Share article
Almost every company I talk to tells me they have a “VoC programme” Voice of Customer, the systematic practice of listening to what customers say. Then I ask what happened to the last big insight from the last survey. Silence. Or: “it goes into a report.” And who reads that report? More silence.
That’s not a VoC programme. That’s a collection of questionnaires with a nice dashboard attached. A real VoC programme can be recognised by one thing alone: whether it actually changes anything. Not by how many NPS surveys (Net Promoter Score – a loyalty metric based on how likely a customer is to recommend the company) get sent out.
Qualtrics XM Institute has been tracking this for years, and the numbers are pretty grim. In its latest major survey on the state of CX management, over two-fifths of companies are still stuck at the very first maturity stage, where CX isn’t even seen as a strategic opportunity yet, and only two percent have reached the top level, where customer experience is genuinely built into everyday decision-making. And that’s talking about large companies with a thousand-plus employees, not smaller players, where the reality tends to be even rawer.
So let’s build this up step by step, from the ground.
What a VoC programme is actually made of
A VoC programme rests on four building blocks. Miss even one, and the programme doesn’t work it just takes a while before anyone notices.
Data sources. This is where most companies make their first mistake assuming VoC equals a survey. It doesn’t. A real VoC mix combines three types of data: transactional feedback (CSAT or CES after a specific interaction – a purchase, a support call, a complaint), relational feedback (a regular NPS-style survey measuring the overall relationship with the brand), and passive data that customers don’t fill in themselves support tickets, calls, reviews, website behaviour, chat transcripts. Smaller companies almost always ignore this third category, even though it’s the cheapest and most honest. A customer tells you more truth in a complaint than in a survey where you’re asking them to rate you right after a purchase.
Analysis. This isn’t just about averaging a score. It’s about finding the pattern why people give a bad rating, not just how many of them do. Without the “why”, you have nothing to act on, just a number that goes up or down with nobody knowing the reason.
Action. This is the heart of the whole programme, and also where most companies bury it. Every finding needs an owner, a deadline, and a clear definition of what counts as “resolved”.
Reporting. Not a report for the sake of a report. A report that reaches the people who actually have the power to change something, and shows them exactly what needs to change not fifty charts that everyone reads a different conclusion into.
Bain and Harvard Business Review have been making a similar point for years Rob Markey, Fred Reichheld and Andreas Dullweber, in their article on closing the feedback loop, show that the most successful companies don’t route feedback through a centralised research team, but send it straight to the frontline staff who actually spoke with the customer. Those staff then call the customer back themselves and find out, in a direct conversation, exactly what happened. The data isn’t filtered through three layers of management before it turns into action.
How to start small and why that’s the only way
Almost every company launching a VoC effort for the first time makes the same mistake: they want a fully-fledged programme straight away. Multi-channel collection, predictive analytics, CRM integration, a dashboard for every department. It sounds great on a slide for leadership. In practice, it means that three months later, nothing has actually launched, because there’s simply too much to coordinate.
The opposite approach works. Pick one touchpoint say, a resolved complaint, or the first month after onboarding – and build a complete, small loop around it: collection, analysis, action, feedback to the customer. Not a big programme with a weak action component, but a small one that genuinely and visibly changes something.
In practice, that looks like this:
- Pick one moment in the customer journey that hurts the most where you’re losing customers, or where most complaints come from.
- Set up short, targeted feedback collection at exactly that point. Not a generic annual survey.
- Decide who reads the results every week a name, not a department.
- Set a rule: every systemic finding (not a single complaint, but a recurring pattern) must get a response within two weeks – either an action, or an explanation of why no action will be taken.
- Once this has run smoothly for three months straight, add another touchpoint.
Only once this first loop is alive and people respect it do you widen the scope. Not before. I’ve seen it happen again and again – companies that wanted to “do VoC properly” from day one ended up with a beautiful platform and zero action, because the organisation simply couldn’t process the volume of data it had generated for itself.
Connecting to decisions, not to a report
This is where it all comes down to. VoC data needs to land where decisions are actually made not in a standalone CX report read by a handful of people on the CX team and nobody else.
A practical question I get asked a lot: who should own the action arising from feedback – the CX team, or the operational department? The answer, backed up by practice across companies, is clear: operational teams. If the CX team designs the action plan and a completely different team has to deliver it, you get exactly the tension that kills programmes: “nice idea, but we have other priorities.” When the team that actually runs the process has both access to the data and responsibility for acting on it, things happen faster and stick longer.
The second piece is executive sponsorship. Qualtrics XM Institute reports that over 60 percent of companies with a functioning CX programme have senior executive sponsorship, and the gap between mature companies and those lagging behind in CX comes down precisely to how consistent that top-down support is. Without someone from leadership bringing VoC data into meetings and asking what’s being done about it, the programme stays isolated within a single department and dies a quiet death at the first reorganisation.
One more specific thing that’s often underrated: an SLA (service level agreement, a clearly defined timeframe and standard) for response. Not for resolution – for response. A company doesn’t have to solve every problem within a week. But within a few days, it needs to be able to say: “we see this, we’re on it, here’s how long it’ll take.” A customer who gives feedback and hears nothing for two months won’t bother giving it again next time.
Why VoC programmes fall asleep and how to keep them alive
Here are a few reasons I see over and over:
Feedback gets collected, but nobody closes the loop. A customer gives feedback, the company processes it internally, but the customer is never told what happened to it. Over time, customers notice that filling in the survey leads nowhere, and they stop responding. Response rates drop, the data gets less representative, trust in the results drops along with it and it’s a downward spiral.
The programme gets measured by vanity metrics, not impact. The company celebrates NPS going up three points. Nobody asks whether those three points meant lower churn or higher CLV (customer lifetime value the total value a customer generates over their relationship with the company). The score itself doesn’t move the business; only the action the score triggers does.
Competing priorities. No company is short on important things to fix. VoC loses when it doesn’t have a clear owner with real authority and calendar space set aside for action. Qualtrics confirms that “competing organisational priorities” is consistently the most commonly cited barrier to CX success, and significantly more so among less mature companies than among more advanced ones.
The programme grows faster than the organisation can process. A new channel gets added, a new segment, a new set of questions but the capacity for analysis and action stays the same. Data piles up, nobody keeps up with processing it, the programme looks alive from the outside (dashboards are running), but it’s dead inside.
How do you keep it alive? The basic principle is simple, and it works: smaller scope, faster loop, visible action. Better to have fewer questions with a higher response rate and a faster reaction than a monstrous quarterly survey nobody finishes reading. And communicate internally, regularly, what’s changed as a result of feedback – an internal channel showing “here’s what you told us, here’s what we did about it” does more to keep the energy up than any dashboard.
A VoC programme isn’t a project you launch once and let run itself. It’s an operational discipline, just like accounting or workplace safety. It needs an owner, a rhythm, and visible results. Without that, it stays a collection of nice numbers nobody cares about which is exactly where most companies are today.
