Presented by ACV Auctions
ACV Auctions became Buffalo, NY’s first unicorn when it raised $150 million Series E round with a $1.5 billion post-money valuation in 2019. The company’s AI-powered, dealer-to-dealer automotive auction platform, which enables dealers, historically tied to local markets, to buy and sell used vehicles from anywhere, is disrupting the multi-billion-dollar used car industry and putting Buffalo on the map.
The company’s rising trajectory, and its place in Buffalo’s startup ecosystem, are driven by its tech- and data-first best practices. Everything from hiring practices to domain-driven design, automation, digitizing back-end ops, and predictable and fast software delivery combine holistically to create a team-centric culture with world-class product development — and have driven record growth.
“The vision is pretty simple — we’re aiming to be the top technology company in this part of the country,” says Vikas Mehta, ACV’s chief operating officer. “Great products require great technology and great teams. Over the past couple of years, we’ve made significant investments in hiring great talent and building a culture that promotes innovation, autonomy, and speed-to-decision-making, that prioritizes innovation and testing.”
Creating phenomenal teams
To attract and develop top talent and build industry-best teams, the company proactively folds in the best practices that leadership has brought with them from the many established tech companies they hail from, including Apple, Google, PayPal, eBay, Facebook, Lending Tree, and more. The common denominator is creating an atmosphere for engagement and retention, with an emphasis on career pathing, learning, and development, says Mehta.
And they are not only recruiting established talent from the region, but also building the pipeline of upcoming tech stars. They’re closely affiliated with the University at Buffalo’s School of Engineering and Applied Sciences, and host small-class teams and interns to problem-solve real-world projects in the business, and many graduates continue on as full-time team members.
They hire for diversity, intentionally seeking creative collisions of unique perspectives that help challenge standing assumptions and address operational pain points with technology solutions. Their candidates come from a broad array of backgrounds, from brilliant new perspectives courtesy of graduates with degrees in electrical, automotive industrial and manufacturing engineering, physics, computational neural science, aerospace engineering, and computer science — while also scooping talent from companies like Tesla, Apple, Qualcomm, Carvana, and KMX Industries. The team’s backgrounds are in everything from defense to Tier 1 automotive suppliers, IT, insurance, neural science, and technology.
And as part of the TechBuffalo initiative, ACV and other big technology companies in the region are sponsoring the development of a pipeline of technologists, providing funding and training to individuals who don’t come from a technology background in order to develop valuable engineering and product skills. And internally, they’re upskilling and reskilling, developing talent within the company to tap into the deep wells of knowledge and expertise as well as raw skill.
“We’re on a journey to be best-in-class in terms of culture, and determining the best growth plans and career paths for our people,” Mehta says.
Creating phenomenal tech
“We’re pushing our code more than 20 times a day, our change failure rate is under 5%, and our mean time to fix is in the minutes, and that comes down to domain-driven design,” Mehta says. “You’re creating these independent buckets of tech that are concurrently being built and released, and then always thinking through best practices and tech support systems that allow us to move fast.”
The company is organized into nine domain teams, near autonomous vertical units that work in close partnership with business teams. They were able to assign team ownership and autonomy for each domain, treat each of these areas as services, and then assess the build-versus-buy question by feature set or capability on an individual basis.
For instance, the inspection domain maps to the ways a car could be represented on ACV’s website or platform, such as a feed coming from a third party, or through the company’s own inspection app. That inspection domain has the responsibility of taking inputs and then structuring them into a vehicle object output that in turn gets ingested by the marketplace team. The marketplace team works autonomously with those outputs — iterating, testing, and building marketplace experiences around how these vehicle objects can show up to ACV’s customers, without having dependencies on the inspection domain.
This structure allows them to multithread, moving faster with more independent decision-making and peer leadership within teams, while enabling world-class productivity metrics. Teams deliver to production over 500 times a month, which allows team members to shine regularly at the helm of every new project. This kind of organizational structure has a tremendous, often understated impact on two of the most important drivers of career advancement: opportunities for upward mobility and growth development.
“We feel like we’ve unlocked a sense of ownership and autonomy, speed, and the ability to become a true platform, where specific capabilities could be offered internally or externally, independent of others,” Mehta says. “In many ways this was a technology decision, but it’s led to a culture and engagement opportunity.”
Automation and AI as a superpower
The dealer-to-dealer marketplace fulfillment service is a complex undertaking. Payments, title transfers, arbitration, and more were all initially handled manually, but with every new territory or state added, new complexities were added, because rules and regulations vary widely across borders, from liens to odometer readings.
AI, computer vision, automation, and robotics were the game changers, scaling back-office operations, optimizing every step, and dramatically reducing re-work and errors, from scanning titles to auto-dispatching and auto-assigning loads for vehicle transport – all to create a seamless transportation, payment, and title transfer experience. The technologies aren’t simply buzzwords, but tools that every domain relies on under one data umbrella to deliver a much more efficient and engaging marketplace.
“Data can provide the ability to give an instant release slip to a new customer, which in our industry is almost unheard-of,” Mehta says. “Now, because we have all of these signals in the marketplace, we can deliver a phenomenal experience for the customer by leveraging the data we have.”
Their massive data repository and intelligence allows them to price cars on their platform, and otherwise predict conditions of cars that may or may not be in their ecosystem, based on the information they have already amassed. This helps sellers determine the best liquidation channel, the best price, and warns them about likely condition challenges too.
“Over time we’ve built not just very efficient, scalable tech, but we’ve built a platform that collects a lot of data, and over time can be used to make our process more intelligent and more streamlined,” Mehta adds. “This is how large data companies move fast at scale: through leveraging investment in technology and people.”
And R&D is always in the process of testing and iterating new ways to address customer pain points and leverage the customer data and feedback it receives in the field, especially around discovery, personalization, recommendation, and intelligent notifications.
“It’s about creating a trusted, engaging, and efficient marketplace. It’s not just a product. It’s a commitment made across multiple aspects,” says Mehta.
Sponsored articles are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. Content produced by our editorial team is never influenced by advertisers or sponsors in any way. For more information, contact [email protected].
