A vendor upgrade, a new vendor system, or a simple UI tweak can break a legacy RPA bot and create a week-long ticket backlog. The same happens with spreadsheets and macros that rely on fixed cell references. Automation teams spend more time rebuilding bots than delivering value. The real issue is not technology. It is the way we interact with interfaces that are never stable.
Why RPA breaks here
Traditional RPA (UiPath, Automation Anywhere, Blue Prism, Power Automate) automates by binding to selectors, xpaths, and object IDs. When the app or UI changes, the bot breaks and a developer has to rebuild it. That is the maintenance treadmill. Industry research shows that 30 to 50 percent of RPA project time goes into maintenance and troubleshooting, not new automation. A single selector change can halt dozens of bots across the enterprise. The cost is not just time. It is risk. When a critical finance or compliance process halts, teams must fall back to manual handling, which introduces audit gaps and human error.
What changes with computer use agents
- Survives UI changes: agents see the current screen layout and act accordingly, instead of failing on a missing selector.
- No brittle selectors: no xpaths or object IDs to maintain. The agent reads the interface and responds to what it sees.
- Recovers from exceptions: when a process deviates, the agent observes the state and adjusts rather than halting.
- Follows the SOP as written: a standard operating procedure in plain English is already almost a prompt for a computer use agent.
- Works on legacy and Citrix: agents run on real desktops and terminals, including virtualized environments where RPA struggles.
The one line a VP of automation should remember: selectors break, vision adapts.
How to move without the risk
You do not need to rip and replace RPA overnight. Start with a single high-pain process that is exception-heavy, UI-unstable, or documented in SOP form. Identify a process that today relies on a fragile macro or an RPA bot that breaks with every update. Map the current steps and capture the manual workarounds. Then, run a pilot with a computer use agent on the same process. Compare uptime, changeover time, and support tickets before and after the pilot. Use the results to decide where agents add the most value. The goal is not to replace all RPA at once. It is to steadily move from brittle, fixed bots to durable, vision-based agents. RPA still fits high-volume, stable, deterministic backend tasks. Agents are the durable answer for changing UIs, exception-heavy workflows, and SOP-driven operations.
A simple maturity curve
The curve is not a one-time project. It is a shift in how you build and maintain automation. Start with macros and scripts. Add RPA bots for stable, rule-based processes. Introduce computer use agents for dynamic, SOP-driven workflows. Each step reduces the time spent on maintenance and increases the resilience of your automation estate. Over time, agents become your primary engine for the long tail of automation.
The enterprise automation maturity curve is clear: selectors break, vision adapts. Start with one high-pain process, pilot a computer use agent, and measure the difference. If you are ready to see how agents can reduce maintenance time and increase resilience, book a demo with the Coasty team at https://cal.com/coasty/15min .
Want to see this in action?
View Case Studies