What Happens When You A/B Test Hybrid Work? The Results Are Fascinating
An interesting study published in Nature just made a powerful contribution to the work-from-home vs office discourse. Instead of relying on gut feelings or anecdotal evidence, researchers conducted a real-world A/B test with 1,600 employees at Trip.com, one of the world's largest online travel companies. The findings? They might surprise even the most ardent skeptics.
The experiment
Trip.com randomly assigned 1,612 China-based employees (395 managers and 1,217 non-managers) into two groups:
- Control group: Five days in the office
- Test group: Three days in the office (Monday, Tuesday, Thursday)
They tracked everything from productivity to quit rates for six months, then followed up with performance reviews for two years. Here's what they discovered.
The unexpected results
Before the experiment, managers predicted hybrid work would decrease productivity by 2.6%. Non-managers were actually slightly optimistic (and realistic), expecting a +0.7% boost.
The actual results painted a valuable picture:
- No difference in productivity, performance reviews or promotion rates between groups
- Job satisfaction increased significantly
- Quit rates dropped by 35% (even more dramatically for women at 54% and employees with long commutes at 52%)
After experiencing hybrid work for six months, managers dramatically shifted their view, now seeing a +1.0% productivity impact, while non-managers landed at +1.6%, essentially removing the perception gap between the groups.
The financial impact
The numbers get even more interesting when you look at the bottom line. According to the Society of Human Resource Management, each employee departure costs at least 50% of their annual salary. For Trip.com, that's $30,000 per person.
By reducing turnover through hybrid work, the company saved millions without sacrificing performance or innovation.
What made it work?
Trip.com's success wasn't accidental. They identified three critical elements:
- Performance management: Instead of monitoring desk time, they implemented rigorous six-month reviews based on measurable outcomes
- Coordinated schedules: Teams came to the office on the same days, preventing the frustration of empty offices and unnecessary video calls
- Leadership alignment: The entire executive team supported the initiative, creating consistency throughout the organization
These elements demonstrate that it's not just about location - it's about the way you work.
Providing the space for flexibility
One particularly fascinating finding: hybrid employees worked about 1.5 hours less on home days. However, they compensated by working longer hours on office days and occasionally on weekends.
The flexibility allowed them to integrate personal needs - doctor's appointments, school runs, or even leisure activities - while maintaining their overall productivity. It's why flexibility often results in improved mental health and an improved perception of work-life balance.
What this means for the future
As companies continue debating return-to-office policies (without any solid evidence to back up the return-to-office mandates), this research provides valuable, data-driven insights. It suggests that the question isn't whether hybrid work can be effective, but rather how to implement it successfully. How to adjust the way of working to unleash the potential workplace flexibility has to offer.
Looking ahead
What's particularly valuable about this study is its methodology. By using A/B testing - a practice common in product development but rare in management - Trip.com has shown us a more robust way to approach workplace innovation.
Trip.com is now conducting further experiments to fine-tune their approach, including testing different combinations of office days to find the optimal balance.
My recommendation
Set up an experiment where teams can decide themselves when to work from where. I'd put my savings on it being even more powerful.
