Numerical Reasoning in Banking Assessments
Numerical reasoning tests are one of the most important hurdles in banking recruitment. Whether you are applying to Goldman Sachs, JPMorgan, Morgan Stanley, or any other major financial institution, you will almost certainly face a quantitative assessment early in the hiring process. These tests measure your ability to interpret financial data, perform calculations under pressure, and draw logical conclusions from numbers. This guide covers everything you need to know, from the specific question types banks use to worked examples and targeted preparation strategies.
Why Banks Prioritize Numerical Reasoning Skills
Banks use numerical reasoning tests because every role in finance depends on accurate quantitative analysis. Analysts at Goldman Sachs build financial models. Relationship managers at HSBC evaluate credit risk. Traders at Morgan Stanley make split-second decisions based on market data. Even operations roles at Citi and Bank of America require the ability to spot errors in transaction data and reconcile figures across systems.
Numerical reasoning tests give recruiters a standardized, objective way to measure these skills across thousands of applicants. A single graduate program at JPMorgan might receive over 50,000 applications in a single cycle, and a numerical test is one of the fastest ways to identify candidates who can handle the quantitative demands of the job.
Beyond raw calculation ability, these tests also measure how you work under time pressure, how accurately you read data, and whether you can distinguish relevant information from background noise in a financial dataset. These are exactly the skills you will use every day on the job.
💡Banks do not use numerical tests to trick you. They use them because the day-to-day work genuinely requires strong quantitative skills. If you can handle the test, you can handle the role.
Question Types Specific to Banking Assessments
Banking numerical tests go beyond basic arithmetic. The questions are designed to mirror real financial scenarios you would encounter on the job. Here are the most common types you will see:
Profit margin and profitability analysis. You will be given revenue and cost data for a company or division and asked to calculate gross margin, operating margin, or net profit margin. These questions test whether you understand the relationship between revenue, costs, and profit.
Revenue growth and year-over-year comparison. Questions present financial data across multiple time periods and ask you to calculate percentage growth, compound annual growth rates (CAGR), or identify trends in performance.
Ratio analysis from balance sheets. You may be given a simplified balance sheet and asked to calculate debt-to-equity ratios, current ratios, or return on equity (ROE). Investment banking and asset management assessments are especially likely to include these.
Currency conversion and exchange rate calculations. Global banks like HSBC, Deutsche Bank, and UBS frequently include questions involving multi-currency transactions, requiring you to convert between currencies and calculate net values.
Interest rate and compound growth problems. These questions ask you to calculate simple or compound interest, future values, or the effect of rate changes on loan portfolios or investment returns.
Data interpretation from charts and tables. Nearly every banking test includes questions where you must extract data from bar charts, line graphs, pie charts, or multi-row financial tables before performing calculations.
If you want to start practicing these question types right away, try the free practice tests available on our platform.
Worked Examples With Financial Data
Working through realistic examples is the best way to prepare. Here are three problems that reflect the style and difficulty of actual banking assessments.
Example 1: Profit Margin Calculation
A retail banking division reported the following results for Q3:
| Metric | Amount |
|---|---|
| Total Revenue | $48,500,000 |
| Cost of Goods Sold | $29,100,000 |
| Operating Expenses | $11,640,000 |
| Interest Expense | $1,940,000 |
Question: What is the operating profit margin for Q3?
Solution:
- Operating Profit = Revenue - COGS - Operating Expenses
- Operating Profit = $48,500,000 - $29,100,000 - $11,640,000 = $7,760,000
- Operating Profit Margin = ($7,760,000 / $48,500,000) x 100 = 16.0%
Note that interest expense is not included in operating profit. Banking tests often include extra data to see if you know which figures to use and which to ignore.
Example 2: Revenue Growth Rate
An investment bank's advisory division reported the following annual revenues:
| Year | Revenue ($m) |
|---|---|
| 2022 | 3,200 |
| 2023 | 3,520 |
| 2024 | 3,960 |
Question: What was the percentage revenue growth from 2023 to 2024, and how does it compare to the growth from 2022 to 2023?
Solution:
- Growth 2022-2023 = ((3,520 - 3,200) / 3,200) x 100 = (320 / 3,200) x 100 = 10.0%
- Growth 2023-2024 = ((3,960 - 3,520) / 3,520) x 100 = (440 / 3,520) x 100 = 12.5%
- The growth rate increased by 2.5 percentage points from one year to the next.
Example 3: Ratio Analysis From a Balance Sheet
A bank's simplified balance sheet shows:
| Item | Amount ($m) |
|---|---|
| Total Assets | 850 |
| Total Liabilities | 680 |
| Shareholders' Equity | 170 |
| Net Income | 34 |
Question: Calculate the debt-to-equity ratio and the return on equity (ROE).
Solution:
- Debt-to-Equity Ratio = Total Liabilities / Shareholders' Equity = 680 / 170 = 4.0
- ROE = (Net Income / Shareholders' Equity) x 100 = (34 / 170) x 100 = 20.0%
A debt-to-equity ratio of 4.0 is common for banks, which are naturally more leveraged than companies in other industries. An ROE of 20% would be considered strong performance.
Example 4: Currency Conversion
A client holds EUR 2,400,000 in a Deutsche Bank account and wants to convert to USD. The current exchange rate is EUR 1 = USD 1.085. The bank charges a 0.3% conversion fee on the USD amount.
Question: How much USD will the client receive after fees?
Solution:
- Gross USD = 2,400,000 x 1.085 = $2,604,000
- Conversion fee = $2,604,000 x 0.003 = $7,812
- Net USD received = $2,604,000 - $7,812 = $2,596,188
These worked examples reflect the difficulty level you can expect at most major banks. Practice similar questions with the numerical reasoning test to build both speed and accuracy.
Financial Terminology You Need to Know
Knowing the right terminology saves valuable seconds during your test. You do not need a finance degree, but you should be comfortable with these terms before test day:
- Revenue (Sales/Turnover): Total income before any deductions.
- Gross Profit: Revenue minus cost of goods sold (COGS).
- Operating Profit (EBIT): Gross profit minus operating expenses, before interest and tax.
- Net Profit: The bottom line after all expenses, interest, and taxes.
- Profit Margin: Profit as a percentage of revenue. Can be gross, operating, or net.
- ROE (Return on Equity): Net income divided by shareholders' equity, expressed as a percentage.
- Debt-to-Equity Ratio: Total debt divided by total equity. Measures financial leverage.
- Current Ratio: Current assets divided by current liabilities. Measures short-term liquidity.
- CAGR (Compound Annual Growth Rate): The smoothed annual growth rate over a multi-year period.
- Basis Points (bps): One basis point equals 0.01%. Used to describe interest rate changes.
💡You do not need to memorize complex financial formulas. Focus on understanding profit margins, growth rates, and basic ratios. These cover the vast majority of banking test questions.
Test Providers That Banks Use
Different banks partner with different test providers, and each provider has its own format and style. Knowing which provider your target bank uses allows you to practice with the right type of questions.
SHL is the most widely used provider in banking recruitment. Goldman Sachs, Barclays, HSBC, and Citi all use SHL numerical reasoning tests. SHL tests are known for tight time limits, typically giving you about 60 to 90 seconds per question. Questions are data-heavy and usually involve interpreting tables or charts.
Kenexa (IBM) is used by JPMorgan, among others. Kenexa tests tend to have slightly longer time allowances per question but may include more complex multi-step problems. The data sets can be larger, and questions sometimes require you to combine information from multiple tables.
cut-e (Aon) is used by some European banks including Deutsche Bank. These tests are adaptive, meaning the difficulty adjusts based on your performance. Getting early questions right leads to harder subsequent questions, which can earn you a higher score.
Saville Assessment is used by select banks and financial services firms, particularly for management and leadership roles. The format emphasizes data interpretation with business-relevant scenarios.
Talent Q (Korn Ferry) uses adaptive testing and is employed by UBS and some other institutions. The format is similar to SHL but with dynamic difficulty adjustment.
| Bank | Primary Test Provider | Typical Format |
|---|---|---|
| Goldman Sachs | SHL | 20-25 questions, 25-35 minutes |
| JPMorgan | Kenexa (IBM) | 20-30 questions, 30-40 minutes |
| Morgan Stanley | SHL | 18-24 questions, 25-35 minutes |
| Citi | SHL | 18-25 questions, 25-30 minutes |
| Bank of America | SHL / Kenexa | 20-25 questions, 25-35 minutes |
| Barclays | SHL | 18-24 questions, 25-30 minutes |
| HSBC | SHL | 20-25 questions, 30-35 minutes |
| Deutsche Bank | cut-e (Aon) | Adaptive, 20-25 questions |
| UBS | Talent Q | Adaptive, 18-24 questions |
| Credit Suisse | SHL / cut-e | 20-25 questions, 25-35 minutes |
Banking Sector Comparison: Test Characteristics by Role Type
The type of banking role you apply for affects the difficulty and focus of the numerical test. Here is how tests differ across sectors:
| Sector | Difficulty Level | Key Focus Areas | Time Pressure | Typical Cut-off |
|---|---|---|---|---|
| Investment Banking | High | Financial modeling, ratio analysis, multi-step problems | Very High | 80th percentile+ |
| Retail Banking | Moderate | Customer data, transaction analysis, percentages | Moderate | 65th-70th percentile |
| Asset Management | High | Portfolio returns, compound growth, fund performance | High | 75th-80th percentile |
| Insurance | Moderate | Risk calculations, statistical data, probability | Moderate | 65th-75th percentile |
| Fintech | Moderate-High | Data interpretation, growth metrics, unit economics | High | 70th-80th percentile |
Investment banking roles at firms like Goldman Sachs and Morgan Stanley have the toughest numerical tests, both in terms of question difficulty and cut-off scores. Retail banking roles at HSBC or Bank of America are still demanding but typically allow slightly more time per question and focus more on practical business calculations.
💡Tailor your preparation to your target sector. If you are applying to investment banking, focus heavily on ratio analysis and multi-step financial problems. For retail banking, prioritize data interpretation and percentage calculations.
Preparation Strategies Specific to Finance
Generic numerical reasoning practice is a good starting point, but you need finance-specific preparation to score well on banking assessments. Here are the most effective strategies:
Build financial literacy first. Before you start timed practice, make sure you understand basic financial statements. Know the difference between gross profit and net profit. Understand what a balance sheet tells you versus an income statement. This foundational knowledge lets you focus on calculation speed during the test rather than struggling with terminology.
Practice under realistic time pressure. Banking tests are fast. SHL tests typically give you 60 to 90 seconds per question, and every second counts. Start your practice untimed to build accuracy, then gradually reduce your time limits until you can consistently answer questions within the target window.
Learn to identify irrelevant data. Banking test questions almost always include more data than you need. A table might show eight columns, but the answer requires only two. Practice scanning data quickly and identifying exactly which numbers feed into your calculation.
Master mental math shortcuts. Calculating 15% of a number is faster if you find 10% and add half of it. Finding 12.5% is faster if you divide by 8. These shortcuts save seconds that add up over 25 questions.
Review your mistakes systematically. After every practice session, go back and analyze every question you got wrong. Was it a calculation error, a data reading mistake, or a conceptual gap? Each type of mistake requires a different fix.
Ready to start practicing? The All Test Package gives you access to numerical reasoning tests, abstract reasoning tests, and more, all designed to mirror real banking assessments.
Typical Cut-off Scores and What They Mean
Most banks do not publish their exact cut-off scores, but industry patterns are well established. Your score on a numerical reasoning test is typically reported as a percentile, meaning your performance is ranked against a comparison group of other test-takers.
For graduate analyst roles at top-tier investment banks like Goldman Sachs, Morgan Stanley, and JPMorgan, you generally need to score at or above the 80th percentile. This means you need to outperform 80% of the people who take the same test.
Mid-tier and retail banking roles at institutions like Bank of America, Barclays, and HSBC typically set cut-offs around the 65th to 75th percentile. These are still competitive, but the margin for error is slightly more forgiving.
Asset management firms and insurance companies usually fall somewhere in between, with cut-offs around the 70th to 80th percentile depending on the specific role and firm.
It is worth noting that cut-off scores can vary by region, role level, and recruitment cycle. During peak hiring periods when application volumes are especially high, some banks raise their thresholds to manage the volume of candidates advancing to the next stage.
The best approach is to aim well above the minimum. If the expected cut-off is the 75th percentile, prepare to score at the 85th or higher. This gives you a comfortable buffer and demonstrates strong quantitative ability to the hiring team.
How to Handle Test Day
The numerical reasoning test itself is only part of the challenge. How you manage the test experience matters just as much as your preparation.
Check your setup in advance. If you are taking an online proctored test, verify your internet connection, webcam, and browser compatibility the day before. Technical issues can cost you time or even void your test session.
Read the instructions carefully. Not all banking tests allow calculators. Some use negative marking, meaning incorrect answers reduce your score. Knowing the rules before you start prevents costly assumptions.
Manage your time actively. If a question is taking more than 90 seconds, mark it and move on. It is better to answer 22 out of 25 questions correctly than to spend four minutes on one difficult problem and run out of time for the last five.
Use estimation before calculating. Before you calculate a precise answer, estimate the rough range. If the answer choices are 12%, 18%, 24%, and 31%, and your estimate is around 17-19%, you can focus your calculation on confirming 18% rather than working through every option.
Frequently Asked Questions
Which banks use numerical reasoning tests?
Nearly every major bank uses numerical reasoning tests in their recruitment process. Goldman Sachs, JPMorgan, Morgan Stanley, Citi, Bank of America, Barclays, HSBC, Deutsche Bank, UBS, and Credit Suisse all include quantitative assessments for graduate, analyst, and internship positions.
Do I need finance knowledge for banking numerical tests?
You need a working understanding of basic financial concepts like revenue, profit, margins, and percentages. Deep finance knowledge or a finance degree is not required, but familiarity with common financial terminology will save you valuable time during the test.
Can I use a calculator for banking numerical tests?
It depends on the test provider and the bank. SHL-based tests typically allow an on-screen calculator. Kenexa (IBM) tests sometimes restrict calculator use. Always check the instructions provided before your test begins, and practice both with and without a calculator.
What is a good score on a banking numerical test?
Most banks expect candidates to score in the 70th percentile or above. Top-tier investment banks like Goldman Sachs and Morgan Stanley often set the bar at the 80th percentile for analyst roles. Scoring is relative to other candidates in the same test window.
How long are banking numerical reasoning tests?
Most banking numerical tests last between 25 and 40 minutes and contain 18 to 30 questions. The exact length depends on the bank and test provider. SHL tests tend to be shorter and faster-paced, while Kenexa tests may give slightly more time per question.
Can I retake a banking numerical test if I fail?
Most banks enforce a waiting period of 6 to 12 months before you can reapply and retake the test. Some banks, like JPMorgan, allow only one application per recruitment cycle. Use this time to practice intensively so you pass on your next attempt.
Prepare With Assessment-Training.com
Getting a strong score on a banking numerical reasoning test comes down to targeted practice with realistic questions. Start with the numerical reasoning test to build your skills, or explore the All Test Package for complete preparation across all assessment types.