A Voice of Customer programme works when it does three things: it hears from customers at the right moments, it turns that into numbers and reasons, and it puts the findings in front of people who can act on them every month. CSAT and NPS are the easy part. The harder part is making somebody with authority change something because of what customers said.
Measuring customer sentiment was the fifth pillar of the plan I described in building a three-to-five-year support plan. At Q4 we built a programme around two measures, CSAT and NPS, and I’ve done similar work at AudienceView, where it was NPS, and earlier at Tyco. This is what I’ve learned about making it useful.
What’s the difference between CSAT and NPS?
They answer different questions. CSAT, customer satisfaction, asks how happy the customer was with a specific interaction: the ticket they just had resolved, the call they just finished. It’s close to the work, so it tells you how your team is performing. NPS, the Net Promoter Score, asks how likely the customer is to recommend you. It’s about the relationship, not the transaction, so it tells you something about loyalty and, indirectly, about the risk of losing the account.
Using both makes sense for a support organisation because they catch different problems. A customer can be satisfied with every ticket and still be a poor bet to renew, because of something support can’t fix, like the product, the price or a missing feature. And a customer who loves the company may still rate a particular ticket badly because it took too long. If you only had one you’d mistake one situation for the other. I looked at the second in understanding NPS and whether it makes sense and the argument hasn’t changed: NPS is useful when you treat it as a conversation starter, and misleading when you treat it as a verdict.
Be careful with small numbers. Show me a support operation with fewer than 50 survey responses a month and I’ll show you a metric that’s being driven by two or three outliers. One angry customer can swing a score for a quarter. If your volume of responses is low, report the trend over a longer period and say how many responses sit behind the number, so nobody reads a swing as a change in performance.
How do you get customers to answer?
By asking at the right moment, with the right question, in the fewest clicks. A CSAT survey that arrives with the ticket’s closing message, with one question and one click, gets far more answers than a long form sent a week later. Keep it short, and make the optional comment box obvious, because the comment is where the information is.
Think about who you’re asking. Our customers are investor relations professionals, busy and often under deadline. A survey that interrupts them during an earnings week won’t get a good reception. We paid attention to timing so we weren’t adding to their workload when they were at their most stretched. Know your customers’ calendars and respect them.
And close the loop. When a customer gives a low score or a critical comment, someone should get back to them promptly. It’s one of the best retention actions available, because it shows the feedback went somewhere. A customer who gets a call about a bad experience often ends up more loyal than one who never had a problem. A customer who’s ignored remembers it.
How do you turn scores into reasons?
With tagging. A score tells you how someone felt. It doesn’t tell you why, and without the why you can’t fix anything. In my Tyco work, documented in the Six Sigma and Voice of Customer case study, we asked agents to tag each ticket with a secondary reason code from a list of 18 categories covering product gaps, documentation problems and support experience failures. It added roughly 30 seconds per ticket. In return, we got a monthly view of the real distribution of customer pain points, and three product-level changes came directly from it in the first year.
That principle has stayed with me. VOC isn’t a survey. It’s a data collection discipline built into the operation. If your agents tag tickets consistently, you already hold more customer intelligence than most product teams see in a year. The survey adds the customer’s own words and a score. Put the two side by side, a low score and a tag on the same ticket, and you can see which kinds of problem drive dissatisfaction.
Be disciplined about the taxonomy. Too many categories and people stop using them properly. Too few and everything lands in “other.” Start with a short list, test it on real tickets, and revise after a quarter. The same cleanliness that matters for routing matters here. In fact, they’re the same job: a clean set of categories serves routing, reporting and VOC at once.
Who needs to see the results?
The people who can change things, and not only the support team. This was the main lesson from Tyco. VOC data is only valuable if it reaches decision-makers, so I formatted it for product and leadership audiences and put it in front of them every month. Data that never leaves operations changes nothing. A product manager who sees that a particular feature generated the most frustration last quarter will make a different decision from one who has never seen the number.
Give each audience what they need. Leadership wants a short summary: the scores, the trend and the three biggest themes. Product wants the detail behind the themes, with real customer quotes. The support team wants to know what customers said about them, good and bad, and what’s being done. A single report for everyone satisfies nobody, so build a few views from the same data.
Also report what changed as a result. Nothing destroys a VOC programme faster than customers and staff asking “what happens to this?” and getting no answer. A short “you said, we did” list, circulated every quarter, shows that the effort has consequences. It encourages customers to keep answering and it earns the programme its budget.
How do you connect VOC to the rest of the operation?
Through the scorecard and through training. When customers consistently mention something about the way a type of request is handled, it should show up in coaching and in the training curriculum. I explained the principle in 9 crucial metrics to measure customer support team performance: choose measures that reflect the customer’s experience, and connect them to what the team does day to day.
And let it shape the culture. I wrote about cultivating a customer-centric culture, and the practical way to do it is to bring the customer into the room. Read real comments out in team meetings, positive and negative. Let the people who earned a compliment hear it. Share the harsh ones without blame, as problems to solve. A team that hears its customers regularly starts to think about them without being asked.
Think about governance, because a VOC programme needs an owner. Someone has to maintain the survey, check the taxonomy, produce the monthly report and chase the actions. If it’s everyone’s job, it’s nobody’s. In a support organisation that owner is usually someone close to the operation who has the trust of both the team and the people the findings are meant to reach. Give them time to do it properly, and protect that time when a peak arrives.
Be careful about incentives. Once scores become targets, people find ways to move them. Agents ask customers for good ratings, difficult cases get quietly excluded, and the number improves while the experience doesn’t. Use scores to understand and to coach, not to rank people against each other, and look at the comments as well as the number. The words are harder to game than the digits.
Last, be patient. A VOC programme takes a few quarters to produce anything you can trust, because you need enough responses and enough history to see a pattern. The first report will be thin. Publish it anyway, say what it can and can’t tell you and improve it. A modest report that arrives every month is worth more than a polished one that arrives once.
Measurement without action is a hobby. The scores matter less than what you do on Monday because of them.
Hutch Morzaria is a Director-level CX and Support Leadership professional with 19 years of experience building global support organizations across SaaS, Fintech, and enterprise technology. He has hired dozens of support and CX leaders across his career and holds ITIL Expert certification across V3 and V4.
