London Investment Banking Modelling Tests: Formats, Sample Tasks, Tips

IB Modeling Test Bank: London Formats and Tactics

A modeling test is a timed Excel build or audit that shows how you turn a short brief into a clean, traceable model under pressure. An IB Modeling Test Bank for London candidates is the common set of tasks banks and funds draw from – three-statement builds, LBOs, accretion or dilution, and credit cases – to gauge speed, structure, and judgment. Think of it as a practical screen: can your work stand in front of a client with minimal edits?

London tests reward clarity, ties, and a short read-out that a VP can hand to a client. This guide explains common formats, scoring, and tactics, then adds a few fresh tips on reviewer triage and IFRS nuances so you walk in prepared and walk out with an offer.

Context and objectives – who uses these tests and why

London banks, elite boutiques, and private credit funds run these tests at different stages. Bulge brackets use short screens around assessment centers. Boutiques and buyout funds prefer longer cases with a short write-up and discussion. Private credit teams focus on debt capacity, covenants, and liquidity. The clock is tight: 30 to 90 minutes for Excel screens and 2 to 3 hours for case studies. The objective is narrow – tie the three statements, size buyout returns, estimate accretion or dilution, or underwrite downside liquidity – and the tools allowed are limited.

What firms, candidates, and reviewers want

Firms’ priority: A file a VP can send to a client with minimal rework. Speed helps, but auditability and error control decide outcomes.
Candidate signal: Clear logic, minimal hardcoding, conservative assumptions, and simple checks. Show control under time.
Reviewer trust: Stable links, consistent signs, sources and uses that tie, and a crisp summary of sensitivities. A tidy model earns a real conversation.

Formats you will see in London

Excel skills screen: Formula fluency, data cleaning, shortcuts, and a small build in 30 to 60 minutes – the quick filter.
Three-statement build: Integrate P&L, balance sheet, and cash flow with working capital, D&A, capex, and debt – often with a light DCF or ratios.
Accretion or dilution: Pro forma P&L for an acquisition with purchase accounting, financing mix, synergies, and EPS impact.
LBO mini-case: Sources and uses, operating cases, debt schedule, cash sweep, returns, and sensitivity tables – sometimes with a short read-out.
Paper LBO: No Excel. Estimate IRR from entry or exit multiples, leverage, and cash generation.
Model audit or repair: Find errors, re-link statements, and add checks – the hygiene test.
Private credit underwriting: Debt capacity and covenants with downside. Focus on DSCR, FCCR, leverage, liquidity, and cash dominion.

What good looks like under the clock

A clean test model has one assumptions sheet and a compact build – inputs, calculations, outputs, and checks. Outputs should show either the three statements or transaction schedules, a summary box with key results and sensitivities, and a short note if asked. Checks matter: the balance sheet balances; sources equal uses; cash equals cash; tax does not exceed pre-tax income; debt never goes negative; covenants compute; and interest flows reconcile.

How you will be scored

Hygiene: Consistent styles and signs, separated inputs, no buried constants, and no uncontrolled circularity.
Structure: Clear blocks, transparent steps, and minimal cross-sheet jumps.
Accuracy: Ties, reconciliations, and plausible tax logic that survives a quick wiggle test.
Judgment: Reasonable, stated assumptions that are easy to flex and defend.
Communication: One-page summary of conclusions, drivers, risks, and the sensitivity grid.
Efficiency: Working core in time, extras only after the spine is complete.

Sample tasks and winning approaches

Three-statement build – 60 minutes

Build five-year forecasts from provided historicals with simple drivers: tapering growth, margin expansion, D&A and capex as percent of revenue, steady working capital, a simple debt roll, and a statutory tax rate. Keep the interest loop conservative unless iteration is allowed. Outputs should include FCF, net debt, leverage, coverage, and a two-way sensitivity on growth and margin. If you need a refresher on structure and pacing, practice a compact three-statement model end to end in under an hour.

LBO mini-case – 90 minutes

Price at a given EV or EBITDA and finance with a mix of TLB and notes, adding a revolver if liquidity is tight. Build sources and uses, an operating case anchored on revenue and margin, debt schedules with a cash sweep, and equity returns (IRR and MoIC). Sensitize entry multiple and EBITDA growth. One-line illustration helps: if EV is 1,000, debt 500, equity 500 – exit EV 1,200 with 300 net debt – equity value 900 – MoIC 1.8x pre-costs, then compute IRR from timing. For more ideas on set-up, review this LBO modeling framework to stress-test your approach.

Accretion or dilution – 60 minutes

Build standalone P&Ls for acquirer and target. Layer purchase accounting – intangible amortization and any step-ups – new debt cost, and new shares for stock consideration. Combine results, apply synergy timing and integration costs, tax, and compute EPS. List three risks: synergy capture, amortization drag, and rating headroom. If you want a quick primer, see this succinct accretion or dilution analysis.

Core mechanics under the clock

Sources and uses first: Reconcile equity, debt, fees, and uses. A first-minute tie saves last-minute panic.
Revenue to EBITDA: Use compact drivers. If historicals are noisy, normalize margins to a credible base.
PPE and capex: Roll PPE with capex and D&A and keep PPE non-negative to maintain cash realism.
Working capital: Percent-of-sales is fine under time pressure; switch to days if asked. Practice typical working capital schedules.
Interest: Use average debt and include base-rate floors. Keep loops intentional if iterations are allowed.
Taxes: Anchor to pre-tax income and keep cash or statutory differences simple unless directed.
Cash flow: Reconcile net income to cash or present FCF to firm with tight ties for trustable cash.
Debt sweep: Sweep excess cash after mandatory items. The revolver is the buffer of last resort – revisit a quick primer on a cash sweep if needed.
Purchase accounting: Include intangible amortization in accretion tasks and state the assumption.
Sensitivities: Link only to drivers. Two-way tables on first-order variables for decision clarity.

Architecture and triage under time pressure

Use a simple template you can rebuild fast: Inputs, Calcs, Outputs, Checks. One color for inputs and one for formulas. One operation per row. No buried constants – if you must hardcode a one-off, comment it. Add a checks block with pass or fail flags for the balance sheet, sources versus uses, and debt non-negativity. Read the prompt twice, build the spine first, and park frills. If stuck, move on and leave a note.

Frequent kill tests to avoid

Unbalanced balance sheet: Add a visible check from minute one to reduce screening risk.
Negative debt or cash: Use MAX or MIN and a revolver buffer to protect credibility.
Interest on wrong base: Include an explicit base rate and floor for cost of capital accuracy.
Buried numbers: Centralize inputs to build reviewer trust.
Overcomplication: Simple, auditable steps beat fancy features that do not work.
Missing costs: Do not model synergies without costs to achieve – show realism.
Miscounted shares: Compute issuance and, if needed, weighted averages for EPS accuracy.
Capex placement: Keep capex in investing for cash flow clarity.

Sector nuances that pop up

FIG: Regulatory capital, RWA, provisioning, and net interest margin – not EBITDA.
Infrastructure: DSCR, LLCR, sculpted amortization, and reserve accounts drive debt sizing.
Real estate: NOI, cap rates, LTV or LTC, ICR, and tenancy schedules if provided.
Energy: Production profiles, decline curves, price decks, and unit opex with capex phasing.
TMT or software: ARR, MRR, gross and net retention, churn, and cohort logic where relevant.

Tools and IT reality

Expect bank-provided Windows machines with no macros, add-ins, or internet unless stated. Dynamic arrays and XLOOKUP may be available, but only use them if you can revert to SUMIF or INDEX-MATCH quickly. Python in Excel is off-limits. Treat these limits as baseline and draft the model as if you will hand it to someone on Office 365 with default settings.

Preparation that moves the needle

Replicate the formats: Timed three-statement, LBO, and accretion or dilution builds from a blank file.
Follow a standard: Adopt FAST-like structure for auditability and speed.
Mimic constraints: Windows, no macros, built-in functions only – get comfortable with constraints.
Rehearse communication: Write a 10-line summary after each practice test.
Build a skeleton: Inputs, three-statement core, debt schedule, and sensitivity grid you can rebuild on site.
Benchmark: Validate speed and quality with independent timed exams.

London vs. other markets – IFRS items to watch

Assessment centers pack Excel tasks, group exercises, and interviews into one day. Time-boxed Excel builds are more common in London than in the US, where paper LBOs dominate. Expect IFRS context: watch EPS definitions, lease accounting under IFRS 16, and exceptional items. Private credit has a larger footprint, so lender-style underwriting shows up often. As an extra check, capture IFRS 16 lease interest and depreciation flows cleanly and call them out in your summary.

Data handling and closeout

Firms must handle candidate data under UK GDPR with clear notices on processing, retention, and any automated scoring. Tests are proprietary, so files or notes do not leave the room. Remote tests are proctored with screen monitoring and timestamps; unapproved tools lead to disqualification and HR flags. Closeout pattern: archive artifacts – index, versions, Q&A, users, audit logs – hash the archive, apply retention, obtain vendor deletion with a destruction certificate, and honor legal holds over deletion.

Reviewer mindset – the 5-minute triage

Reviewers have minutes, not hours. They open inputs, scan checks, glance at outputs, and wiggle a couple of drivers. If ties pass and sensitivities move the right numbers by sensible amounts, you are on to the conversation. If checks break or outputs swing wildly, the review ends. As a fresh angle, assume the reviewer will do exactly three actions: hit F9, alter revenue growth by 100 bps, and change capex by 50 bps. If your cash, leverage, and EPS respond predictably, you pass the sniff test.

Assumptions when inputs are thin

Base rates: Use the policy rate or a round proxy and apply floors where loans specify them.
Taxes: Use the headline rate unless NOLs or special items are indicated.
Capex: Percent of revenue within historical ranges or maintenance near D&A for industrials.
Exit multiples: Match entry unless there is a clear case for change.
Synergies: Quantify only with clear drivers and include costs to achieve.

What not to do

No visuals early: Do not add charts before core ties pass.
No debates: Do not argue with the brief. State your interpretation and proceed.
No hidden assumptions: Keep all drivers on the inputs sheet.
No naked numbers: Always pair outputs with drivers and a sensitivity.
No ungrounded downside: Tie harsh scenarios to explicit levers.

High-impact habits

First principles: Sources and uses reconcile, cash and debt stay non-negative, and taxes do not exceed pre-tax income.
Driver-based modeling: Avoid unnecessary line-by-line growth rates.
Readable formulas: Dodge long nested IFs and prefer helper rows.
Targeted comments: Comment at the module level when logic needs a cue.
Pre-link sensitivity: A two-way table on one or two core drivers pays off in review.
Save early: Old PCs and IT hiccups are real – build resilience into your workflow.

If a late error appears

Fix core tie-outs immediately. If a secondary feature misbehaves, freeze the core, disclose the limitation, and proceed. In the read-out, lead with what works, then explain the limitation and the likely directional impact. This preserves credibility while keeping the discussion moving.

Quick hand-in checklist

Core ties: Statements tie and sources equal uses.
Inputs: Centralized, labeled, and no buried constants.
Checks: Visible and passing.
Sensitivity: Table runs and changes the right outputs.
Summary: Assumptions, drivers, outputs, and risks in one concise section.
File hygiene: Correct naming, versioning, and optional one-page print area.

Why this matters for practitioners

For associates and VPs, these tests predict day-one performance. Teams need bankers who can build client-ready files fast, defend the logic, and pivot when facts change. The same skills drive tight Q&A, clean diligence bridges, and crisp committee materials. In the long run, disciplined structure and plain-spoken conclusions save time, avoid rework, and keep you in the room where decisions get made. If you want a quick refresher on credit-focused metrics to add to your toolkit, skim these debt financing metrics.

Conclusion

London modeling tests reward clean structure, conservative judgment, and crisp communication. Build the spine first, prove the ties, show a simple sensitivity, and summarize like you are sending the file to a client. Simple, correct, and useful beats fancy, fragile, and late.