In the House: How Southwest Saved Time and Money with an Inhouse MSP

• Ten years ago, Southwest Airlines spent approximately $85 million per year on their contingent workforce program.
• Recognizing this program would continue to grow in size and strategic importance for the company, they decided to bring it inhouse.
• Since then, they’ve seen a $20-million reduction in cost per year, including $5-6 million per year through self-sourcing.
• Visit Staffing.com's homepage for more.

Change is hard.

I've been in the people department at Southwest Airlines for 14 years, most recently involved in establishing our contingent workforce program and bringing our managed service provider (MSP) inhouse. (An MSP delivers services on their customers’ premises, in this case sourcing and placing contingent workers. MSPs also offer support and manage all related IT and administrative operations for their customers.) Southwest currently has 17,000 people in our vendor management system (VMS), and we spend about $65 million each year on our contingent workforce (primarily in staff augmentation).

Bringing the program inhouse was a decision we made carefully, largely based on the rising importance of contingent workforces. Ardent research reports 41% of the average enterprise’s workforce is non-employee labor, a number which has grown more than 2.5 times in 10 years. According to a study by BCG and Harvard Business School’s Managing the Future of Work Initiative, roughly 40% of respondents said they expected freelance workers to account for an increased share of their organization’s workforce over the coming five years.

Not all companies have the right systems in place to manage this growing workforce, however. Deloitte’s 2019 Human Capital Trends report states only 8% of companies have processes to “manage and develop alternative workforces, and fully 54% of respondents said they either managed alternative workers inconsistently or had few or no processes for managing them at all” (see figure below).

We recognized our own contingent workforce program would continue to grow in size and strategic importance for the company, and we wanted to ensure we had control. We were able to see the potential for cost savings, reduced time to fill requests, and perhaps most importantly, ensure our contingent workforce was as much a part of the Southwest culture as our 67,000 full-time employees. The culture at Southwest is one of our primary value propositions not only for our customers but for our employees. Our culture began with the founder of Southwest, Herb Kelleher, who famously focused on people. “You have to treat your employees like customers. When you treat them right, then they will treat your outside customers right. That has been a powerful competitive weapon for us,” he told Fortune in 2001.

Almost 50 years after Kelleher started the company, people continue to be our focus. Because my team lives and breathes our culture every day, we understand what it looks like to hire the right people. Our contingent workforce has become an important pipeline for full-time employment, especially in our customer service department, where culture is especially important. Our contingent workforce is not remote. When we bring these workers into Southwest, they get a badge, come into our building in Dallas, and join a team, whether they’re here for six months or three years. It's critical, therefore, that we get the correct fit from the beginning.

In any change management process, adoption will be a huge issue. Here’s how we did it right.

Demonstrate Wins

From the onset, we realized we would have to present our positive results often. A large part of gaining the trust of the hiring managers and the leadership team was proving that the inhouse program fills positions more quickly, for less money, and with better cultural fit. We also calculated costs related to saving the hiring managers’ time. We looked at their typical salary and how much time they spent reviewing resumes, phone screening, and interviewing candidates. (Hiring managers had resumes sitting in their inboxes for an average of 40 days.) We were able to remove those tasks from their plate, allowing them to focus on their primary role, and we quantified the results.

In addition to culling the suppliers from 43 to 13, we expanded Beeline as our vendor management system and TalentNet as our provider for our private talent pools. Just under half of our placements each year are through our self-sourcing initiatives, which saves us $5-6 million per year.

Other wins came in the form of clear guidelines and processes. For example, one of our first steps was to create procedures for granting and limiting access to our facilities and networks. As our contingent workforce had grown organically, no one had established a central authorization point for determining access. When we expanded Beeline and our VMS, we closed down all other paths for access; all contingent workers now had to go through the program. Whether it's getting access to the network or getting a hard badge by facilities, none of those things can happen unless it comes from my team.

Expect Pushback

One of the biggest hurdles we faced was asking our hiring managers to trust that our team could source and place candidates better than their favorite suppliers. Many of the managers had worked with their suppliers for years, and they did not want to give up those relationships.

In order to be a vendor-neutral program, however, we had to evaluate our suppliers fairly, hold them accountable, and ultimately reduce the number. The hiring managers challenged us. They continued to reach out to their preferred supplier, slowing down the adoption of our inhouse program. The suppliers, too, challenged us. When they submitted candidates, they wanted to work directly with the hiring manager. They’d ask, "What qualifies your team to either move my person forward or reject them?" Ultimately, we had to prove ourselves with both the hiring managers and the suppliers so that we could say, "No, I know what my customer is looking for, and this is it.”

Give It Time (But Not Too Much)

Because we were moving from an existing MSP to an internal one, we scheduled six months for the transition. In retrospect, this was too much time. My team had industry experience and understood their new roles. The six-month time frame prevented hiring managers from adopting the new processes more quickly and allowed more time to continue relationships with their previous suppliers. (For a company creating an inhouse MSP from scratch, I would recommend a six-month time frame or longer for the whole process: planning through implementation.)

We rolled out the new MSP with a phased approach, department by department. Our communication method was heavy with face-to-face meetings. We hosted town halls and lunch-and-learns. We sat in small group department meetings and discussed what we were doing and what they could expect. While challenges and pushback are to be expected, I believe the amount of time we invested in these in-person meetings was critical to the success of our implementation.

It’s been five years since we launched the inhouse MSP. My team and I continue to learn, and the program continues to mature. For example, when we first introduced it, we wanted it to be a white-glove service for our hiring managers. We didn’t want them to have to open requisitions in Beeline or manage any of the administrative tasks. Rather, we wanted them to call their Southwest recruiter. Over time, we’ve had to reset those expectations in favor of more self-service. We do total talent now, and I've got a much smaller team that manages the program.

Our program will continue to evolve, along with the contingent workforce industry and talent management, more broadly. The changes associated with such a dynamic talent economy will continue to be challenging, but necessary. I’m excited to be a part of it and looking forward to the future.