NPS measures one thing with great efficiency: how willing your customers say they are to recommend you, compressed into a score a board can track. What it cannot do, by design, is tell you why the score is what it is, or why it just moved. Customer feedback beyond NPS means pairing the number with the customer's actual voice, verbatim words and, increasingly, short video, so the same program that detects a change can also explain it. Keep the score for the board. Add the voice for the fix.
What NPS does well
Any serious conversation about going beyond NPS should start by being fair to it, because the metric earned its place. Fred Reichheld introduced it in a 2003 Harvard Business Review article titled "The One Number You Need to Grow," and it spread for good reasons.
One question gets answered at rates no long survey will ever see. The score is comparable across quarters, regions, and industries, which makes it legible to boards and useful for spotting movement. And movement detection is real value: when a branch or a release sends the number down eight points, something true happened.
The honest way to describe NPS is as a smoke detector. It's reliable at telling you something is burning. It is completely mute about what, where, and why.
What the number can't tell you
Picture the quarterly review every CX leader has sat through. The score dropped from 58 to 44. Everyone in the room asks the same question: what happened? And the score, having done its only job, has nothing further to say.
The open-text box under the rating was supposed to cover this, and mostly doesn't. A minority of respondents type anything at all, and what they type tends toward "it was fine" and "slow service," words stripped of the specifics and tone that would make them actionable. Averages compound the problem by blending segments: a score of 44 may contain delighted regulars and a furious cohort from one location, presented as a single lukewarm middle.
Then there's the gaming problem. When bonuses hang on the number, frontline staff learn to ask for tens, survey timing gets curated, and the instrument bends. Reichheld himself has publicly criticized how some companies game and misuse the score he created. None of this makes NPS worthless. It makes it incomplete, and the gap is precisely the part a CX leader needs to act: the reason.
What a 60-second video carries that a score can't
A short video of a customer talking about their experience delivers four things no rating scale produces.
The specifics, unprompted. Customers on video name the dish, the employee, the step in the booking flow that broke. They volunteer the detail you would never have thought to put in a questionnaire, because they're telling a story rather than completing a form.
The tone. "It was fine" said with a smile and "it was fine" said through a tight jaw are opposite data points that an open-text box records identically. Video preserves the difference, which is often the entire finding.
The face, and what it does internally. A branch manager can scroll past a row in a spreadsheet; almost no one shrugs off a real customer looking into the camera and describing their evening. Video creates urgency and alignment inside an organization in a way charts never have. Played at the start of a quarterly review, two clips reframe the following hour.
A reusable artifact. The same clip that explains the score is, with the customer's consent and your curation, social proof for your website and ads. No survey response has ever been worth publishing.
The honest trade-off: fewer customers will record a video than will tap a number, every time. That's fine, because the two instruments do different jobs. The score gives you breadth; the voice gives you depth. Pair them; don't swap one for the other.
How to pair the score with the voice
Keep NPS as the smoke detector
Change nothing about the score's cadence or its place in board reporting. Its comparability over time is the asset; protect it.
Put a voice capture at the moment of experience
The reason open-text boxes underperform is timing and effort: they arrive by email days later, attached to a chore. A voice capture works in the warm moment instead: a QR code where the experience ends for physical locations, a recording link in the post-purchase email for online ones, opening a browser page where the customer can talk for up to 60 seconds with no app and no account. For multi-location brands, attribution to the specific branch should be built into the link itself; the mechanics are covered in detail in the voice of the customer playbook.
Route by need, not by score
Every video lands privately with whoever owns the experience, alongside the customer's rating. Every customer, regardless of that rating, gets the same two optional choices: an invitation to share their experience publicly, and a personal follow-up if they want to hear from you. The customers who ask for follow-up get a human reply within 48 hours, which is where the post-purchase relationship is actually won. This is the design Outhentik ships: browser-based recording, per-location attribution, identical opt-ins for every customer, and a recovery inbox for the ones who raised a hand. Whatever you build with, the routing principle stands.
Bring video into the rooms where the score is read
Institutionalize one habit: no score is presented without voice behind it. A quarterly review opens with the number, then two or three short clips from the segment that moved. The number says what changed; the clips say why; the meeting spends its time on fixes instead of speculation.
What NOT to do
Don't bolt a "tell us why" text field onto the NPS email and call it voice of the customer. It measures the memory and patience of the few, days after the fact, in flat text.
Don't screen by score before asking for public reviews. The standard playbook of "send your promoters a Google review link" is review gating: selecting who gets asked based on their stated sentiment. The FTC's 2024 final rule on fake reviews makes the practice actionable with civil penalties of up to $51,744 per violation, and Google's review policy prohibits selectively soliciting positive reviews. If a public review invitation exists in your program, every customer receives it on identical terms.
Don't tie frontline bonuses to the score alone. You'll get tens, begged for at the counter, and an instrument you can no longer trust.
Don't curate the video stream into a highlight reel. The praise is pleasant; the uncomfortable clips are the program. If leadership only ever sees the smiling ones, you've rebuilt the problem you were solving.
And don't judge the video channel by survey-volume expectations. Plan for a minority of customers and a majority of insight.
What to expect, realistically
The capture point produces its first clips within days of being placed in the right moment, and a watchable pattern emerges over a few weeks. Participation stays a minority of customers indefinitely; depth, not volume, is what you're buying. The pairing proves itself the first time the score moves and, in the same week, you can sit in a room and show exactly why. Anyone promising a flood of videos or a guaranteed score lift is selling you a dashboard. What this actually delivers is the end of the "what happened?" silence in quarterly reviews.
Frequently asked questions
What is NPS and how is it calculated? Net Promoter Score comes from one survey question: how likely are you to recommend us, on a scale of 0 to 10. Customers answering 9 or 10 are promoters, 7 or 8 are passives, and 0 to 6 are detractors. The score is the percentage of promoters minus the percentage of detractors, giving a figure between -100 and +100.
What does customer feedback beyond NPS mean? It means collecting the reasons alongside the rating: verbatim customer words, and ideally short video, captured at the moment of the experience and attributed to the specific location or segment. The score detects that sentiment moved; the voice explains what to fix.
Should we replace NPS with video feedback? No. The score's strength is breadth and comparability over time; video's strength is depth and cause. Replacing one with the other trades one blindness for another. Keep the score as the detector and add voice as the explanation.
Is it legal to ask only promoters for Google reviews? No. Inviting only customers with high scores to post public reviews is review gating, which the FTC's 2024 rule on fake reviews penalizes at up to $51,744 per violation and Google's content policy prohibits. A compliant program offers every customer the same public review invitation, regardless of their score.
How many customers will actually record a video? A minority, and any honest vendor will say so. Asks made at the moment of the experience, in a browser with no app or account, perform far better than emailed requests days later, and each response carries far more usable detail than a rating ever will.
How should customer videos be used internally? Three ways: branch or product managers watch the clips from their own area and respond to customers who asked for follow-up; leadership reviews open with clips behind the headline number; and selected clips, with the customer's consent, become training material and public social proof.
Does this work for online businesses without a physical location? Yes. The capture point becomes a recording link in the post-purchase email or order confirmation page instead of a QR code at the counter. The customer experience is identical: a browser, up to 60 seconds, no app, and the same optional choices afterward.
Try Outhentik free for 7 days - no credit card required →
Ahmed Mustafa is the founder of Outhentik, which opens a direct channel between businesses and their customers: video testimonials, compliant Google review growth, and customer recovery from one flow. He built it for every leader who has stared at a moving score and wished it could talk.