Latest / The Operator's Manual with Fexingo: Daily Decisions, Tactical Wins, and Business Execution / How a Single Question Cut IT Support Tickets 70 Percent
Transcript
- Lucas: So you're a SaaS company, you've got 15,000 active users, and every month your support team fields around 4,200 tickets. That is a lot. Most leaders would look at that number and think — we need more headcount, or we need a better knowledge base, or we need to build a chatbot. Luna: Right. The classic reflex. Throw more bodies or more software at it. Lucas: Exactly. But the VP of Customer Experience at a company called CloudHive — mid-sized, B2B, project management and collaboration software — she did something different. She spent three months categorizing every single ticket by root cause. And what she found was that 70 percent of them fell into one bucket: users couldn't find a feature that already existed. Luna: So not bugs, not missing features, not billing questions. Just 'I can't find the thing that's already there.' Lucas: Correct. And that's a design problem, but not in the way most people think. It's an information-architecture problem. The features were there, they worked, but the user didn't know that clicking a certain icon or navigating to a certain tab would do what they wanted. So the VP asked herself: what's the smallest possible intervention that could bridge that gap? Luna: And the answer was one question. Lucas: One question. On the user's first login screen — and actually, for existing users they rolled it out as a one-time modal — they added a dropdown that said: 'What is your primary goal today?' with five options. 'Set up a new project,' 'Integrate with another tool,' 'Generate a report,' 'Invite team members,' 'Something else.' That's it. Luna: And when the user selects one, what happens? Lucas: If the user picks an option that matches a feature they haven't used yet — say they pick 'Integrate with another tool' but they've never opened the integrations page — a single sentence appears. 'You can do this under Settings > Integrations.' No video, no tutorial, no link to a help article. One sentence. Luna: And that cut tickets by 70 percent? Lucas: Within 60 days. They went from 4,200 tickets per month to 1,260. And the drop was almost entirely in that 'I can't find it' category. The other categories stayed flat. Luna: So the fix wasn't about adding more information. It was about giving the right information at the exact moment the user needs it. Before they even know they need it. Lucas: That's the key insight. Most companies, when they see high support volume, they build more content — more FAQ pages, more walkthroughs, more onboarding emails. But the user doesn't want to read a manual. They want to accomplish their goal, and if the path isn't obvious, they bounce. CloudHive's solution works because it intercepts the user's intent before frustration sets in. Luna: Let's talk about the economics for a second. What did that save them? Lucas: So their support team was 12 people, handling about 140 tickets per person per day. With the reduction, they were able to drop to 4 support staff without increasing wait times. That's a headcount reduction of 8 people. At an all-in cost of roughly $60,000 per support rep per year in that region, that's $480,000 in direct salary savings. But if you factor in avoided hiring, training, and the fact that they didn't need to hire 3 more people they'd budgeted for — it's about $1.2 million annually in avoided costs. Luna: And that's before you consider the customer satisfaction upside. Fewer tickets means faster response times for the ones that do come in, which means happier customers. Lucas: And the fix took one developer maybe three days to implement. A dropdown, a simple logic table, and a data store to track which prompts have been shown. That's it. Luna: But I want to push back on something. This worked for a B2B SaaS company with a relatively small feature set. Would this work for, say, a bank, or a hospital? Those environments have way more complexity. Lucas: It's a fair question. I think the principle transfers, but the implementation gets harder. Let me give you an example. A bank's online portal might have hundreds of features. A single dropdown with five options won't cover everything. But you could do a tiered approach — first question narrows the domain, second question gets more specific. The key is the same: surface the path to the user's goal at the moment of intent. Luna: So instead of a dropdown, maybe it's a short wizard. But the spirit is the same. Lucas: Exactly. And I think one of the reasons this works so well in SaaS is that the user already has a goal in mind when they log in. They're not browsing randomly. They came to do something. So asking them what that something is — that's just respecting their time. Luna: It's almost like the classic 'walk up to a receptionist' problem. You walk into a building, you have a purpose, but you don't know which floor to go to. A good receptionist asks 'How can I help you?' and points you in the right direction. CloudHive just digitized that. Lucas: That's actually a really good analogy. And it's striking how few companies do this. Most products throw you into a dashboard and expect you to explore. But exploration is slow and frustrating when you have a deadline. Luna: I want to talk about one more thing. The VP at CloudHive — she didn't just guess the five options. She spent three months categorizing tickets. That's the unglamorous work that made the simple fix possible. Lucas: Huge point. The intervention looks simple, but it was built on a deep understanding of user behavior. She knew that 70 percent of tickets were about feature discovery, and she knew which features were most commonly missed. So the dropdown options weren't arbitrary. They were the top five uncompleted tasks that generated support tickets. Luna: That's the part most organizations skip. They want the quick fix without the diagnosis. Lucas: And that's why this episode fits our show. The operator's manual is about the small, tactical decisions that create outsized impact. But those decisions depend on data. If you don't know what's actually causing your support tickets, you'll build the wrong solution. Luna: Before we go deeper, I want to mention something. These conversations — they take time to research and produce. But they stay ad-free, and they stay useful, because listeners like you chip in a few bucks a month. If you've gotten something out of them, you can find us at buy me a coffee dot com slash fexingo. It genuinely makes a difference. Lucas: Yeah, and it keeps the lights on without any sponsor pressure. So thanks to anyone who's supported that. Luna: Alright. Back to CloudHive. One thing I'm curious about: did they see any negative effects? Like, did users find the question annoying? Lucas: They tracked that. The modal was dismissible — you could click 'Skip' and go straight to the dashboard. About 12 percent of users skipped it. Of the 88 percent who answered, the average time spent on the prompt was 4 seconds. So no, it wasn't perceived as a burden. And the support ticket drop was immediate. Luna: Four seconds for a 70 percent reduction in tickets. That might be the best ROI I've seen on this show. Lucas: It's up there. But the real lesson is about diagnosis. The VP could have said 'our support team needs to be bigger' or 'we need a better knowledge base.' Instead, she asked: what is the actual problem? And she took the time to answer that question before acting. Luna: So for a manager listening, the takeaway is: before you add headcount or software, look at your ticket data. Find the one category that dominates. Then ask yourself: what's the smallest intervention that could eliminate that category? Lucas: And don't assume the intervention has to be big. CloudHive's fix was a dropdown and a sentence. That's it. The complexity was in the analysis, not the solution. Luna: Alright. Next episode, I want to talk about a hospital that cut emergency room wait times with a single sign change. Lucas: I know that case. It's a good one. Let's do it.