I’ve spent a large part of my career in the financial sector, and if there’s one hard-earned lesson I keep coming back to, it’s this: you don’t really know your business until you understand your collections.
For many lending startups, collections are treated as an operational afterthought. Something to be optimised later. Something that sits at the end of the value chain, far removed from growth, product, or underwriting. That’s a dangerous assumption. Collections is not a back-office function, but rather, the final determinant of whether your unit economics survive contact with reality.
I’ve seen otherwise promising lending models collapse not because of poor underwriting or lack of capital, but because collections quietly bled margins month after month. CAC was carefully tracked, disbursals were celebrated, cohort curves looked fine in the first 30 days – and yet the business struggled to ever become meaningfully profitable.
If you’re building or scaling a lending startup, here are seven collections mistakes that can quietly but decisively kill your unit economics.
1. Over-reliance on manual processes and human judgment
Human intuition has its place, but manual-heavy collections processes don’t scale and they leak money in ways that are hard to detect.
When outcomes depend largely on individual agent judgment, inconsistency creeps in. Promises-to-pay are recorded differently, follow-ups are missed, escalations are delayed, and compliance risks increase. Each small failure compounds into real financial loss.
Manual operations also make it harder to measure what’s actually working. You can’t easily attribute recoveries to strategies, scripts, or channels when execution varies wildly across agents and agencies.
Automation doesn’t mean removing humans from the loop. It means using technology to standardise decisions, prioritise effort, and free up human bandwidth for cases where judgment truly matters. Startups that fail to make this shift early often find their collections costs rising faster than their loan book.
2. Using the same collections strategy for every borrower
Not all delinquencies are created equal, yet many lenders treat them that way.
A salaried borrower who missed an EMI due to a temporary cash flow mismatch is very different from a chronically delinquent borrower gaming the system. Applying the same tone, channel, and escalation path to both is inefficient and expensive.
When you don’t segment your borrowers, your collections engine ends up overworking low-risk cases and underperforming on high-risk ones. This results in higher costs, lower recoveries, and frustrated customers.
Effective collections require intelligent segmentation, by risk profile, behaviour, past repayment patterns, and even channel responsiveness. Digital nudges may work for one segment; human intervention may be necessary for another.
Without this nuance, collections become blunt-force trauma instead of a precision tool.
3. Ignoring early-stage delinquencies
Many lenders focus disproportionate energy on late-stage collections – accounts that are already 60, 90, or 120 days past due. By then, a lot of damage is already done.
Early-stage delinquencies are where unit economics are won or lost. A missed EMI at day 1 or day 3 is a signal, not a failure. How you respond in those first few days has an outsized impact on eventual recovery.
Gentle reminders, flexible repayment options, and timely nudges can prevent accounts from ever rolling forward. These interventions are cheaper, more customer-friendly, and far more effective than aggressive action later.
Startups that underinvest in early collections often find themselves spending exponentially more to recover the same rupee months down the line.
4. Optimising for short-term recovery at the cost of long-term value
Aggressive collections tactics can boost short-term numbers, and quietly destroy long-term economics.
When borrowers feel harassed or disrespected, they don’t come back. They don’t refinance. They don’t recommend your product. And in a world of social media and instant reviews, they often make sure others don’t either.
This is especially dangerous for startups building mass or near-prime lending businesses, where repeat usage and cross-sell are critical to profitability. A single bad collection experience can wipe out the lifetime value of an otherwise good customer.
Sustainable collections balance firmness with fairness. The goal is not just to recover dues, but to preserve trust wherever possible. Founders who ignore this trade-off often overestimate their true unit economics.
5. Poor visibility into agency performance
Outsourcing collections doesn’t mean outsourcing accountability.
Many startups work with multiple agencies but lack granular visibility into how each one performs beyond headline recovery numbers. Without detailed metrics, such as cost per recovery, resolution time, customer complaints, or compliance adherence, it’s impossible to know which partners are actually creating value.
Worse, some agencies may boost short-term recoveries through practices that increase disputes, legal exposure, or reputational risk. The bill for that behaviour usually arrives much later.
High-performing lenders treat agencies as extensions of their own operations. They invest in monitoring, data integration, and continuous performance evaluation. Without this discipline, collections costs quietly spiral while quality erodes
6. Not feeding collections insights back into credit and product decisions
Perhaps the most underestimated mistake is treating collections data as terminal data.
Collections outcomes are some of the richest signals you have about borrower behaviour, product design flaws, and underwriting blind spots. When this data sits in a silo, the organisation loses a powerful feedback loop.
Which customer segments are repeatedly delinquent? Which repayment schedules create friction? Which loan tenures or ticket sizes perform poorly under stress? Collections knows the answers, but only if someone is listening.
Startups that fail to close this loop keep repeating the same mistakes at scale. Those that integrate collections insights into credit policy, pricing, and product design steadily improve their unit economics over time..
7. Treating collections as a pure cost centre
The most common mistake is viewing collections purely through the lens of cost reduction.
Founders often ask: How do we reduce cost per call? How do we shrink our agency bill? How do we automate more so we can run leaner? These are reasonable questions – but incomplete ones.
Collections are not just about cost. It’s about recoveries, timing, and customer lifetime value. Cutting costs aggressively without understanding recovery impact is one of the fastest ways to destroy unit economics.
A Rs. 100 saved on collections that leads to Rs. 1,000 less recovered is not optimisation – it’s self-sabotage. Worse, poorly handled collections increase customer churn, brand damage, and future credit losses. The real cost rarely shows up immediately on a P&L sheet.
The more mature way to look at collections is as a revenue-protection and value-maximisation function. Every decision should be evaluated on net recovery impact, not just operational expense.
Collections are an uncomfortable topic that forces founders to confront the gap between expected and realised value. It exposes operational weaknesses that growth metrics often hide.
But that discomfort is also where clarity comes from.
The lending startups that survive and scale are not the ones with the most aggressive growth or the cheapest capital. They are the ones that respect collections as a strategic function – one that deserves the same thoughtfulness as underwriting, product, and risk.
If you’re serious about your unit economics, don’t wait for collections to become a problem before you address it. By the time the strain is clearly reflected in your numbers, the underlying damage has usually been compounding for months.
