Press "1" if you're tired of being on hold! Nobody likes to wait, but since customer service can't be packaged to sit on a shelf ready when we need it, everybody has to "get in line" for help from banks, credit card companies, stores, healthcare or tech support. These days, that usually means dialing a call center somewhere in the world, which makes understanding caller patience increasingly important. A recent study, co-authored by the University of Chicago Booth School of Business professors Baris Ata and Che-Lin Su, and published in the journal Management Science, offers a more accurate approach to modeling caller patience than ever before, which could help call centers reduce customer waiting time on hold while helping businesses too.

"Knowing when a person decides to hang up or hang on is vital to streamlining call center operations, minimizing caller frustration and maximizing each customer service encounter," says Su, an expert in operations management.

More precisely predicting caller behavior can help call centers design better systems going forward as well as fine-tune those already in place; particularly important when companies plan changes in their business or major marketing promotions that produce a surge of calls.

"It's no use spending millions on advertising a new product, service or event if your call center can't cope with the customer response," says Ata, also a professor of operations management.

Improving call center speed and service makes everyone happier; customers are calmer, phone agents less agitated（焦虑的）, and business gets done more quickly and efficiently. But these new insights can also help companies negotiate smarter contracts for out-sourcing such services.

"Since the model produces more realistic results for how long a caller will stay on the line, it enables a more precise estimate for the number of callers who can be served per hour, day and month," explains Ata.

Based on data drawn directly from 1.3 million calls to an Israeli bank's customer service center, the authors present a dynamic model of the caller patience and decision process. They tested their new model against previous research assumptions that caller patience never changes -- even if the call center improves call priority and routing systems.

"The previous models used assumptions of caller abandonment that were easy to apply and analyze, but didn't provide a reasonable picture of people's patience," explains Su.

Four scenarios simulating changes in call priority handling were run with both models, proving that the older-style assumptions may be misleading, as they produce caller-abandonment predictions that can be off substantially -- at times significantly overestimating how long a caller is willing to hang on, or somewhat underestimating a caller's tendency to hang up.

"When a call center alters its discipline to improve speed and service, add agents, or change call routing and priority, we theorized those things should influence caller patience -- and our model shows that such improvements do indeed make a difference in whether people decide to hang up or hang on," concludes Su.