In such a fast-changing world, an Agile Workforce is a tremendous enterprise advantage. Achieving it means leveraging technology solutions to discover candidates on demand—whether inside or outside your organization—that can achieve the set goals for new gigs. Online labor markets have greatly reduced the friction involved in finding suitable candidates, but that only addresses part of the challenge. What’s left? Sifting through the qualified candidates to assemble an effective team.
To do that, you need the right mix of skills at the right cost, matching time zones, and compatible work styles, along with consideration for factors like motivation and commitment to the project. Assessing all this has always been a challenging, time-consuming and error-prone part of the hiring process. But it has become even tougher since the Agile Workforce model was introduced, and the frequency of forming and re-forming teams has increased dramatically.
Meet Crowd Advisor, @accenturelabs’ #AI-enabled tool to build the #AgileWorkforce. Our experts explain in a blog post: accenture.com/CrowdAdvisorBlog
Without better tools to help streamline and automate the selection of team members for gigs, organizations will find it very difficult to apply the gig economy paradigm at scale. This is why Accenture Labs has been developing tools like Crowd Advisor.
Imagine you’re managing a project and it’s time to hire someone with specific skills to perform a task for a fixed duration. You know what your requirements and constraints are, including timeline and level of effort. You probably begin by identifying sourcing channels and talent networks where you can post the job, and receive inquiries from several candidates. Then what?
Online talent platforms will deliver lists of candidates whose skills match your listed needs. But a great deal of analysis and filtering must still be done to avoid subtle mismatches, over-provisioning skills, and paying for additional experience that falls outside the scope of your requirements. What’s more, you’ll want to check for elements like candidate work style, work reputation, integrity, compatibility, commitment and communication skills.
Lacking a way to apply all of these criteria automatically, you, the hiring manager, must do so manually. We’ve found that this requires hiring managers to spend an average of 10 minutes or more per candidate. That adds up to significant point of friction when considering multiple candidates for each of several roles on what is likely a quick, short term project. And even then, the process may ignore key factors.
Crowd Advisor applies advanced analytics and machine learning to the process of sifting candidates discovered in online talent networks. It applies a holistic, data—driven approach to help hiring managers assemble a team that’s well-matched to the project at hand—not on skills alone, but on the many other qualities and considerations that create effective teams.
Shown in the figure above, the lifecycle begins with an employer posting a job to a sourcing platform, and ends by identifying the best matches to the employer's selection criteria.
Crowd Advisor can be plugged into any portal that facilitates crowd-worker hiring. State-of-the-art machine learning algorithms, trained on real-world data, recommend the best matching candidate for each job (see figure below). It’s the use of AI to assess overall compatibility between employer/gig/candidate that really sets Crowd Advisor apart from other talent search tools. It sifts through large volumes of data at high speed, to effectively and quickly extract the best candidates.
Rather than limiting itself to skills and requirements, Crowd Advisor takes a multidimensional approach to assessing candidates. Other elements (or dimensions), like personality and work reputation, are also taken into account, giving hiring managers a complete assessment rather than a simple profile. The hiring manager has discretion to adjust the weights that the algorithm gives to each element (see image below).
Consider a simple scenario where a hiring manager has a tight deadline and a flexible budget for a project. Crowd Advisor can provide default recommendations based on past data from successfully executed projects. But this particular project is business-critical and the hiring manager doesn’t want to take any chances, so they decide to adjust the weighting attached to certain metrics, like prior experience with the hiring manager (or similar task posters) and availability. Once these changes have been made, the candidates identified by Crowd Advisor will likely be different from the algorithm’s default recommendations. Real granularity of criteria allows the hiring manager to experiment with different "what if" scenarios when entering job requirement metrics. And the hybrid sourcing platform’s powerful algorithms take all the guesswork out of hiring the best possible candidate for the job.
So does Crowd Advisor’s vetting and hiring process actually work? The answer’s an emphatic "yes": in a recent evaluation based on 7,254 tasks and 96,271 freelancers, Crowd Advisor’s algorithms had a 98 percent success rate in quickly and effectively matching the right freelancer with the right task, compared to only 15 percent when using traditional hiring methods. What does that mean for you? Less time spent finding the right people, and more time spent on the projects at hand.
For more information or to schedule a demo of Crowd Advisor, contact Alpana Dubey [email@example.com] and Kumar Abhinav [firstname.lastname@example.org]. For more about the Accenture Labs Digital Experiences group and our other crowdsourcing work, visit our website or contact Alex Kass at email@example.com.