Technology

Average Bank Balance Calculation for Underwriting

|Posted by Hitul Mistry / 18 Nov 26

Last reviewed: November 2026

Average bank balance calculation for underwriting is the method a lender uses to turn a borrower's bank statement balances into one average bank balance (ABB) figure that drives loan eligibility. When the method lives in analysts' spreadsheets instead of the credit policy, two people can read the same statement and approve different amounts.

This guide is for heads of underwriting and credit policy heads at non-banking financial companies (NBFCs), banks and fintech lenders. It covers which ABB method to write into policy, which transactions to adjust, how ABB should feed eligibility and how Digiqt's custom AI development team can build an engine that applies it the same way every time.

Key Takeaways

  • ABB has no regulatory formula. The method must be fixed in your board-approved credit policy.
  • Dates and adjustments decide the answer. Sampled dates, months covered and excluded transactions move ABB more than any other choice.
  • Check the equated monthly instalment (EMI) date, not just the average. An account can meet ABB and still bounce.
  • Consistency can be engineered. A rules-plus-model engine can compute ABB the same way on every file and show its working.

What Is Average Bank Balance Calculation for Underwriting?

Average bank balance calculation for underwriting is a policy-defined averaging method, not a single industry formula. It is narrower than spotting edited statements, covered in fake bank statement detection, and separate from ratio work on audited accounts in financial statement spreading for lenders.

How Is ABB Different From ADB and AMB?

ABB is different from average daily balance (ADB) and average monthly balance (AMB) mainly in which balances it samples. Lenders use the three labels loosely, so the policy must define each one.

MeasureUsual MethodBest For
ADBEvery day's closing balance, divided by daysFullest view of funds held
AMBMonth-end or in-month daily balances, averaged per monthSix- or twelve-month trend
ABBClosing balances on sampled dates, such as the 5th, 15th and 25thQuick view of funds around EMI dates

Why Do Two Analysts Get Different ABB From the Same Statement?

Two analysts get different ABB from the same statement because five choices are left open: dates, months, accounts, adjustments and holiday handling. According to Terra Insight's guide to bank statement analyser metrics, ABB, fixed obligation to income ratio (FOIR) and income rules are not set by the Reserve Bank of India (RBI); each lender fixes them in its credit policy.

Open ChoiceAnalyst AAnalyst B
Sampled dates5th, 15th, 25th1st, 10th, 20th, month-end
Months coveredLast 6Last 12
AccountsPrimary current accountAll accounts, netted
Own-account transfersLeft inRemoved
Date is a holidayPrevious day's balanceNext day's balance

According to RBI's NBFC Credit Risk Management Directions, 2025 (RBI/DOR/2025-26/350, 28 November 2025), NBFCs must put in place board-approved policies for their credit activities. An ABB method that differs by analyst is a gap in that policy.

What Does Manual ABB Calculation Cost a Lender?

Manual ABB calculation costs analyst time on every self-employed file and creates eligibility differences that nobody can explain later. In this illustrative scenario, a lender assesses 1,500 bank-surrogate files a month with 12 months of statements.

MeasureSpreadsheet ABBEngine ABB With Review
Minutes per file205
Analyst hours a month500125
Hours released–375, about two full-time analysts
ABB methods in useOne per analystOne policy version

The time saving is the smaller gain. The larger one is that a credit committee, auditor or co-lending partner gets the same answer when they recompute ABB.

Which ABB Method Should a Credit Policy Specify?

A credit policy should specify five ABB parameters: sampled dates, months covered, accounts included, holiday treatment and minimum-balance checks. Each one should be written as a rule an engine can apply without judgement.

Which Dates Should an ABB Calculation Use?

An ABB calculation should use dates that bracket the borrower's cash cycle and the EMI date. Common choices are the 5th, 15th and 25th, or the EMI debit date itself, and ADB is the safer option where balances swing sharply within the month.

Borrower PatternSuggested MethodReason
Salaried, steady creditsThree sampled datesLow intra-month swing
Trader with weekly salesDaily average (ADB)Sampled dates can be gamed
Seasonal business12 months, monthly averagesCaptures off-season
EMI date already fixedSampled dates plus the EMI dateTests real debit-day cover

How Many Months Should ABB Cover?

ABB should cover 12 months for self-employed and seasonal borrowers and at least six months for salaried borrowers. Shorter windows reward balance-building just before an application.

How Should ABB Treat Multiple Accounts and Overdrafts?

ABB should treat multiple accounts by computing each account separately and combining them only after own-account transfers are removed. Netting first double-counts money that simply moved between the borrower's own accounts.

Cash credit and overdraft accounts need a different measure, because their balances are negative by design. Average utilisation against the sanctioned limit, plus the headroom left on sampled dates, shows the same repayment comfort that ABB shows on a current account.

Account SituationABB Treatment
One current or savings accountStandard policy method
Two or more accounts of the borrowerPer-account ABB, transfers removed, then summed
Cash credit or overdraftUtilisation and headroom, not ABB
Joint account with a non-applicantInclude only if the policy allows, and flag it

Which Transactions Should Be Adjusted Before Computing ABB?

Four transaction types should be adjusted before computing ABB: own-account transfers, round-trip credits, loan disbursal credits and returned cheque deposits. Each one lifts balances on sampled dates without adding repayment capacity.

AdjustmentHow to DetectTreatment
Own-account transferCounterparty matches the borrower's other accountRemove from credits and balance
Round-trip creditLarge credit returned to the same party within daysDeduct from balances in between
Loan disbursal creditNarration or lender name in the creditRemove for eligibility
Returned cheque depositCredit reversed with a return entryRemove both entries

IDfy has described how some borrowers lift ABB through temporary transfers from friends and family over the Unified Payments Interface (UPI). Matching counterparties across months is the practical defence, while edited statements belong to a lending fraud detection AI agent.

Do your analysts compute ABB with their own dates and their own exclusions?

See Digiqt's Custom AI Development Services

See how Digiqt scopes and builds credit models for lenders, or talk to our AI engineering team.

How Can Custom AI Development Make ABB Calculation Consistent?

Custom AI development makes ABB calculation consistent by encoding the credit policy as rules and adding a model that tags transactions those rules cannot catch. Digiqt can build this ABB engine for your team and connect it to your loan origination system (LOS).

Digiqt's custom AI development service builds AI solutions through a discovery sprint, feasibility study, model selection and validation, then deployment and continuous improvement. Its services include custom machine learning (ML) modelling, explainable AI, business process automation and analytics.

StepManual TodayWith a Digiqt-Built ABB Engine
InputPDF statements retyped into a spreadsheetStatement data from parsed PDFs or the account aggregator
DatesChosen by the analystPolicy dates applied, holiday rule included
AdjustmentsSpotted if the analyst noticesCustom ML model tags own-account and round-trip credits
CalculationFormula varies by sheetOne versioned policy calculation
ExplanationHard to reconstructExplainable output shows each balance and adjustment
MonitoringNoneAnalytics on ABB, exceptions and overrides by branch

Final eligibility decisions and policy approvals stay with your credit team. Where statements arrive through the account aggregator framework, the engine can work from structured data rather than scanned pages.

Worked Example: One Statement, Two ABB Results

In this illustrative example, adjusting one round-trip credit cuts ABB from ₹1,50,000 to ₹1,35,000 and the eligible EMI from ₹75,000 to ₹67,500. The policy samples the 5th, 15th and 25th, and caps EMI at half of ABB; three months are shown for brevity.

Month5th15th25thMonthly Average
July₹45,000₹1,20,000₹2,85,000₹1,50,000
August₹30,000₹1,05,000₹2,55,000₹1,30,000
September, as reported₹60,000₹1,40,000₹3,10,000₹1,70,000
September, adjusted₹60,000₹1,40,000₹1,75,000₹1,25,000

A ₹1,35,000 credit arrived on 24 September from a relative and went back on 27 September, so the engine deducted it from the 25th balance. Adjusted ABB is ₹4,05,000 divided by three, or ₹1,35,000.

The lowest 5th-of-month balance was only ₹30,000, below the ₹67,500 EMI. Credit therefore moved the EMI date to the 15th, where every balance was above ₹1 lakh.

How Should ABB Feed Loan Eligibility?

ABB should feed loan eligibility through a product-specific EMI multiple plus two guard checks: the balance on the EMI date and the number of returned debits. RBL Bank, for example, runs a banking surrogate programme with an ABB variant whose eligibility is calculated on the average balance over the review period.

Should Lenders Use ABB or Banking Turnover?

Lenders should use ABB for borrowers who hold balances and banking turnover for traders whose money moves through the account quickly. RBL Bank offers both variants, with loans from ₹5 lakh to ₹35 lakh.

Whichever measure you choose, run it through your credit decision engine and record overrides with credit policy deviation tracking. Thin-file borrowers need wider evidence, covered in income estimation for thin file borrowers, and our cash flow underwriting AI agent page shows the wider cash-flow view.

When the ABB method changes, roll it out the way credit policy change rollout describes, so old and new files can be told apart.

Could your credit committee recompute any file's ABB and get the same number?

Plan an ABB Engine With Digiqt

Find out how a discovery sprint turns your ABB policy into a tested model, or visit Digiqt to arrange a scoping call.

Sources

Read our latest blogs and research

Featured Resources

Technology

Fake Bank Statement Detection: Catch Edited PDFs Early

Fake bank statement detection for lenders: how edited PDFs are made, the forensic and cross-checks that catch them, and where to place them before disbursal.

Read more
Technology

Financial Statement Spreading for Lenders: Ratios You Trust

Financial statement spreading for lenders: what a spread must capture, a 7-step workflow, 8 checks using Schedule III and CARO disclosures, and metrics.

Read more
Technology

Account Aggregator for Lenders: Faster Credit Decisions

Account aggregator for lenders explained: how consented bank data flows, where it speeds underwriting and monitoring, and what an FIU integration needs.

Read more

About Us

We are a technology services company focused on enabling businesses to scale through AI-driven transformation. At the intersection of innovation, automation, and design, we help our clients rethink how technology can create real business value.

From AI-powered product development to intelligent automation and custom GenAI solutions, we bring deep technical expertise and a problem-solving mindset to every project. Whether you're a startup or an enterprise, we act as your technology partner, building scalable, future-ready solutions tailored to your industry.

Driven by curiosity and built on trust, we believe in turning complexity into clarity and ideas into impact.

Our key clients

Companies we are associated with

Life99
Edelweiss
Aura
Kotak Securities
Coverfox
Phyllo
Quantify Capital
ArtistOnGo
Unimon Energy

Our Offices

Ahmedabad

B-714, K P Epitome, near Dav International School, Makarba, Ahmedabad, Gujarat 380051

+91 99747 29554

Mumbai

C-20, G Block, WeWork, Enam Sambhav, Bandra-Kurla Complex, Mumbai, Maharashtra 400051

+91 99747 29554

Stockholm

Bäverbäcksgränd 10 12462 Bandhagen, Stockholm, Sweden.

+46 72789 9039

Malaysia

Level 23-1, Premier Suite One Mont Kiara, No 1, Jalan Kiara, Mont Kiara, 50480 Kuala Lumpur

Lewes

16192 Coastal Highway, Lewes, Delaware 19958, USA

software developers ahmedabad
ISO 9001:2015 Certified

Call us

Career: +91 90165 81674

Sales: +91 99747 29554

Email us

Career: hr@digiqt.com

Sales: hitul@digiqt.com

© Digiqt 2026, All Rights Reserved