I just spent 34 minutes doing a robot's job. This is what happens when IT builds the app but outsources the experience.
Do tech leaders think "Customer Support" is someone else's problem?
I ask because I recently spent 34 minutes functioning as a "Human API."
I use a connected health device that started logging unusual results. The app knew the data was wrong; the error logs were generated locally. Yet, to resolve it, I was forced into a manual workflow that belongs in 1995: calling a number, waiting on hold, and verbally reading error codes to a human agent.
This is what I call the "Silo Tax." It is the time and money wasted because the App Development team doesn't speak to the Support Operations team.
This is an Architecture Failure, Not a Service Failure
A traditional business leader sees this 34-minute call and thinks, "We need to reduce hold times."
A modern IT leader sees this and asks, "Why did this call exist?"
In the era of Agentic AI, the app should have owned the resolution. The technology exists to detect the anomaly, package the logs, and securely transmit them to the support queue with a single user tap. Instead, the burden of data transfer was pushed to the customer.
The "Oscar Health" Contrast
It doesn't have to be this way. Look at Oscar Health. According to the recent State of Enterprise AI 2025 report, they integrated their AI stack directly with their medical records and claims systems.
Because their AI has context, it answers 58% of questions instantly. It doesn't ask the user to "read the log files"; it reads them itself. That is the difference between IT owning the outcome versus just building the interface.
The "Proactive" Opportunity
A savvy product leader would have built this workflow:
1. Detection: The app detects the anomaly pattern locally.
2. Interception: An AI agent prompts the user: "This looks unusual. Send diagnostics to support?"
3. Resolution: The logs arrive at the support center before the ticket is even opened.
The Verdict for CIOs
If you are a technology leader, you cannot view "Support" as a downstream function. Support is a feature of your product architecture.
Every time a user has to manually transfer data that your system already possesses, you have failed a basic architectural test. You have built a silo, and your customers are paying the tax.
Sources
- OpenAI. "The State of Enterprise AI 2025 Report." OpenAI, 2025.
Disclaimer: This blog post reflects my personal views only. AI tools may have been used for brevity, structure, or research support. Please independently verify any information before relying on it. This content does not represent the views of my employer, Infotech.com.
Comments