Oscar’s office is not what Sam expected.
She expected glass and chrome. A big screen with dashboards. Maybe a robot in the corner. Instead, it’s a modest space in London, Ontario, with a whiteboard that takes up an entire wall, a desk buried in coffee cups, and a framed screenshot of what appears to be a medieval video game character standing in a mine.
“Is that RuneScape?” Sam asks.
“Long story,” Oscar says. “Involves scorpions and a permanent ban. I’ll tell you sometime.”
He gestures to the whiteboard. “I want to try something. You told me on the phone that you tried three things: hiring, buying software, and building your own with Zapier and ChatGPT. All three failed. But they didn’t fail for the same reason. And I think you already know why, even if you haven’t put words to it yet.”
He writes three problems on the whiteboard:
Problem 1: When a shipping label is created, someone needs to email the tracking number to the customer.
Problem 2: Lisa needs to open QuickBooks, navigate to the same screen, copy data from one field to another, and do this across three apps every single time.
Problem 3: A PDF purchase order arrives and someone needs to read it, understand what it’s asking for, and extract the line items, even though every customer’s PO looks different.
“Three problems,” Oscar says. “If you had to solve each one, what would you reach for?”
Sam thinks. “I mean… Zapier for the first one? It’s just a trigger. When this happens, do that.”
“Good. What about the second?”
“Some kind of… I don’t know, a macro? Something that clicks through the screens the same way every time?”
“Close enough. And the third?”
Sam pauses. “That’s the one where you’d need AI. Because the document is different every time.”
Oscar nods slowly. “You just described the entire technology landscape in about thirty seconds. Most vendors will take forty minutes to sell you a platform that blurs all three of those together. But they’re fundamentally different problems.”
He labels them on the board:
Problem 1: Automation. Rules. If this, then that. Simple, cheap, usually the right starting point.
Problem 2: RPA. Scripts. A software robot that mimics what a human clicks, in the same order, every time. More complex, but still predictable.
Problem 3: AI. Judgment. Reading, interpreting, deciding, even when the input changes.
“Now here’s where most people go wrong,” Oscar says. “Which of these three did CloudSyncPro promise to solve?”
Sam thinks. “All of them? They said it would handle everything.”
“Right. And which of these three did Jordan’s Zapier experiment try to solve?”
“The third one. The AI one. Reading POs and extracting data.”
“Was Zapier the right tool for that?”
Sam is quiet. “No. Zapier is… Problem 1 stuff. Triggers and rules. Jordan bolted ChatGPT onto it to try to handle Problem 3, but the connection between them had no error handling, no monitoring, no…”
She trails off. She can see it now.
“Jordan used the right brain but the wrong body,” Oscar says. “ChatGPT can read a PO. But Zapier couldn’t hold the whole thing together reliably. And nobody built the part in between.”
Sam stares at the whiteboard. Then she says something that surprises both of them.
“It’s like the sugar shack. The tubing is automation. It just flows. The person connecting the buckets to the evaporator is RPA, doing the same physical steps every time. And the grader, the one who tastes the syrup and decides what grade it is, that’s AI.”
Oscar picks up the marker and draws a maple syrup bottle next to the three problems. “Keep going.”
“So CloudSyncPro tried to do all three at once, without understanding which parts we actually needed.”
“And Jordan’s experiment?”
“Was like asking the grader to also connect the tubing and run the evaporator and do quality control, all at once, with no one watching.”
Oscar caps the marker. “Most businesses I work with find that 70 to 80 percent of their pain points are Problem 1. Simple automations. Rules and triggers. AI is the right answer maybe 10 percent of the time. But when it is, it’s a game-changer. The trick is matching the tool to the problem, not buying the biggest tool and hoping it fits everything.”
Sam does the math. “So for most of my problems, we’re talking about the cheap end of the spectrum?”
“For most of your problems, yeah. Your monthly bank reconciliation? That’s mostly rule-following with some pattern matching, medium complexity. Your PO data extraction? That’s AI territory, because every customer sends a different format. But your tracking number emails? That’s a simple trigger: when a shipping label is created, email the tracking number to the customer. That’s a $200 automation.”
“I’ve been paying Lisa to do that manually for three years.”
“I know.”
Oscar pours two coffees from a machine in the corner and hands Sam one.
“Can I ask you something?” Sam says. “Jordan’s been using ChatGPT for months. I use it sometimes too. Doesn’t that count as having AI in the business?”
Oscar sits down. “It counts as having a smart coworker you can ask questions. It doesn’t count as having AI in your business.”
He explains: when you open ChatGPT and type a question, you’re using an AI assistant. You ask, it answers, you decide what to do with the answer. You’re still the one copying, pasting, making decisions, and taking action. You’re still the bottleneck.
An AI system is different. An AI system runs on its own. It reads incoming purchase orders, extracts the line items, checks your inventory, creates a sales order in QuickBooks, and sends a confirmation to the customer. All without you touching it. You see the exceptions. You make the judgment calls. The system handles everything else.
“Using ChatGPT at your desk,” Oscar says, “is like having a brilliant intern who you have to supervise every single minute. Building an AI system is like having an employee who handles an entire workflow and only comes to you when something is weird.”
“Both are valuable. They’re not the same thing.”
Sam thinks about all the times she’s asked ChatGPT to help her draft an email or summarize a report. It was useful. But nothing changed. She still did the same work the next day.
“So ChatGPT was never going to fix my PO problem,” she says.
“Not by itself. ChatGPT is the brain. What you need is the brain plus the hands plus the eyes plus the memory, all wired together. That’s a system.”
“Okay,” Sam says. “So I need to understand what AI actually can and can’t do. Because honestly, half the stuff I read online makes it sound like magic, and the other half makes it sound like it’s going to replace all my employees.”
Oscar laughs. “It’s neither. Here’s the short version: AI is great at reading messy documents, classifying things, and summarizing. It’s bad at being right 100% of the time, remembering anything between conversations, and keeping secrets.”
“Keeping secrets?”
“If your team is pasting customer data into ChatGPT, that data is going to OpenAI’s servers. We’ll talk about governance later, but this matters.”
Sam thinks about Jordan. She pushes the thought aside.
“The most important thing AI can’t do,” Oscar says, “is understand context it hasn’t been given. AI doesn’t know that Mrs. Chen prefers Amber even though she orders Dark. It doesn’t know that Henderson’s PO format changed last quarter. That context lives in Lisa’s head, and if you don’t teach it to the system, the system won’t know.”
“So it’s not a replacement for Lisa.”
“It’s not a replacement for anyone. It’s a replacement for the boring parts of everyone’s job. The reading, the copying, the sorting. Not the judgment. Not the relationships. Not the ‘I know this customer is about to have a problem because their order pattern just changed.’ That’s still human work.”
Sam is scribbling notes. This is the first time anyone has explained this to her without either overselling it or dismissing it.
Sam’s phone buzzes. She glances at it. It’s Lisa. She almost ignores it, but Lisa doesn’t call during business hours unless something is on fire.
“Sorry, one second.” She picks up.
Lisa’s voice is tight. “Henderson’s just called. They want a breakdown of their Q3 order history by grade, by distribution center, by month. Their procurement team needs it for a budget review.”
“Okay. How long will that take?”
“If I drop everything else? Two days. Maybe three. The data is split across QuickBooks, the Google Sheet, and that one spreadsheet from when we switched to the new carrier. I’ll have to cross-reference all three manually.”
Sam closes her eyes. “Do it. Henderson’s comes first.”
She hangs up and looks at Oscar. He hasn’t said anything. He’s just watching.
“That was Lisa. Henderson’s wants a report. It’s going to take her two days to pull together because the data is in three different places.”
Oscar nods. “How long should it take?”
Sam opens her mouth. Closes it. “Five minutes. If our systems talked to each other, it should take five minutes.”
Oscar writes nothing. He doesn’t need to. Sam just heard herself say it.
“So the version my team is using, just ChatGPT, on their own, pasting stuff in, that’s like having a powerful engine with no car around it.”
“That’s exactly what it is. The engine is impressive. But without the chassis, the steering, and the brakes, it’s just noise and heat.”
Oscar tells Sam one more thing before she leaves.
“Everything you tried taught you something. The hire, the software, the Zapier experiment — each one failed for a different reason, and together they tell you exactly where the real problem is.”
He pauses.
“And that is what we’re going to do next. We’re going to look at your business, really look at it, and figure out where you actually stand.”
Sam drives home with something she hasn’t felt in months.
Not confidence. Not excitement. Something more useful than both.
Clarity.
The next week, Sam drives to Quebec to visit one of MapleCo’s longest-standing producers, a family operation near Portneuf that’s been supplying them since 2003.
She makes this trip once or twice a year. Usually it’s about pricing, quality, production forecasts. This time she doesn’t have a specific agenda. She just wants to see the operation. She’s not sure why.
The sugar shack has changed since her last visit. New tubing. New monitoring equipment. The owner, Marc, walks her through the upgrades.
“See this?” He points to a small screen mounted near the mainline. “Every tap is on a sensor. If a line drops pressure, I get a notification. I don’t walk the bush every morning checking tubes anymore. The system tells me where the problems are, and I go fix those.”
“How many taps do you run?”
“Twelve thousand. Used to be four thousand. With the old system, I couldn’t manage more than four thousand because I was the monitoring system. I had to physically check everything. Now the system checks everything and I handle the exceptions.”
Sam watches the sap flow through clear tubing into the collection tank. She watches it get pumped to the evaporator. She watches the operator, Marc’s daughter, monitor a screen that shows temperature, density, and flow rate in real time. The daughter adjusts one dial, checks a reading, and goes back to her coffee.
“How many people run this now?” Sam asks.
“Three during sap season. Used to be seven. But the three aren’t doing less work. They’re doing different work. Better work. My daughter is a better grader than I ever was because she has time to actually grade instead of spending half her day walking tubing lines.”
Sam drives back to Ontario in silence. She doesn’t listen to the radio. She doesn’t make phone calls.
She keeps thinking about Marc’s sentence: I couldn’t manage more than four thousand because I was the monitoring system.