In Part 1, we focused on the underlying challenges many contact centers face today: managing volatility and introducing real flexibility into workforce models.
Volatile demand is unpredictable by nature, making it one of the most difficult problems to solve. Addressing it requires intentionally designing your workforce and the operational systems that support it.
Start with the workforce mix; not the tool
One of the most common mistakes organizations make when exploring flexible labor models is starting with tools or scheduling mechanics before answering a more fundamental question:
What role does flexibility play in our overall staffing strategy?
In our case, it would be a small, strategic layer designed specifically to absorb seasonal volatility.
At a high level, the workforce mix looks like this:
- ~30% scheduled full-time employees
- ~60% scheduled part-time employees
- ~10% unscheduled, employee-based, gig-style part-time employees
That last group — the unscheduled, gig-style employee role — is intentionally small, designed to absorb seasonal volatility and relieve pressure on scheduled staff without creating permanent overstaffing.
Importantly, these roles are not available to newly hired employees. Eligibility begins only after an employee has fully ramped and demonstrated proficiency.
Scheduled vs. unscheduled is not binary
“Unscheduled” doesn’t have to mean “never scheduled.”
For us, during peak seasons these employees do receive schedules. Outside of peak periods, they operate without fixed schedules and pick up hours to create their own schedules.
This approach adds structure when demand is predictable, preserves flexibility when it’s not and avoids locking in hours that won’t be needed later.
Meanwhile, scheduled part-time employees provide another layer of flexibility. They receive schedules one week at a time, several weeks in advance. They have significant control over their schedules by providing availability within defined parameters, which is then matched to forecasted demand.
The result is flexibility aligned to demand.
How it works
Unscheduled agents pick up hours in two primary ways:
- Demand-based availability
Available hours appear in the WFM tool when forecasted demand exceeds scheduled coverage.
- Shift trades
In a trade-heavy environment, this becomes a major source of hours.
Because shift trades are already a big part of how we operate, unscheduled agents fit into the existing model without disrupting it. They also help cover time-off requests through trades, benefiting both employees and the business.
There’s also a third way hours get covered: when volume spikes unexpectedly, we can quickly broadcast the need for help to agents not already working. We can often increase staffing levels by 40% or more within a short window. That kind of response to volatility simply isn’t possible in a traditional, office-based environment.
How to maintain proficiency without over-scheduling
A common concern with part-time employees — particularly those in unscheduled gig-style roles — is staying current, as policies and procedures evolve.
Agents are scheduled for “prep time” at the start of every shift — even if they only work a single shift that week. This time is used to:
- Review knowledge base updates
- Complete LMS courses
- Stay aligned on policy or procedural changes
We haven’t observed a correlation between quality and hours worked once agents reach full proficiency. While new part-time agents may take slightly longer to reach proficiency in calendar time, they tend to progress efficiently relative to the volume they handle.
The hardest part of flexibility
The most meaningful challenge with unscheduled roles isn’t quality or performance. It’s predictability.
Specifically:
- Scheduling one-on-one coaching sessions
- Losing some control over when agents choose to work
To address this, we’re introducing a simple anchor: each unscheduled agent commits to a monthly, scheduled 30-minute touchpoint. If they need to move it, that’s fine — but having one predictable connection point dramatically improves coaching consistency.
Beyond that, leads and managers handle time-sensitive issues in real time, connecting with agents whenever they work next.
Interestingly, as a percentage of total working time, unscheduled agents often receive more focused coaching than their scheduled counterparts — even if coordinating it requires more effort.
Why this model attracts a different workforce
One of the most underappreciated benefits of this approach is who it attracts and retains.
Many participants in these roles are:
- Parents returning to the workforce
- Semi-retired professionals
- Individuals who highly value flexibility
For these employees, flexibility isn’t a perk — it’s the primary reason they’re able to participate at all. Shorter shifts, control over when they work and the ability to scale hours up or down make the role sustainable in a way traditional schedules often are not.
Those dynamics have meaningful downstream effects on burnout, retention and overall workforce stability.
Analyzing burnout, retention and stability
Shorter shifts and greater control over when employees work reduce burnout in very practical ways. A four-hour shift is fundamentally different from an eight-hour one, particularly in high-volume or emotionally demanding environments.
For many employees, that difference makes the work more sustainable, and greater sustainability means lower attrition.
Across our broader operation, part-time attrition has historically been at least half that of full-time roles. While other factors play a role, flexibility is a meaningful contributor.
Less frequent hiring means reduced training spend and retaining more institutional knowledge within the team.
Taken together, these dynamics point to a broader lesson: flexibility, when thoughtfully designed, doesn’t just help absorb volatility. It can also improve the long-term health and resilience of the workforce itself.
The point of this series is not to turn everyone into a gig worker. It’s about acknowledging that different segments of demand require different solutions — and that solving the last 10–20% of volatility often demands a different approach than solving the first 80%.
In Part 3, we’ll step back and look at outcomes, tradeoffs and the leadership mindset shift required to make models like this work.