Randall-Reilly is the recruiting powerhouse for fast growing companies. They utilize more than 30 web properties, an assortment of strategic media channels, and in-house high-tech call center campaigns to rapidly find, promote, and assist candidates through the recruiting and partnering processes of innovative companies. By engaging the services of Randall-Reilly, one of the largest ridesharing companies can scale their partnering process without losing focus on the important things - finding the right people.
Challenge: Discovering that Unique Gem
Finding the right people starts simply - often with a phone call. But when you handle an enormous amount of calls, you're faced with a massive challenge: sorting through millions of minutes of recorded audio in hopes of discovering a unique gem buried within complex and diverse conversations.
Randell-Reilly analyzes millions of minutes of phone interviews in an effort to identify specific call characteristics consistent with the ridesharing company's quest to partner with high quality safe drivers.
"They receive an enormous number of driver-partner applications per year. Analyzing their phone contact processes is key to finding amazing safe drivers. Working with Deepgram is the most effective way to do it." —Dennis Evanson, Head of Quality Assurance, Randall-Reilly
Better Speech Processing, Better Outcomes
"We tried Google's Cloud Speech API and Nuance Dragon, and investigated several other products from companies including Amazon, Tethr, Prosodica and Zoom. Deepgram had the best accuracy and program by far." —Brett Evanson, CTO, Randall-Reilly
After facing inaccurate results produced by leading providers, and struggling through cumbersome process designs, Randall-Reilly turned to Deepgram. They found that last generation solutions simply couldn't offer the services, accuracy, design, or speed needed to review every single call. Deepgram's neural networks, however, easily provided the scale, speed, and accuracy that they required to programmatically analyze phone interviews.
Before Deepgram, Randall-Reilly found that when it audited its speech-to-text query results from other vendors, only 80% of the search results produced confirmable matches. Using Deepgram they were able to increase audit-confirmed results to levels exceeding 95% accuracy.
Faster, Better, and More Reliable Compliance
"The reason we went with Deepgram is because it's a way that we can touch and have some level of QA assessment on every single call. It gives us the ability to scale compliance operations without hiring a hundred people. We're able to get more comfortable with compliance in a more flexible and efficient manner; rather than wading through speech-to-text and its inherent inaccuracies, Deepgram's approach is a faster and more accurate assessment."
While speaking with potential driver-partner candidates, Randell-Reilly's agents guide potential drivers through rigorous approval process. Deepgram's fast and accurate assessments of those conversations provides an optimized method for Randall-Reilly to gain a suitable comfort level concerning both the depth and effectiveness of its compliance efforts. Being able to quickly and reliably measure how well clients do at meeting both client and regulatory standards puts them miles ahead of other standard assessment methods. Deepgram is a crucial part of their process in streamlining both regulatory and client-mandated processes.
"As a company we are very concerned that we are doing and saying what our clients want, and doing it in the most efficient manner. Deepgram helps us track that effort efficiently and speedily. We love working with Deepgram. They're crucial in our automated analysis of every recorded interview, and in our capacity to make sure that [our client's] needs are being met." —Corbin McCabe, Randell-Reilly, General Manager
Find out how you can achieve what Randell-Reilly did.
If you have any feedback about this post, or anything else around Deepgram, we'd love to hear from you. Please let us know in our GitHub discussions.
More with these tags:
Share your feedback
Was this article useful or interesting to you?
We appreciate your response.