Agents were incentivised on contact volume and the console made contacting easy and hardship invisible. Regulatory complaints were rising and recovery rates were not.
STUDY 20 / 30 · ANONYMISED · NDA · NBFC & LENDING · INDIA · 2.4M BORROWERS
NBFC & LENDING · INDIA A collections interface designed against harassment.

Built contact limits into the system as hard constraints, made borrower hardship signals prominent, and put a settlement pathway in front of the agent at the moment it becomes relevant.
Complaints fell sharply while recovery improved, because agents spent their contacts on borrowers who could pay and offered structure to those who could not.
THE ARGUMENT
Why the obvious solution was wrong.
The study matters because the product problem was reframed before the interface was polished.
Collections consoles are built to maximise contact throughput, because that is what the incentive measures. The interface therefore makes the next call one click away and makes everything about the borrower's circumstances two clicks away or absent. An agent under a volume target will follow the path the interface makes cheap, and the resulting behaviour is the harassment the regulator eventually acts on.
Contact governance is now enforced by the system rather than by policy: attempts per day, per week, permitted hours and channel limits are hard constraints the agent cannot exceed. Hardship signals - a recent partial payment, a stated job loss, a medical flag, a prior plan - render at the top of the console, above the balance. The settlement pathway is one action from the call screen rather than an escalation. Recovery rose because a structured plan a borrower can meet recovers more than a fourteenth phone call, and complaints fell because the fourteenth call became impossible.
THE INTERFACE CRAFT
The interaction, rendered as a working product surface.
The specimen below is code-native and uses the study's own design logic. The client interface remains protected.
5 factors · 2 overrides
DESIGN DECISIONS
Positions we would defend.
Each decision names the principle and the product consequence, not a stylistic preference.
Constrain the console, not the agent
Policy that depends on an agent under a volume target resisting the cheapest path in the interface is policy that will fail.
Surface circumstance first
An agent who sees hardship before balance has a different conversation. The information order is the intervention.
Make the humane path the fast path
If settlement is slower than another call, agents will make another call. Recovery follows from making the right thing easy.
PRODUCT LEADER READOUT
What transfers, and what should remain specific to this product.
A case study is useful when its operating principle travels without turning the original interface into a template.
Read the operating condition
For Collections platform · consumer lending NBFC, the transferable lesson is not a copied screen. It is the condition the interface had to make legible: Attempts, hours and channels enforced by the system. The agent cannot exceed them, so the policy is not dependent on the agent under a target. Rebuild that visibility for your own roles, risk, terminology, and operating cadence.
Protect the design rule
Policy that depends on an agent under a volume target resisting the cheapest path in the interface is policy that will fail. Keep that rule in the acceptance criteria, component states, and production QA record so later visual cleanup cannot erase why the interaction exists.
Measure behaviour after ship
The evidence record is -78% for regulatory complaints, vs baseline. Recreate the baseline and outcome window before rollout, segment the result by role and context, and state clearly what the measure cannot prove.
RESEARCH RECORD
The work behind the interface.
These artefacts connect the final interaction back to the evidence and product model that produced it.
Call audit
Listened to 200 recorded collections calls with the compliance team, coding for the moment a call became a complaint risk.
Agent shadowing
Sat with 12 agents through full shifts to map how the console's affordances interacted with the volume target.
Hardship signal design
Worked with risk and compliance to define which signals were both predictive and lawful to surface prominently.
Recovery model comparison
Compared 12-month recovery from structured plans versus continued contact for matched borrower cohorts.
“Complaints fell and collections went up. Nobody on my team believed both could happen.”
Head of Collections, consumer NBFC · under NDA
NEXT STUDY · 21 / 30 · NBFC & LENDING