The interns discovered the cheat code earlier than the bosses realized that there was a whole game to be played.
Someone shared that there would be a bonus multiplier based on badge swipes. The intern's optimal move was clear within roughly four seconds: swipe, purchase coffee, swipe out. 💸
You didn't measure presence, you measured a door
Let's be honest about something no one in that room ever seems to acknowledge. Swiping a badge isn't "an engineer doing focused work close to their peers."
Yes, a badge swipe simply opens a door.
When you add a bonus to opening the door, you're making it the goal. Not collaborating. Not shipping. Opening the door.
Once a metric becomes a target, it ceases to be a good metric.
The receipts are ugly
This is not just a concept. The monitoring systems have been integrated into evaluating performance and pay.
In June 2023, an internal Google memo from Chief People Officer Fiona Cicconi declared that badge swipes would be used to ensure three days in the office, with bosses told to "co-locate high-performing, low-performing, and average-performing employees" and managers "encouraged to incorporate badging, tool and resource logins" into reviews. By the end of 2025 and early 2026, Amazon was deploying dashboards that automatically highlight those clocking up the fewest hours as "low-time badgers" or "zero badgers" off a rolling eight-week swipe average.
Next, the money came into play. A survey conducted by ResumeTemplates in October 2024 asked 713 business leaders:
→ 18% of companies use badge data to dock bonuses
→ 12% let it hit base salary
→ 21% treat it as grounds for termination
According to the same survey, 60% of companies monitor employee attendance, and the most common method used is the employee badge. Hence, for the most part, companies are essentially evaluating employees based on their ability to walk through a door.
And the door is lying
We will most likely see a predictable outcome. In Owl Labs' 2025 report, 43% of hybrid workers were actively "coffee badging," and another 12% were waiting their turn.
That's true. The intern experienced coffee badging. You arrive, show your face, then vanish.
The data has been manipulated to a point where it cannot be considered true anymore. Occuspace provided numbers from late 2025 indicating that badge data on building occupancy is exaggerated by 15% to 20% due to hit-and-run swipes.
I'll repeat that in other words. Leadership decided on salaries based on an inaccurate amount. About a fifth lower. Then, they turned attendance into a game and can't believe employees are exploiting it to get compensated.
The surveillance war nobody wins
So the metric got gamed. Naturally, the response isn't "maybe the metric was dumb." The response is more surveillance.
For example, you can use tools such as CurrentWare, which costs $12/user/month, advertises the capability to "detect coffee badging" by monitoring your IP and MAC addresses; the exact hours your machine is latched onto the office network.
Think about that for a moment. You're essentially hiring a vendor to care for the emotional well-being of your engineer's laptop and its relationship with the WiFi.
This is the arms race that the thesis anticipates. If you use money as a proxy, you will not achieve the desired result. You will experience fraudulent activities, followed by MAC-address forensics, and then more sophisticated fraud.
The part everyone skips
No one leaves a job they love without a backward glance. John Frehse, head of global labor strategy at Ankura, is very clear on this. 'Coffee badging is a clear indicator of dysfunction in a workplace.' ☕
The coffee badge is a symptom of the underlying problem: The office isn't a good enough place to be, so rather than work out how to make it better, people optimize for looking "most of the office" rather than actually being in the office.
Instead of investing in tracking down MAC addresses at $12 a head, why not examine why people don't show up and reward that.
The metric was always a proxy for something tangible: individuals working on projects collectively. If you reward the proxy, the tangible asset no longer matters, and everyone concentrates on optimizing the proxy.
Takeaway
Adding a monetary value to anything changes it from a measurement of reality to a measurement of motivation. The interns simply happened to read the spec more quickly than you could put it down.
If you cannot quantify "great work happening in the room," should you even be paying for the room, or for the work?
What's the most obviously-gamed metric you've watched a company defend with a straight face?
