An employee once spent about 10 hours a week checking and updating online listings. Reviewing a new dashboard takes about two minutes thanks to AI.
Christine Kotler, Chief Marketing and Communications Officer at Baptist Health South Florida, reports those results from an automation project that grew out of her team’s overall AI transformation effort.
It all started because the healthcare organization needed accurate online listings so people could find its locations, physicians, and services. Maintaining those listings, however, required employees to look for problems and make updates across dozens of sites and platforms.
In Episode 242 of The Artificial Intelligence Show, part of the AI Transformations series presented by Google Cloud, Kotler walks through how one team changed that work. A collaboration between employees who understood the problem and an internal AI guide produced an automated dashboard, with plans to bring the approach to other teams.
Online listings are one of the ways a healthcare organization connects people with care. So they need ongoing attention to remain accurate.
At Baptist Health, that attention added up. Kotler describes a highly manual process, spread across different teams in the department, of checking listings, finding problems, and making corrections. For one employee, the work could take about 10 hours each week.
The process relied on spreadsheets. Its recurring nature meant that completing one week’s updates did not remove the obligation to do the work again.
There was a useful distinction inside that problem. Keeping information accurate was necessary. The amount of manual effort required to do it, however, was open to change.
The client-facing team brought that problem to an “AI Sherpa” within Baptist, one of the department’s more experienced AI users who helped colleagues apply the technology.
“All right, we need to solve this problem. Help me understand the best way to do this,” Kotler says, recalling the team’s request.
The request reflected how Baptist Health’s peer support had evolved. Sherpas initially tested tools and tutored colleagues. As employees gained experience, teams began asking for help with business challenges and workflow changes.
The broader model was collaborative: work through the problem, get advice, build, and return for more help. Teams remained involved in developing their own solutions.
That is a useful place to begin a workflow project. The people doing the work can identify what is repetitive, what information matters, and what a useful result should look like. Someone with more experience applying AI can help them find a way to change it.
In the listings project, Kotler says the team spent a couple of weeks connecting the data sources and making the process more automated.
That work produced a visual dashboard. Information that had required manual attention across different sites could now be reviewed through red, yellow, and green indicators.
About a week before the interview, the team unveiled the dashboard. It brought listings information into a visual display that a person could review quickly.
“What once took somebody about 10 hours every week now takes about two minutes,” she says.
The result Kotler highlights is a shorter review. The employee can look at the dashboard’s indicators instead of manually checking for problems across the different sites. Having the information brought together changes how much attention that recurring check requires.
Her next step was to extend the approach to the other teams managing listings. The first team had demonstrated a use for work that recurred elsewhere in the department.
Kotler’s interest in the result goes beyond a faster weekly process. She sees an opportunity to use the employee’s abilities differently.
“That was not a great use of a very talented individual,” she says of the spreadsheet-driven updates.
For leaders examining their own work, the case suggests a practical starting point: look for information that employees repeatedly check, reconcile, or update across multiple places. Make the current steps visible, then bring the people who perform them into the discussion about what could change.
A useful design question is what the employee should see when that information comes together, and what they need to do when it calls for attention. A clear review can remove much of the searching, while the response to a problem still needs an owner.
Then comes the management decision: what should the employee be able to do with that capacity? Kotler’s answer is to redirect talent toward work where it can have greater impact. Identifying that work belongs alongside the automation project, so the organization knows what the saved time is meant to make possible.
Did you enjoy this transformation story? Go deeper on how AI is reshaping work and business with The Artificial Intelligence Show. Each week, we break down what matters in AI and what leaders should do about it.