Login to disbursal TAT: how to measure it and where the days actually go

Define the clock before you argue about the number

Half the TAT disputes inside a lender are definitional. Sales counts from customer consent, credit counts from a complete login, operations counts from a clean file. All three are defensible and all three produce different numbers, which is why the monthly review turns into a debate about the denominator instead of the delay.

Fix three things in writing: what event starts the clock, what event stops it, and whether the clock counts calendar hours or working hours. Then publish the definition next to the metric on every dashboard, so nobody has to ask.

• Start: application logged with the minimum mandatory set - not the first phone call, not the pre-qualification.
• Stop: disbursement instruction released, not sanction issued. Sanction-to-disbursal is where files quietly die.
• Pause rules: define exactly when the clock stops for customer-pending items, and cap the number of pauses per file so 'customer pending' does not become a parking bay.
• Working hours: a file logged at 6pm on Friday should not carry a weekend penalty in your operations metric, though it absolutely does in the customer's experience. Report both.

Report median and 90th percentile side by side. If your median is 2 days and your P90 is 14, you do not have a TAT problem - you have a small set of file types that behave completely differently, and they need a separate queue.

Indicative stage-wise TAT by product

The numbers below are the working ranges we see across NBFC and bank retail portfolios in India. Treat them as a shape to compare against rather than a target handed down from a board deck.

The five queues where the days actually go

When we instrument a portfolio and look at where files sit rather than where work happens, the same five queues account for most of the elapsed time. None of them are solved by hiring more credit analysts.

• Incomplete login. The file enters with a missing income proof or an unsigned page, and the first credit touch is a rework request. Checking document completeness and a rough FOIR at the point of sourcing removes this queue almost entirely.
• Sequential verification. Field investigation, tele-verification, employer check and legal-technical run one after another because each one waits for the previous outcome. Most can be triggered in parallel the moment the file is logged.
• Unrequested approvals. A file needing a deviation sits until someone notices it needs a deviation. Auto-routing on the deviation matrix instead of manual escalation typically removes a full day.
• Post-sanction documentation. Sanction goes out, then the customer is asked for the NACH mandate, the insurance nomination, or a co-applicant signature. Collecting these alongside the credit documents shifts them off the critical path.
• Banking and cut-off timing. A disbursement cleared at 4:10pm against a 4pm cut-off is a full extra day to the customer. Publishing an internal cut-off an hour before the banking one is unglamorous and works.

What to instrument first

You cannot manage a queue you cannot see. Before any process redesign, get stage-entry and stage-exit timestamps on every file, with the owner named at each stage. Once ageing is visible by stage and by owner, most of the obvious delay corrects itself within two review cycles, because nobody wants to be the person holding the oldest file in the room.

• Stage ageing report: median, P90 and count of files older than the SLA, by stage, refreshed daily.
• Owner-level ageing: the same view by credit officer, branch and DSA, so accountability is not diffused.
• Reason codes on every pause and every rework, with a monthly Pareto - three reasons usually explain seventy percent of it.
• Rework rate at first credit touch: the cleanest single indicator of sourcing quality.

Where Toolyt fits

Toolyt sits on the sourcing and field side of this clock, which is where the recoverable time usually is. Documents are captured and checked on the phone at the customer's location, eligibility and obligations are computed before the file is logged, allocation and escalation rules move files without anyone chasing, and stage-wise ageing is visible to the branch manager on the same app the team already uses. Files reach credit complete, which is the cheapest way to compress TAT.