Good industry conferences rarely provide ready-made answers. Instead, they expose you to different ways of thinking about the same problems.
That was certainly our impression after two days of discussions at this year’s Credit & Collections Summit Asia in Jakarta.
Bringing together banks, digital lenders, fintechs and technology providers from across Southeast Asia, the conference covered a wide range of topics from AI and digital collections to governance, portfolio quality and customer engagement. The region’s rapidly evolving lending markets provide a particularly interesting context for these discussions, shaped by differences in financial infrastructure, regulation, digital adoption and financial inclusion.
Yet what stood out was how familiar many of the underlying questions were. European financial institutions may operate in a more mature and highly regulated environment, but banks and lenders everywhere are increasingly asking the same questions. How can risk be identified earlier? How should AI support, not replace, human decision-making? And perhaps most importantly: should collections still be viewed as the last mile of the credit journey, or as a strategic function that continuously improves every stage of the credit lifecycle?
This article explores three ideas that emerged across several presentations and discussions during the conference. Ideas that deserve attention well beyond Southeast Asia. Because although markets differ, the opportunity to learn from each other’s experiences has never been greater. The future of credit management will not be shaped by where innovation happens first, but by how effectively institutions adapt the best ideas to their own operating models.
Collections is no longer the last mile
For a long time, collections was treated as the final stage of the credit lifecycle. A customer was approved, the loan was serviced, risk was monitored, and only when repayment problems emerged did collections enter the picture.
The discussions in Jakarta challenged that linear view.
A collection case contains information that is valuable far beyond the recovery process itself. Which customers are struggling? What behaviours preceded the deterioration? Which interventions led to a sustainable resolution, and which merely delayed the next missed payment? What did the institution get right or wrong when the credit was originally granted?
These questions make collections an important source of intelligence for the wider credit organisation. When collection outcomes are systematically captured and fed back into underwriting, behavioural scoring, limit management, pricing and treatment strategies, the credit lifecycle becomes a continuous process rather than a sequence of disconnected functions.
The value is not simply in collecting more data. It is in using what happens during the life of a loan to improve the decisions that come next. A better understanding of what drives delinquency can inform earlier risk detection, more relevant interventions and, ultimately, better credit decisions.
The implication is significant: collections is no longer simply where the credit journey ends. It is where the institution can learn whether the journey worked.
And that insight should travel back upstream to the next risk assessment, the next credit decision and the next customer interaction.
Early warning creates room to act
If collections is no longer the last mile, the next question is how early a lender can recognise that a customer is moving towards difficulty.
Early warning is not a new concept. Most banks and lenders already have some combination of risk scores, behavioural indicators, dashboards, alerts and increasingly, AI-driven models designed to identify emerging risk. The challenge is often not detecting the signal but deciding what to do with it. An early warning should create a window in which the outcome can still be changed.
An early warning can provide an opportunity to act while the customer is still performing, through proactive communication, a change in treatment, closer monitoring or other forms of early intervention. The aim is not to start collections earlier, but to intervene early enough to prevent the situation from becoming a collections case.
This is also where early warning and collections become part of the same feedback loop. Early warning identifies emerging risk, interventions aim to prevent deterioration, and collections provides valuable insight into what happened and what worked. That insight can then improve the next risk assessment and the next intervention.
The important shift is from periodic risk monitoring to continuous risk management. The question is no longer simply whether a customer is at risk, but whether the institution can recognise changes early enough to act while there is still an opportunity to change the outcome. The stronger the feedback loop between risk, intervention and collections, the more the credit lifecycle moves from a series of handovers to a continuous process of intervention and improvement.
From channels to orchestrated journeys
The next step is not more channels. It is better orchestration.
Collections has never had more ways to reach a customer. SMS, email, WhatsApp, IVR, chatbots, voicebots and human agents can all play a role. But adding channels does not automatically create a better customer journey. The opportunity is to make those channels work together dynamically.
AI can help orchestrate that journey around the customer rather than around the channel. Segment, product, behaviour, contactability, timing and the response to the previous interaction can all influence what happens next. A digital reminder may be enough for one customer; another may need a voice interaction, a human conversation or a different treatment altogether. If an interaction generates no response, the next action can change. If the customer engages, the journey can move in a different direction.
This is where workflow intelligence becomes particularly relevant. AI can connect segmentation, decision rules, channels and outcomes into a single flow, while keeping human intervention where it adds the most value. It can also help reduce unnecessary contact rather than simply increasing it, suppressing over-communication and allowing agents to focus on conversations that genuinely require them.
The goal is not to automate every interaction. It is to make every interaction more intentional.
But orchestration should not be confused with automation for its own sake. The more AI becomes embedded in credit and collections workflows, the more important governance, explainability, human oversight and clear business ownership become. AI should strengthen decision-making and execution at scale, while keeping the right decisions under appropriate human control.
That is a more meaningful AI opportunity for collections: not simply doing individual tasks faster, but orchestrating the next best interaction throughout the customer journey.
Conclusion – beyond the silos
The most interesting takeaway from Jakarta was not a particular technology or a single new approach to collections. It was the opportunity to look at the credit lifecycle differently.
Risk assessment, early warning and collections are often treated as separate capabilities, owned by different teams and measured through different lenses. Yet they are part of the same cycle. A risk assessment shapes the initial decision; customer behaviour creates new signals; early warning creates an opportunity to intervene; and collections generates evidence about what happened and what worked. That insight should then feed back into the next risk assessment, early-warning signals and intervention strategies.
Making this cycle work requires more than technology. Risk and Collections need a shared view of the customer and the portfolio, supported by common data, clear ownership and measures that follow the customer journey rather than organisational boundaries. AI can strengthen this model by orchestrating interactions and supporting decisions at scale, but within a framework of appropriate governance, human oversight and measurable outcomes.
The result is a more connected credit lifecycle: one in which information does not stop at organisational boundaries, decisions are informed by what happened before, and execution creates insight for what comes next.
Perhaps that is the broader lesson from Jakarta. The markets may be different, but the opportunity is shared: to connect data, decisions, people and technology so that the credit lifecycle becomes more responsive and better with every outcome.
Stay tuned for more insights as we continue to explore the latest trends shaping the future of finance, and feel free to book an appointment with our expert any time.
