BPO’s next chapter: Survive the AI wave or sink

In late 2023, SkyTech Solutions, one of Bangladesh's top outsourcing firms, faced a sudden reckoning. The core tasks of a major project, which generated a steady source of revenue, were handed over to artificial intelligence.

"Artificial intelligence wiped out over 80 percent of the jobs in that project in a single blow," said Musnad E Ahmed, founder of SkyTech Solutions. "They [client] informed us that the core task would now be handled by AI automation, leaving only the quality control work for humans."

The task involved processing freight invoices for Uber Freight and uploading the information into a customer relationship management (CRM) system -- a repetitive but lucrative assignment that had employed more than 50 people and paid $8.50 an hour per worker.

SkyTech's experience is emblematic of a growing disruption in Bangladesh's business process outsourcing (BPO) sector, where AI is rapidly displacing entry-level workers in customer service, transcription, data entry, and other routine jobs.

THE AI OVERTAKE

While not on par with global hubs like India and the Philippines, the BPO sector has been flourishing in Bangladesh with a total market value of $850 million, comprising around 400 firms, according to the Bangladesh Association of Contact Center and Outsourcing (BACCO). It employs more than 80,000 people, around 40 percent of whom are women.

Once hailed as a key driver of youth employment and a reliable source of foreign exchange, the sector now finds itself in the throes of transformation. Companies are increasingly adopting automation to stay competitive in an evolving global market. But in doing so, they are shrinking their workforces.

Yasin Arafat, who worked in customer chat support at a Dhaka-based BPO firm, lost his job when a client introduced an AI-powered chatbot.

"As the task I worked on was replaced by AI, I was left jobless for months," said Arafat, who eventually rejoined under a different project.

For ASL BPO, another leading player, automation has already reshaped the business. The firm reduced its team by 20-30 percent on a recent e-commerce project, citing the efficiency gains of AI-driven quality assurance.

"Team sizes, especially in customer support and graphic design, have decreased due to automation," said Zayed Uddin Ahmed, CEO of ASL BPO. "AI increases efficiency, especially with generative tools. Our digital marketing team now uses ChatGPT for research, saving hours."

The company had over 250 employees in mid-2023. Now it has 150.

"We've brought automation into our workflow, but the overall effect has been a reduction in staff," Zayed said. "We must embrace AI solutions to survive in this new era. It's not optional, it's essential."

At The KOW Company, a post-production studio specialising in image and 3D processing, AI has not just streamlined operations, but it has even changed hiring patterns. "Earlier, we used to post job openings for 25 people at a time. Now we hire around 10," said CEO Kowser Ahmed.

In one year, the company's workforce dropped from 650 to 550. Yet its daily image processing capacity increased from 25,000 to 40,000.

"As AI continues to evolve, especially with agentic AI, human involvement in repetitive work will decline sharply," he said.

NEW OPPORTUNITIES

Despite job cuts, many in the industry see AI as an opportunity for reinvention. Some firms are using it to create value-added services and explore new business models.

"We've been working with AI for 7-8 years. Earlier, image editing was 100 percent manual. Now, around 30 percent is AI-assisted, which speeds up delivery and creates more opportunities," said Md Riajul Islam, CEO of Bright River Bangladesh.

The company, which serves clients in 16 countries, has built its own AI development team and now processes higher volumes with fewer people.

At Mevrik, a firm specialising in digital customer experience in finance and taxation, AI now acts as the first responder in client interactions.

"We're handling nearly 100,000 repetitive queries daily for a single client, which otherwise would need hundreds of human agents," said Reyad Hossain, head of operations.

Newer entrants like Digipro Solutions Ltd, founded in 2022, are also using AI to develop accounting and finance tools.

"Our AI-driven reconciliation tools can detect mismatches in financial reports far quicker than a large manual team," said the company's manager, Maruf Mosharraf.

FUTURE WILL DEPEND ON ADAPTABILITY

As AI takes over repetitive, routine tasks, humans are having to evolve and switch to more complex problem-solving.

"Initially, our call centre system connected customers directly to human agents. Now, the initial interaction is handled by a pre-recorded AI agent," SkyTech's Musnad said. "Only when the AI identifies a potential lead does it transfer the call to a human."

Tasks like reporting lost cards or checking account balances are now largely automated. However, more complex services, such as new card issuance or detailed inquiries, still require human involvement, he said.

"Even in invoice processing, where AI handles data extraction and validation, SkyTech still relies on people for data quality checks and report generation. We saw a reduction in basic-level roles and lost significant man-hours as repetitive tasks were automated," Musnad said.

Yet demand for second-level support remains steady, and SkyTech is now investing in upskilling.

"We are allocating a budget for training and skill development," he added. "But this transformation requires government support. If the government doesn't step in to subsidise training costs, we risk falling behind."

Rakibul Hassan, a leading AI expert in Bangladesh, said the key to adapting to the changes is upskilling -- training staff in AI supervision, data analysis, higher-order thinking, and communication.

Companies are also beginning to offer new AI-powered services, such as predictive analytics and fraud detection, he noted.

"By specialising in these advanced offerings, they can attract new clients and lead in business process transformation," Rakibul said.

Bangladesh's BPO industry was built on manpower, but its future will depend on adaptability. "The firms that survive the AI wave will be those that can reinvent themselves, retrain their workforce, and redefine what outsourcing means in a world where machines increasingly do the talking."

LESSONS FROM GLOBAL HUBS

The influence of AI is not unique to Bangladesh. In India and the Philippines, the world's two largest BPO hubs, companies are also navigating similar transitions, albeit with more robust infrastructure and government support.

In India, many BPO firms have already integrated AI into their service offerings, shifting focus from basic call centre operations to high-value services like robotic process automation (RPA), AI-driven analytics, and customer experience design.

Multinational firms like Tata Consultancy Services and Infosys are investing in AI labs and partnerships, while mid-tier companies are reskilling their teams through industry-academia collaborations, according to a Washington Post report published last month.

In the Philippines, the world's second-largest BPO market, 86 percent of knowledge workers already use AI tools, including emotional-recognition and generative systems, reads a report by The World Financial Review published this March.

These augment tasks like call handling and real-time support, although they are also increasing work pressure and employer monitoring.

NEW POLICY IN THE MAKING

As the impact of AI grows, the government is working on new policies.

"Layoffs in the BPO sector are a concern. In response, we are formulating a National AI Policy that will specifically address these issues," said Faiz Ahmad Taiyeb, special assistant to the chief adviser with executive authority over posts, telecom and ICT.

"In addition, several scattered training programmes currently underway in different wings of the ICT Division will be consolidated to provide structured AI upskilling opportunities. We have also developed training curricula, and a dedicated project will soon be undertaken to implement this," he added.

Skill development initiatives will focus on enabling the workforce to use AI tools for software development, image editing, and other tasks alongside legacy systems.

"We are addressing AI adoption slowly, but we believe that by listening to all stakeholders, we can develop a more comprehensive and inclusive response," he added.