Years of work, spread across Excel files
For a restaurant operations specialist, Excel had become an essential part of his business. He used it to analyze sales, purchases, staffing, and financial performance, helping restaurant owners understand how their businesses were doing.
Over time, the spreadsheets became increasingly complicated. His work relied on three interconnected Excel files, with sales exports sometimes containing 10,000 rows per month. Some calculations and queries could take around 15 minutes.
He knew the numbers, understood how the calculations worked, and knew what restaurant owners needed to see. But working across multiple spreadsheets was becoming increasingly time-consuming, especially when dealing with several clients.
He wanted to bring everything into one application, where reports would be easier to manage and clients could eventually explore their own restaurant data.
The problem was that he didn't have a software development background.
Learning to build with AI
In August 2026, he started working with AAI Labs through Product Factory.
He had already spent years developing his approach to restaurant analysis, so he wanted to stay closely involved in building the application. With AI coding tools like Codex, he could work on the software himself, but he needed guidance on how to approach development and someone to turn to when things got complicated.
AAI Labs helped him plan the application, organize the codebase, and break larger features into manageable tasks. Through regular coaching sessions, they reviewed what he had built, worked through problems, and discussed what should come next.
One important part was making sure the numbers were correct. A dashboard might look impressive, but it wouldn't be much use to a restaurant owner if its financial calculations were wrong. The existing Excel reports provided a reference for checking the application's results.
AAI Labs also took care of the hosting and deployment setup, allowing him to test changes before making them available in the live application.
Seeing his work take shape
A few months later, the application had come a long way.
He could now import sales data, compare performance across different periods, and switch between multiple clients and restaurants. He also added AI-generated reports and began building purchasing analysis.
One feature allows him to compare ingredients purchased by different restaurants, including their quantities and prices. It's the kind of analysis that could help his clients identify opportunities to purchase supplies together or negotiate better deals.
By late October, the application was running on AAI Labs-managed servers, and he had successfully imported several years of real sales data.
There are still things to work through. Larger files have caused performance problems, and the team is reviewing user permissions and data security before opening the application to clients. As of November 2026, the plan was to begin testing with two clients covering three restaurants.
He continues building with Codex, deciding which features matter based on his experience working with restaurants. When he runs into technical problems or needs help making decisions about the application, AAI Labs is there to work through them with him.
What started as a collection of Excel workbooks is gradually becoming software he can use in his own business and eventually offer directly to his clients.
FREE PRODUCT READINESS CHECK
How ready is your own app?
Answer 12 plain-language questions and see which areas deserve attention first. About 3 minutes, no sign-up.
Take the readiness checkLet’s talk about your product.
Discuss where you are today and the engineering support you need next.
Book a Free 30-Minute Call