When a global technology company evaluated its workplace food program, it discovered a costly pattern: its catering vendor, Andros, was preparing meals for the full office headcount rather than the number of people actually eating each day. With roughly 700 employees assigned to the building but only about 525 showing up for lunch, the company was paying for hundreds of unnecessary meals every week.
The mismatches didn’t stop at attendance. Nearly 70 percent of employees preferred protein-forward meals, yet Andros continued producing large volumes of starches and salads that were consistently under-consumed. High-demand items ran out early while low-demand items went untouched, driving both waste and frustration.
Sifted was brought in to rebuild the program from the ground up using precision portioning, its data-driven model that aligns production with real consumption patterns. By studying historical attendance, tracking individual meal elements, and adjusting portion ratios for proteins, starches, and greens, Sifted replaced guesswork with accurate forecasting.
The impact was immediate. Daily food costs fell by nearly 30 percent, and the company saved more than $300,000 in its first year—even though Sifted’s per-person pricing was higher than Andros’s. Waste dropped sharply, employees consistently received the meal components they preferred, and the workplace team gained clear visibility into participation and cost performance.
The result is a catering program that is more efficient, more sustainable, and far more attuned to the people it serves—proof that smarter portioning outperforms outdated headcount-based models every time.