How Welcome Pickups Cut QA Time by 93% While Reducing Agent-Driven Dissatisfaction by 22%

Alex

December 15, 2025

Welcome Pickups transformed their quality assurance from manual ticket reviews to AI-powered automation, achieving 100% coverage on negative reviews while cutting DSAT analysis from 2-3 days per week to just 2 hours.

In two months, they reduced agent-driven dissatisfaction from 50% to 39%, and freed their QA team to focus on strategy instead of spreadsheets.

The Problem: When Manual QA Can't Keep Up with Global Scale

Rodica Moroi, Quality Specialist at Welcome Pickups, had a routine that was slowly crushing her team's ability to improve.

Every single ticket review took 15 to 30 minutes. Multiply that across a support operation handling 8,000 to 20,000 interactions monthly (volume that could double during peak travel seasons) and the math became impossible.

"Each ticket review took 15 to 30 minutes," Rodica explains. "We couldn't review enough tickets to get statistically meaningful insights. As a result, QA focus was mostly on evaluations and coaching, not on strategy, insights, or proactive quality improvements."

Welcome Pickups operates in over 370 destinations across 116 countries, providing pre-booked, personalized airport transfers and sightseeing experiences.

Founded in 2015 in Athens, Greece, the company built its reputation on a simple promise: stress-free arrival experiences powered by carefully selected local drivers. But that promise extended to three distinct personas, including travelers, partners, and drivers, with each requiring tailored support at every step of their journey.

Want to understand why things didn't go so well in another interaction, and I just love digging into things and understand the why behind them. Okay, so my name is Radhika and I am a quality specialist here at Welcome Pickups. Welcome Pickups was founded back in twenty fifteen here in Athens, Greece and what we do is personalized and pre booked transfers. We offer a very seamless experience to travelers when they arrive at a destination. We are currently operating in more than three seventy destinations across the world. It's about one hundred and sixteen countries. We are using Intrigg our QA process in our support department. Basically though, we might use it in the support department, but the insights we get out of it can be shared with other teams as well. We try to collaborate closely with product, with operations, so we get insights both on the product itself and how the functionality and operational aspects, for example, the availability of drivers and how can it affect the overall experience of a traveler and the journey. Real case scenarios, I mean, having trainings on what we actually do is way better. We can move from the theoretical side of things from the PowerPoint slides, for example, to practice. And I think this is way more effective for training someone. And even for onboarding new people to the team, they get way more confidence and they are more ready to open their desktop and just respond to chat. It's very simple to use. The workloads are very easy to set up. The cycles of evaluations are there, easy to find, easy to extract any data that we need. But as I said, the reporting is there. It has trends monitoring. It has individual reporting. It has reporting for the evaluators as well. So it has everything someone might need. So it's very important to us to provide a seamless, stress free experience to travelers, but also to partners and drivers. We have different personas, and we have to make sure that we are there, we are present at every step of their journey and support them with everything they might need. So from a general question to how do I make a ride, to any issue that might arise with a transfer or a sightseeing experience. So it's very important for us to make sure that every interaction is warm and also welcoming, just as welcome stands for. INTRIC helps us in many ways, I'd say. For sure, it's very important, to talk about the time efficiency. For example, in the past, reviewing one ticket manually would take from, I don't know, fifteen minutes, or thirty to some more complex cases. Now in the same amount of time we can grade the whole amount of tickets from one day for the whole support team, not just for one agent. So it's very important because we have more time to actually dig into the insights that we get. We have more time for more strategic decisions, data driven decisions. A recent feature actually in TRIC released is the AI simulations. So basically we can assign some real case scenario trainings to agents. They have to respond to an AI bot just like they would do with a customer, with a partner, and they can train on a specific aspect, and we can monitor their performance and improvement. We have full visibility and control on both the insights and the training process. And I would also say that the sampling is also very nice and great. We can sample tickets based on so many different factors. As I said, we have many personas and we want to ensure that the support we offer is tailored and personalized to everyone. I, will not replace human judgment. It is here to enhance it. As a quality specialist myself, I felt like a big productivity boost. I had so much more time to focus on more significant, tasks rather than repetitive processes. So I, I would say to definitely give it a try. We wanna, discover how can we turn every experience into a great one.

The support team worked closely with operations and product to maintain service quality, which meant their insights needed to inform decisions beyond just agent performance. They needed visibility into product functionality, driver availability, and operational bottlenecks, which are issues that required comprehensive data, not cherry-picked samples.

But comprehensive data was exactly what they couldn't get.

The consequences compounded over time.

With limited ticket coverage, Rodica couldn't confidently identify team-wide trends or systemic issues and coaching became reactive, addressing individual tickets rather than patterns.

During peak travel seasons, when the team doubled in size and interactions surged to 20,000 per month, QA coverage became even more sparse, exactly when they needed it most.

The Solution in Action: From Spreadsheets to Strategy

Intryc fundamentally changed how Welcome's QA team spent their time.

Instead of manually reviewing a small sample of tickets using spreadsheets to track inputs on agent-driven versus non-agent-driven issues, the platform now evaluates 100% of negative reviews automatically.

The transformation is most visible in Rodica's Monday morning routine.

"Every Monday morning, we start the week with a complete set of insights on all negative reviews from the previous week, something that used to take days of manual work," she says.

Those insights inform coaching and flow to operations and product teams, helping identify issues with driver availability, product functionality, and operational processes that impact the traveler experience.

Key workflows enabled

Comprehensive DSAT Analysis: What once consumed 2-3 days per week now takes 2 hours. More importantly, the analysis is based on 100% of negative reviews rather than a small sample, which means Welcome Pickups can act on every dissatisfied interaction and understand exactly what happened.

Personalized Coaching at Scale: Instead of creating lengthy presentations or generic bullet points that need adjustment for each agent, Rodica builds tailored coaching sessions directly in Intryc. Discussion points generate automatically based on each agent's evaluations, ensuring every session addresses that agent's specific needs.

AI Training Simulations: A recent feature allows Welcome Pickups to assign real case scenario training where agents respond to an AI bot exactly as they would with a customer or partner. "They can train on a specific aspect, and we can monitor their performance and improvement," Rodica explains. "We have full visibility and control on both the insights and the training process."

Multi-Persona Sampling: With three distinct personas to support, which include travelers, partners, and drivers, Welcome Pickups needs flexible sampling to ensure quality across all interaction types. "We can sample tickets based on so many different factors," Rodica notes. "We want to ensure that the support we offer is tailored and personalized to everyone."

"As a quality specialist myself, I felt a big productivity boost," Rodica says. "I can finally dedicate my energy to meaningful analysis rather than repetitive tasks. I don't feel trapped in a routine anymore."

Results: 93% Time Savings, 22% Quality Improvement, 100% Coverage

The Numbers

Quality Impact:

Agent-driven dissatisfaction dropped from 50% to 39%, a 22% reduction in just two months

Coverage Transformation:

100% of negative reviews now evaluated (up from a small manual sample that couldn't provide statistical significance)
Complete visibility into every DSAT interaction across all three personas (travelers, partners, drivers)

Time Efficiency:

DSAT analysis: 2-3 days per week → 2 hours per week (93% time reduction)
Per-ticket comparison: 15-30 minutes for one ticket → same time now evaluates an entire day's tickets for the whole support team

Coaching Acceleration:

Significantly faster coaching cycles with automated, personalized session building
AI simulations enable proactive skill development on specific scenarios
Weekly comprehensive reports replace slow, delayed manual collection

The Scale

To understand the magnitude of efficiency gain, consider Rodica's comparison: "In the past, reviewing one ticket manually would take 15 to 30 minutes. Now, in the same amount of time, we can grade the whole amount of tickets from one day for the whole support team, not just for one agent."

At peak volume (20,000 monthly interactions, or roughly 650 daily), that's the difference between reviewing one ticket versus reviewing hundreds, which is a productivity multiplier that's difficult to achieve through any means other than AI automation.

Beyond Efficiency: Strategic Value

The time savings enabled doing QA better. With comprehensive data flowing weekly, Welcome Pickups can now:

Spot patterns early and prevent issues from escalating
Share actionable insights with operations and product teams
Make data-driven decisions about training priorities
Maintain consistent quality standards even as team size doubles during peak seasons
Ensure fair, representative evaluations for all agents

"It's not just about saving time," Rodica emphasizes. "It's about having visibility and control. I believe that the earlier you spot patterns, the faster you can act and prevent bigger issues down the line."

Table: QA Efficiency & Coverage

Metric	Before Intryc	After Intryc	Verified Impact
DSAT analysis time	2–3 days/week	~2 hours/week	Major time reduction
Coverage	Small manual sample	100% of negative reviews	Full visibility
Agent-driven dissatisfaction	50%	39%	Reduction

‍Key Takeaways

What made this transformation successful:

Trust in AI enablement, not replacement: Welcome understood that AI would enhance human judgment rather than eliminate it, freeing QA specialists for higher-value work.
Comprehensive data over perfect sampling: 100% coverage of negative reviews provided pattern recognition that statistically valid sampling never could at their scale.
Proactive practices: Weekly insight reviews allowed the team to spot and address issues before they became systemic problems.
Cross-functional sharing: QA insights flowing to operations and product teams multiplied the value beyond agent coaching.

Rodica's advice for other CX leaders:

"Definitely move from manual QA to AI-powered QA. It will not replace human judgment but will support QA peers by allowing them more time for actual QA work - like actually identifying patterns - with far more precise insights due to larger sample sizes."

She recommends reviewing insights regularly, ideally weekly: "Keeping that rhythm helps maintain a proactive QA culture instead of a reactive one."

The human element remains critical: "It is here to enhance it. As a quality specialist myself, I felt like a big productivity boost. I had so much more time to focus on more significant tasks rather than repetitive processes."

Cut Your QA Time from Days to Hours and Scale Coverage Like Welcome Pickups

See how Intryc helps quality teams achieve 100% coverage while freeing up time for strategic work that actually moves the needle.

Book an Interactive Demo

‍

"In the past, reviewing one ticket manually would take 15 to 30 minutes. Now, in the same amount of time, we can grade the whole amount of tickets from one day for the whole support team, not just for one agent."

Rodica Moroi, Quality Specialist at Welcome Pickups