How Ounass Turns Customer Feedback Into Better Service

A customer who has just spent the price of a small car on a handbag does not complain about the leather. She complains that nobody replied to her WhatsApp message for four hours. That is the paradox of luxury e-commerce in the Gulf: the product rarely disappoints, but the service layer around it — the speed, the tone, the follow-through — is where loyalty is actually won or lost. Ounass, the Dubai-based luxury platform owned by Al Tayer Group, has built its customer experience model around that insight, using customer feedback, data analytics and messaging channels to continuously refine how it serves shoppers.

That is the short answer to how Ounass turns feedback into better service: it does not treat feedback as a satisfaction score to report upward, but as an operating input — read, analysed and routed through the same conversational channels customers already use, so that what a shopper says today shapes what she experiences next time. The more interesting question for other CX leaders is not whether Ounass listens. Every brand claims to listen. It is what separates listening that changes behaviour from listening that just produces a dashboard.

What is Ounass, and why does its CX model matter?

Ounass is a luxury fashion and lifestyle e-commerce platform operating across the Middle East, built by Al Tayer Group — the Dubai conglomerate that also holds the regional franchise rights to brands including Harvey Nichols and Bloomingdale's. It launched to serve a customer who had, until then, been asked to choose between the convenience of mass-market online retail and the service standard of a physical luxury boutique. That positioning only works if the digital experience can replicate the attentiveness of an in-store personal shopper — remembering preferences, anticipating needs, resolving problems before they become complaints.

That is a much harder promise to keep online than in a boutique, because the feedback that would normally come through facial expressions, hesitation at a rack, or a quiet word to a sales assistant has to be captured, interpreted and acted on through entirely different signals: reviews, support conversations, return reasons, and behavioural data. An analysis of Ounass's CX operations shows the platform actively uses customer feedback, data analytics and messaging channels to optimise service delivery — the digital equivalent of a boutique manager watching the floor and adjusting in real time.

How does Ounass turn customer feedback into better service?

The mechanism is straightforward in principle and difficult in execution: feedback is gathered through direct messaging and service interactions, combined with data analytics on behaviour and transactions, and used to adjust how service is delivered. The discipline is in closing the loop quickly enough that the customer who raised the issue — or the next one with the same issue — actually experiences the fix.

Most retailers collect plenty of feedback. Few act on it inside a window that still matters to the customer who gave it. Fred Reichheld, in his 2003 Harvard Business Review article The One Number You Need to Grow, made the case that the value of customer feedback lies almost entirely in what a company does with it afterwards, not in the elegance of the measurement. A platform built on conversational channels has a structural advantage here: a complaint typed into a messaging thread is already sitting in the channel through which a resolution, an apology, or a personalised gesture can be delivered — no handoff to a separate survey tool, no delay waiting for a quarterly report to surface the pattern.

Why do messaging channels change the economics of feedback?

Because they collapse the distance between complaint and resolution. A star rating submitted after checkout is a data point. A message typed to a human, in real time, is a relationship moment — and it carries a behavioural cost that email surveys don't: the customer has invested effort in writing it. Ignoring that effort reads as a far bigger slight than ignoring a tick-box survey, which is precisely why response speed on messaging channels does more for perceived service quality than almost any other single CX lever in luxury retail.

There is also a quieter advantage: reciprocity. When a brand responds personally and promptly to a message, it signals that the relationship is being taken seriously — and customers who feel that effort tend to repay it with patience on the next issue, and with the kind of unprompted positive word-of-mouth that no loyalty programme can buy. This is the same principle behind why a handwritten note from a boutique manager outperforms a generic discount code: the gesture is disproportionate to its cost, and that disproportion is what makes it memorable.

What's wrong with treating feedback as a satisfaction score?

Scores flatten nuance. A Net Promoter Score tells a brand that satisfaction dipped two points this quarter; it does not tell anyone which touchpoint caused the dip, for whom, or why. Bain & Company's own 2005 report, Closing the Delivery Gap, found that 80% of companies believed they delivered a superior customer experience, while only 8% of their customers agreed — a gap that exists precisely because companies measure sentiment in aggregate while customers experience service one interaction at a time.

Luxury retail cannot afford that gap. A shopper spending at the top of the market has, by definition, more alternatives and less patience. Treating her feedback as a line in a satisfaction tracker, rather than as an instruction for how to behave with her specifically next time, wastes the single most valuable thing she has given the brand: context. This is where data analytics earns its place alongside messaging — not to replace the human conversation, but to connect it to purchase history, browsing behaviour and prior service interactions, so that a complaint about late delivery becomes a note on her account rather than a ticket that closes and disappears.

How should luxury brands close the loop without cheapening the moment?

This is where the peak-end rule earns its keep. Daniel Kahneman's research with colleagues — published in their 1993 study on experienced versus remembered utility, and extended in later work — showed that people judge an experience overwhelmingly by its most intense moment and how it ends, largely ignoring its duration. Nielsen Norman Group's summary of the peak-end rule makes the practical point clearly: a service recovery that ends well can outweigh a rocky middle entirely.

For a luxury platform, that means the way a resolved complaint ends matters more than how quickly it was logged. A customer whose delayed order was followed by a genuine apology, a clear explanation and a small, well-judged gesture will often remember the recovery more fondly than she remembers the original failure — provided the gesture feels proportionate rather than transactional. A blanket discount code sent automatically does the opposite: it signals that her specific complaint was processed by a rule, not read by a person. Luxury customers notice the difference between a response that was written for them and one that was written for anyone.

What can other CX leaders copy from this model?

None of this requires Ounass-specific access to copy the underlying discipline. The pattern is portable to any brand serious about turning feedback into service improvement rather than a quarterly report:

1. Collect feedback where the customer already is. A complaint typed into an existing chat thread will get more honest, detailed input than one requested through a separate survey link days later.
2. Route it to the people who can act, not just the people who can report it. Feedback that lands in an analytics dashboard before it reaches a service agent has already lost its usefulness.
3. Close the loop inside the window that still matters. A resolution delivered a week later reads as administrative. One delivered the same day reads as care.
4. Connect the complaint to the customer's history, not just the transaction. Data analytics should tell the agent who this person is before she has to explain it again.
5. Design the end of the recovery deliberately. Apply the peak-end rule on purpose — the last thing the customer hears or receives should be the most considered part of the interaction.
6. Feed the pattern back into design, not just into training. If the same complaint recurs across many customers, the fix belongs in the journey design, not in another round of coaching.

The sixth step is the one most organisations skip. Training agents to apologise better treats a symptom. Redesigning the journey that produced the complaint treats the cause — and it is the only version of "listening to customers" that actually reduces complaint volume over time rather than simply improving how complaints are handled.

Where does feedback-driven service break down?

The honest risk is volume. A platform that actively invites conversation through messaging channels will receive far more unstructured feedback than one hiding behind a ticketing form — and unstructured feedback is expensive to process well. Without disciplined categorisation and clear ownership, a flood of messages produces noise, not insight, and agents start answering reactively instead of pattern-spotting. This is also where loss aversion can distort decision-making internally: a single vivid, angry message tends to get more organisational attention than ten quieter but more representative ones, because the fear of one more escalation looms larger than the slower cost of ignoring a recurring, low-drama pattern. Richard Thaler and Cass Sunstein's work on choice architecture, set out in their 2008 book Nudge, is a useful reminder here: the structure around a decision shapes the decision as much as the information feeding it does, which is why feedback needs clear triage rules, not just an open inbox and good intentions.

The second risk is personalisation without judgement. Data analytics can tell a brand what a customer bought, returned and complained about — but using that history clumsily, in a way that feels surveilled rather than attentive, erodes the very trust the feedback loop was meant to build. The line between "this brand knows me" and "this brand is watching me" is thinner than most CX teams assume, and luxury customers, who are more attuned to subtlety than most, notice when it is crossed.

What does this mean for the wider region's e-commerce sector?

Ounass operates in a market where several platforms are experimenting with similar feedback-to-service pipelines, each adapting the same underlying principle to a different customer base. The comparison is instructive: a platform like Noon has built its credibility on operational trust rather than curated luxury service, as explored in Ounass's regional counterpart Noon's approach to trust-first CX, while international marketplaces such as Flipkart have shown how structured feedback programmes can be scaled across a much larger and more price-sensitive customer base, detailed in how Flipkart converts customer feedback into service improvements. What they share is a refusal to treat feedback as a compliance exercise. The sector is converging on the same conclusion from different starting points: feedback that does not change the next interaction is not feedback management, it is record-keeping.

That convergence has a governance dimension too. As more channels generate more feedback, the question of who owns it — service, marketing, product or a dedicated CX function — becomes a real organisational fault line, one explored in depth in the governance question of who should own Voice of Customer. Without a clear answer, even the best messaging-channel feedback loop eventually stalls on internal politics rather than customer apathy.

A complaint answered in a dashboard has been measured. A complaint answered in the channel the customer used to raise it has been heard — and the difference shows up in whether she buys again.

The underlying lesson for any CX or experience leader watching Ounass's model is this: the channel you collect feedback in is not neutral. It sets the customer's expectation for how fast, how personal, and how final the response will be. Build the feedback loop in a structured voice-of-customer strategy, feed it with the discipline of rigorous customer feedback management, and apply it across the full digital retail journey — not just the support inbox — and feedback stops being a reporting exercise and starts doing what it was always meant to do: making the next interaction better than the last one.

Luxury retail has always sold more than the product. It sells the feeling of being known. The brands that will own that feeling over the next decade are not the ones with the most elegant satisfaction surveys — they are the ones willing to read every message, act on most of them, and let the ending of a resolved problem say more about them than the problem ever did.