the store are great, but the checkout line on Saturdays is way too long.
If they opened more registers, I’d give them a 10, no question.
On a scale of 0 to 10, how likely are you to recommend Company X to a friend or colleague?
A 10, definitely.
What do you value most about us?
I had an issue with my bill and they sorted it all out on the first call.
The agent was really friendly and walked me through every line item.
Comment analysis is part of the project, not an add-on.
Differences between waves and segments are tested, so you don’t chase random swings.
The attributes and themes that weigh most on recommendation, ranked by impact.
Recommendations with an owner and a metric to track each action.
More than 30 years in research and over 500 projects in Brazil and abroad.
Processes certified to ISO 20252 since 2010, and data protection in line with Brazil’s LGPD.
Clients are identified by industry only, as in all our case studies.
Understand why public-sector clients of a financial institution were giving low satisfaction scores.
Online in-depth interviews with public-sector clients and with the teams that serve them.
An improvement agenda built around what clients and staff both saw as priorities.
Understand why customers of a digital reordering service had stopped placing orders.
Phone survey of inactive customers, first confirming each customer’s actual status.
Fixes to the customer journey and different win-back actions for each type of inactive customer.
Market zones give you a first reference point: above 75 is excellence, 51 to 75 is quality, 1 to 50 is improvement, and 0 or below is critical. These zones are a convention widely used in Brazil, not part of the original NPS methodology. The most accurate comparison is with competitors in your industry and with your own track record.
Talk to a specialistNPS is a single metric: it measures, on a 0–10 scale, how likely a customer is to recommend you. A satisfaction survey is the broader study, covering service, price, delivery times, quality and every touchpoint. The two go hand in hand: NPS is usually one of the metrics tracked in a satisfaction program.
Learn about satisfaction researchIt depends on how scores are spread and how many groups you want to compare. As a rule of thumb, about 400 responses give a margin of error of 7 to 8 NPS points, and about 1,000 responses bring it down to around 5 points, at 95% confidence. To compare locations or channels, each group needs a large enough sample.
Open the sample size calculatorYes. Send us your file with scores and comments and we’ll run the same analyses: distribution, segments, period-over-period tests, comment themes and the impact of each theme on the score. It’s a great starting point for a diagnostic before you change how you measure.
Talk to a specialistTransactional NPS works well on an always-on basis, right after each interaction. Relationship NPS is usually measured in quarterly, twice-yearly or annual waves. What matters is keeping the same interval and the same question, so results stay comparable.
Talk to a specialistThe data you share is used only for the research, with access limited to the project team, and handled in line with Brazil’s data protection law (LGPD). Results are reported in aggregate, and a customer is only identified to you with their consent, as with detractor alerts. Our processes follow the ISO 20252 standard.
Talk to a specialistRequest a proposal or, if you already track NPS, send us your data for a diagnostic: you’ll get a sample of our analyses run on your own data. A specialist will get back to you within one business day.
I’d rather chat on WhatsApp