About
When you’re one of the fastest-growing businesses in the lodging industry, customer support can quickly become a bottleneck.
With a product suite spanning PMS, channel management, and revenue management systems, Smartness supports thousands of lodging operators worldwide. But with growth comes a surge in support inquiries, and Smartness’ lean support teams needed a smarter way to scale without compromising experience.
The Support Challenge
Each product at Smartness has its own dedicated support team; small, efficient, and constantly under pressure to move fast.
“Each team handles between 1,000 and 1,500 tickets per month. We deal with 6,000–7,000 requests every quarter.”
Joseph | Customer Support Manager
Most inquiries weren’t complex technical problems. Instead, they were repetitive “How do I…?” questions Smartness already had documentation for.
The real challenge: speed.
With a fast-growing customer base, first response time had drifted into hours. Far from ideal.
That combination, high volume of repetitive questions and small teams, created consistent pressure. Smartness needed a way to keep quality high without endlessly expanding the support team.
So they turned to AI. But not before making a few painful mistakes.
The First Wave of AI: Impressive Numbers, Unhappy Customers
Operational Headaches: Heavy Maintenance and One-Way Integrations
The Pluno Challenge: “You’re Never Going to Be Better Than Our Current Solution”
Smooth Transition: From Skepticism to Plug-and-Play
The First Wave of AI: Impressive Numbers, Unhappy Customers
Pluno wasn’t the first AI solution Smartness tried.
“When I say I tried them all, I tried them all.”
Joseph started with smaller chatbot tools that looked great on paper:
“They had really good stats on their websites, but they lacked complexity. Our business is a bit outside of Zendesk’s standard intents, we’re not e-commerce. We have different flows.”
Those tools struggled with Smartness’ real-world complexity.
So Joseph moved to a major AI solution provided by a competitor to their ticketing system (they use Zendesk).
That tool did move the metrics, at least at first.
“I was happy with the results because our stats really jumped… We had lots of resolved tickets.”
But there was a catch.
What was buried inside those “resolved chats”?
Once Joseph started digging, he saw the real impact; customers weren’t getting faster help. They were getting long, robotic walls of text written in generic chatbot language, and often marked “resolved” even when nothing was fixed.
“You could feel frustration building up in our customer base.”
The experience looked like this:
“It felt like: I’m a chatbot, this is a wall of text. If you want to talk to someone, answer this exactly and you’ll be passed on. That’s not how you solve things.”
On top of that, the AI solution wasn’t escalating issues properly:
Tickets weren’t always routed to the right groups
Some tickets were marked “resolved” when they weren’t
A human had to constantly monitor different views in the Zendesk inbox to catch mistakes
“I needed someone checking a view constantly to catch open tickets or tickets that were not really resolved, and then manually pull them into another view.”
The metrics looked good, but the reality did not:
“The numbers of resolved tickets were high, but when you looked at the quality, it wasn’t great. After a few months, customers started complaining.”
Operational Headaches: Heavy Maintenance and One-Way Integrations
The problems weren’t just with the chatbot’s behavior, they were also structural.
Because the AI provider was a competitor to Zendesk, integration was limited as the communication between platforms was mostly one-way. Smartness had to rebuild and maintain knowledge in a separate system, and so naturally, the reporting was scattered and messy.
“We had to provide all the knowledge base content, maintain it, and do QA on a different platform. Reporting was all over the place. It wasn’t great.”
The AI solution was supposed to save time, but it created more maintenance instead.
That’s where Pluno entered the picture.
The Pluno Challenge: “You’re Never Going to Be Better Than Our Current Solution”
Joseph wasn’t looking for another AI vendor.
Pluno reached out and challenged him directly:
“You came to my door and said: We’re better than this. Let us prove it to you. I was skeptical.”
He agreed to a test, but with low expectations:
“I told Korbi (Pluno Cofounder): I’ll give you these 50 tickets because you’re never going to be better than our current solution.”
Pluno built a simulation using those 50 real-world questions.
When Joseph saw the results, his reaction changed completely:
Smooth Transition: From Skepticism to Plug-and-Play
Given their previous experience, you’d expect the switch to be painful. It wasn’t.
“The transition has been pretty smooth. It was almost just plug and play.”
Because Smartness had already structured a lot of content for the previous AI tool, they could recycle that setup:
“The other solution already had everything set up, so for Pluno we just reused what we had. It was a smooth transition.”
There was one major blocker, though: knowledge base separation by product/brand.
Smartness needed Pluno to split content into knowledge silos and ensure the AI pulled answers from the right product’s KB.
“We needed knowledge-based differentiation so we could switch some products first. You shipped that feature very fast. That was huge for us.”
Deep Zendesk Integration: Goodbye Maintenance Hell
Unlike the previous solution, Pluno integrates directly into Zendesk.
Pluno pulls from Zendesk’s own knowledge base, tickets, and context
Tools and workflows are all connected in one system
No need to maintain a separate AI platform
“Pluno resolves all that. Being integrated with Zendesk gives it a completely different pool to answer questions. Maintenance is super light, apart from a few responses here and there, which is normal.”
Instead of maintaining two parallel worlds, support now lives where it always has: inside Zendesk, with Pluno acting as a powerful layer on top.
Smarter Automation: How Pluno Actually Helps Agents
Pluno helps Smartness in two major ways:
1. Offloading How-To & Known Issues
Pluno’s primary target at Smartness:
“How do I…?” questions
Errors and edge cases already documented
“My main goal is to reduce the pressure that how-to questions put on support. These are all questions we have knowledge base for and ticket history for, so Pluno can tackle all this.”
Pluno now handles over 40% of tickets, and Joseph trusts that number:
“With the previous solution, we had lots of resolved tickets but poor quality. Now I know that if Pluno is taking over 40% of our tickets, those are real resolved tickets.”
2. Acting as a Smart QA Layer
Pluno isn’t just there for customers, it’s a productivity tool for agents too as it summarizes the entire ticket conversation, Identifies what’s missing and suggests the next action!
“Honestly, when I jump on a ticket now, I don’t even look at what the customer said. I have the summary from Pluno. It tells me: You have to enable that. I go, I enable it, and we’re done. Ticket resolved.”
It also knows when to escalate:
“Pluno knows when something needs to be escalated, when something is urgent, and it does it. If something hasn’t come to me, I know it’s not urgent and doesn’t need a human. This is fantastic.”
Human-Like Conversations: Meet Marco, the AI Agent
One of the biggest differences Joseph noticed?
Tone and formatting.
“The other solution was a wall of text, very robotic and formatted improperly. Pluno’s responses are conversational.”
The experience was so natural that the team gave the AI agent a real Italian name: Marco.
“We called our AI agent a real Italian name because it has conversations day and night with our customers.”
The results have been almost comical. Some customers thought Marco was a real person. They’d send kisses and messages like “You never sleep!”. Even the internal teammates tagged Marco in tickets as if he were a colleague.
“I had to tell my team: guys, it’s not real. He’s here, but he’s not a human. That’s how good it is at talking to people.”
Hard Numbers: From Hours to 22 Minutes
Pluno didn’t just “feel” better, it transformed key metrics.
The biggest win?
First Response Time
“Before Pluno, our first reply time was in the hours. Now it’s 22 minutes.”
That improvement came from Pluno handling a large proportion of simple tickets and providing high-quality ticket summaries that let agents respond in minutes.
“Thanks to Pluno’s summaries, we can jump into a ticket, spend a few minutes, and respond. The team is super happy. Our customers are happy too.”
Ticket Resolution & Reliability
40%+ of tickets are now handled by Pluno
Escalations are accurate and routed to the right teams
Agents no longer waste time “babysitting” views to catch misclassified tickets
And importantly, Joseph trusts the automation:
“Now I know that if a ticket is marked resolved, it’s actually resolved. And if something needs a human, Pluno escalates it.”
Changing Mindsets: Internal Teams and Customers
Not everyone is instantly comfortable with AI.
Joseph sees this every day, even though Smartness itself offers AI components to its own customers.
“We have our own chatbot for the lodging industry. Some customers use our AI to answer their guests, but don’t want to use AI for their own support. That’s the paradox.”
He believes the resistance isn’t to AI itself, but to bad AI experiences:
“You go to big companies like Just Eat or Deliveroo, and they all have chatbots. But many of them don’t do it well. Customers get blocked instead of helped. That’s where the skepticism comes from.”
He’s very clear on where AI fits:
“Support starts with real talent. If you think about using AI to replace talent, you’re doing it wrong. You should use AI to make your team’s work much more efficient. That’s what Pluno does.”
There’s also an education piece with customers:
Sometimes a customer will say “the AI gave me the wrong answer”, but never replied to correct it or continue the conversation.
“I tell them: if I gave you a wrong answer, you would tell me. Do the same with the AI. Tell it that it made a mistake. It’s a mindset shift.”
And if Pluno doesn’t know something, that’s also a useful signal:
“If it doesn’t have an answer, it means we need to improve our knowledge base. It’s on us.”
Scalability: Handling More with the Same Team
Joseph sees this every day, even though Smartness itself offers AI components to its own customers.
“We don’t want to replace people. We just don’t want to expand the team further.”
His goal is for Pluno to:
Handle 100% of how-to questions
Take care of light errors that don’t need engineering
Free human agents to focus on complex issues
To get there, Smartness is planning a complete overhaul of is knowledge base structure, internal workflows, and how content is written and organized.
“In a perfect world, Pluno handles 100% of how-to and light errors. To achieve that, we need to change how our knowledge base and workflows are structured. There’s a lot of work, but I know Pluno adapts really well to everything we do.”
Working with Pluno’s Team
Joseph is just as enthusiastic about the people behind Pluno as he is about the product.
“We chat every day, basically. You’re always there. You work day and night, you talk to me on weekends. I couldn’t ask for a better partner.”
He highlights two big things:
Responsiveness – questions and issues get fast answers
Openness to feedback – key features get shipped quickly
The knowledge base segmentation feature is a perfect example:
“That was a crucial feature for us, and you shipped it very fast. I’ve requested other features too, the important ones appear in the app after a very brief time. You know what has value and you ship it fast.”
“There’s No Reason Not to Give It a Try”
So, what would Joseph tell another company similar to Smartness that’s skeptical about AI in a crowded market?
“There’s no reason not to give it a try. Pluno is a very solid solution that doesn’t block customers from getting the answers they need, it does the opposite. It works with the customer to get there.”
And his practical recommendation:
“Give it your 50 questions. They’ll come back with a matrix of answers and you can analyze the results and decide. That’s what convinced me.”
For Smartness, that decision changed everything:
“If you hadn’t insisted on letting us try your product, we’d probably have a very angry customer base right now.”