Each day, Mark Koh and his team at Kuala Lumpur-based data training & content moderation company Supahands help their clients test the limits of the question ‘how much data is too much data?’ Data drives innovation, but it is only valuable if it is assigned a purpose, and it is only usable if it is organized to serve that purpose. As companies continue to move toward digitization and automation, and more data is at their disposal than ever before, the amount of data requiring cleaning, tagging, and categorizing has exploded. Amazon and Google are two examples of companies that have set themselves apart through data-driven innovation, but many organizations don’t have the scale, time or expertise to understand and make use of their data with the speed and accuracy needed to take full advantage of it.
That’s where Supahands comes in. Headquartered amidst the hustle-and-bustle and congestion of Malaysia’s capital city, the Supahands team aims to create a less polluted data universe. The almost 4-year-old startup integrates human and machine intelligence to create the world’s most efficient workforce and to empower its client organizations to scale to new heights. I connected with Koh–Supahands’ Co-founder and CEO–while living in Malaysia for a month and studying its startup ecosystem as part of my year-long journey investigating global startup communities. The focus of our conversation: Supahands’ unique human-machine hybrid approach and using Malaysia as a home base for its growing platform.
A Hybrid Approach
Supahands operates with the core understanding that machines and humans need each other. Humans manipulate data to educate and train machines to maximize what the machines can do, which enables machines to apply this data and complete tasks more efficiently and effectively than humans in many cases. By optimizing the way machines and humans interact and learn from each other within its own walls, Supahands is able to take on the tedious data training and content moderation tasks that its clients crucially need. The company’s initial vision was to function as a virtual assistant service provider, but it started pivoting in 2016 to a B2B model centered on a workforce-for-hire.
This approach has paid off, as business is booming with clients across Southeast Asia, Australia, Switzerland and the United States and a CAGR (compound annual growth rate) of 400% over the past 3 years. Examples of clients include Uber and now Grab (which acquired Uber’s operations in Southeast Asia), as well as automotive website network iCarAsia and stock content agency 123rf. Without high-quality data inputs, these companies cannot properly leverage their own technologies. In the words of prominent data quality expert Dr. Thomas Redman, “if your data is bad, your machine learning tools are useless.” To better serve its existing client base and expand its global presence, the company is in the process of closing its Series A round. This follows a seed round including funding from Asian telecom giant Axiata, leading early-stage venture firm 500 Startups, and Malaysia’s Cradle Seed Ventures.
What Makes Supahands Supa
What sets Supahands apart from its competitors in the busy outsourcing and moderation space is both how it completes projects, and the quality of its output. The company is positioned between crowdsourcing and freelancer marketplaces like Amazon Mechanical Turk and Upwork, and traditional BPO (business process outsourcing) companies and contact centers. It provides a higher level of curation and quality-control than standard marketplaces, without sacrificing the speed, agility, and automation often lacking in old-school BPO companies. How exactly does Supahands accomplish this? By practicing what it preaches–efficiency and data-driven decision making. Supahands has designed exceptionally efficient workflows and operations mixing crowdsourced human talent and highly-intelligent technology. This enables a small full-time team of 35 people working from a single base in Kuala Lumpur to oversee the operations of a curated crowd of 2,000 workers throughout Southeast Asia, known as SupaAgents. Managing SupaAgent output is no small feat, as the crowd of SupaAgents currently works on over 1 million units of data each month!
Members of the Supahands teamSUPAHANDS
Supahands develops all of the technology for its “Workplace” project management platform in-house, with a deep understanding of what’s required to optimize the speed and quality of project work. The centerpiece of this platform is DIANE (Digital Innovation Assistant for kNowledge Engineering), a predictive routing system that matches agents and projects based on timing and relevance. Supahands’ project managers look to DIANE to help them quickly and effectively launch projects. On the supply side (the workers), the company collects and aggregates data about each SupaAgent including skills, availability, and past performance that then flows through DIANE. Each agent-project pairing is governed by the intelligence of the system to control worker quality. Supahands prides itself on understanding and making use of data to recruit high-quality workers, and to provide these workers with the tools they need to succeed, even maintaining a worker happiness tracker. On the demand side (the work), projects can be broken down into micro-tasks or aggregated into larger projects with task forces of multiple workers to ensure SupaAgents work at high levels of accuracy and that the right sets of skills are assigned to each project.
SupaAgents are recommended and paired with projects based on Supahands’ predictive routing engine, DIANESUPAHANDS
Project managers also leverage Workplace’s Enterprise Resource Planning tool to manually manage SupaAgent supply with client demand and make sure that the workforce is “optimised according to the volume of projects Supahands has.” This tooling has compressed the amount the time needed to launch projects and manage workers from hours to minutes, meaning that Supahands’ team can focus its time and efforts on project completion. And the engineering team at Supahands is far from done identifying additional efficiencies; it is constantly identifying ways to make DIANE and the rest of its technology stack more intelligent, whether that means automating more processes or making better use of internal data to feed its systems.
Made In Malaysia
Supahands’ ability to achieve a balance of speed and quality, and to continue releasing impactful proprietary technology is largely a function of its Malaysian roots. Supahands has taken full advantage of Malaysia’s assets, such as a large pool of technical talent and an affordable labor force. It has tapped into the pool of tech talent in standing up and rapidly scaling a formidable engineering team to build its suite of in-house tech solutions like DIANE, and has tapped into Malaysia’s (and more broadly Southeast Asia’s) affordable workforce to build its team of SupaAgents. The company has another valuable resource at its disposal: Malaysia’s technology startup ecosystem. For all of its urban sprawl and chaos, Kuala Lumpur, and Malaysia by extension, contains a surprisingly small and tight-knit community of passionate entrepreneurs and investors. Given its size and scale, there is a widespread feeling of hometown pride and a collective commitment to the success of the ecosystem’s startups. Community members are willing to help one another in any way they can–a theme I noticed across the Southeast Asian ecosystems I explored. In fact, I was initially introduced to Mark Koh by the Co-Founder of Malaysia’s first co-working space, Paper + Toast, as Koh had participated in numerous events at the space. In addition to the assets Malaysia offers within its own ecosystem, it also acts as a gateway to many of Southeast Asia’s other growing markets–a position that has benefitted Supahands as it expands its workforce and client network. Malaysia neighbors Singapore and its wealth of capital, and is centrally located between Vietnam, Thailand, and Indonesia.
What’s Next?
The pressing question for Supahands is not only what’s next (continuing to increase the level of automation within its platform and building out partnerships on both the supply and demand sides to accelerate growth), but also where’s next. For all of Malaysia’s value as an early stage startup market, many startups are compelled to leave and set up operations elsewhere in order to scale successfully (the most notable example being the aforementioned ride-hailing startup Grab, which moved its headquarters from Malaysia to Singapore in 2014). Can Malaysia continue to support Supahands’ quest to expand its global reach? Moreover, will the company be able to maintain its efficient model as it continues to grow? A positive indication that the company is up to the challenge is the leanness of its operations. Supahands does not need to invest in setting up significant operations in different markets around the world, as its SupaAgent network is remote and the services it offers are all online.
While Supahands’ global scalability to help businesses continuously collect and use more data remains to be seen, much can be learned from its approach and the role it is playing in driving more dynamic and efficient ways of working. It has set an example in unifying human and machine intelligence that will be followed by the business outsourcing industry, and that represents the not too distant future of workforce optimization across all industries. It has embraced what many others are beginning to recognize: the cyborg workforce revolution is upon us.
