Ted Taylor left a 25-year career in the semiconductor industry and now operates three co-branded Schlotzsky’s and Cinnabon quick-service restaurants across New Mexico as an independent franchisee. Managing quick-service restaurants requires extensive behind-the-scenes work — menu pricing, staffing, equipment maintenance and policies — and Taylor has turned to OpenAI tools, including ChatGPT Work, to streamline many of those tasks.
What changed and by how much
- Menu repricing time dropped from five days to one–two hours, a reduction of about 95%.
- Monthly cash reconciliation time fell from three hours to 5–10 minutes after uploading each store’s cash transaction data and bank records to ChatGPT Work.
- About four hours of manual work are saved each payroll cycle by using ChatGPT Work to summarize payroll inputs.
Taylor also estimates that better scheduling could lower hourly labor costs by one to two percentage points of sales across his three restaurants. That improvement could free roughly $50,000–$100,000 in annual cash flow that could be used for higher pay, bonuses, business investment, or increased profit distributions.
How ChatGPT Work is used for pricing
Annual menu repricing requires analysis of recipe costs, item popularity, local competitor pricing, portion sizes and the differing economics of in-store versus delivery orders. Taylor first uses ChatGPT Work to research local pricing trends, portion sizes and relative food quality to establish pricing targets and strategies. He then uploads a categorized list of menu items with current in-store and delivery prices and specifies the overall desired price increase and which items should be changed more or less.
If he wants a deeper analysis, Taylor adds item-level sales history and supplier costs. ChatGPT Work can use those inputs to estimate sales impacts of price changes and optimize item-level pricing around cost-of-goods targets. The output is a spreadsheet with recommended in-store and delivery prices and an explanation of how those changes align with an overall pricing strategy that considers customer retention, minimizing price shocks, and positioning add-ons.
The first time Taylor ran a pricing analysis, he also completed a manual review. “ChatGPT Work made a deeper local comparison and helped me avoid increases that would have made some items uncompetitive,” he said. After that experience he began using ChatGPT Work’s recommendations as his starting point.
Labor management and scheduling
Taylor applies ChatGPT Work to labor management in two ways: reviewing how labor is tracking across his three restaurants and testing an improved method for building the weekly schedule.
For labor analysis he provided read-only access to sales and labor data. The analysis compares scheduled, budgeted and actual hours with forecast and actual sales, and it reviews the previous 12 weeks to account for stores that consistently over- or under-perform. The resulting report flags overtime, shifts running longer than planned, and coverage gaps, and highlights items a manager should review.
For scheduling, Taylor uploaded an employee roster with roles, training, availability and pay, along with staffing needs for different sales levels. ChatGPT Work proposed a full schedule for the location; the manager reviews it, makes adjustments as needed, and enters it into the restaurant’s scheduling system. In initial tests they identified missing rules (for example, when to add staff during busy periods) and refined the instructions for ChatGPT Work accordingly.
Day-to-day operations and other uses
Beyond pricing and scheduling, Taylor uses ChatGPT Work for routine operational tasks: drafting HR policies, reconciling cash, and managing supply and repair requests. When New Mexico required restaurants to keep a written procedure for bodily-fluid incidents, Taylor used ChatGPT to generate a draft in about 30 seconds. He also used the system to draft a social media policy for an employee handbook; an external HR firm reviewed it and suggested no changes.
Monthly cash reconciliation now takes 5–10 minutes after ChatGPT Work compares store cash transaction files with bank deposits. Payroll cycles save roughly four hours of manual work because ChatGPT Work summarizes payroll inputs.
Taylor also built a web portal — powered by outputs from ChatGPT Work — where employees can request supplies or repairs. The portal sends notifications via a dedicated email address and tracks the status of each request.
Process and safeguards
Taylor’s approach is pragmatic: he gives ChatGPT the same data and guidelines he would use to make a decision, then compares the AI’s output with how he would do the work. As his confidence grows, he expands the tool’s use to more tasks, but he always reviews results before implementing them.
Why this matters for small businesses
The example shows how a small quick-service operator can apply a large-language-model tool to speed up routine, data-heavy tasks. Beyond reclaimed hours, the tool enables analyses that might otherwise be skipped for lack of time, allowing owners and managers to focus more on customers, staff and business growth.
