A Spring Week modeling test is a timed Excel exercise banks use to check whether you can build, link, and explain a basic financial model under pressure. A Spring Week is a short insight program in London that feeds summer internships and, eventually, analyst seats; the test is a conversion filter. A three-statement model links the income statement, balance sheet, and cash flow to produce a forecast and a view on value or credit.
The payoff for mastering these tests is simple. You prove you can structure a clean model fast, avoid errors, and communicate what the numbers mean. If you can do that in 60 to 120 minutes, you look like day-one help on a live deal.
When tests happen and what they prove
Banks run these exercises at three points to stress different skills. Online pre-screens track keystroke accuracy and logic. Assessment centers and Spring Week conversions look for correct three-statement links, sensible assumptions, and a defensible valuation or credit view. The best candidates stabilize the core model quickly and leave time to explain the so what.
Common formats and what each tests
- Online Excel drills: Timed, locked environment that measures speed, formula logic, and data cleaning discipline; error rate and consistency are key.
- In-person case: Build or extend a simple three-statement model, then answer valuation or credit prompts in 60 to 120 minutes.
- Paper LBO or credit case: Minimal Excel; focus on leverage, coverage, and returns common in private credit or PE insight days.
- Valuation-only tasks: DCF or trading comps from extracts, still graded on assumptions and sensitivities.
What banks actually want to see
Interviewers are not rewarding associate-level bells and whistles. They want clean structure, correct accounting, reasonable drivers, and a crisp explanation. The exercise measures preparation and composure, not perfection.
Why each stakeholder cares
- Recruiters: Verify Excel fluency and reduce reliance on pedigree; fewer mis-hires saves time and cost.
- Deal teams: Protect file hygiene on the shared drive; eliminate integrity breaches and patchwork models.
- Candidates: Prove applied knowledge in a controlled setting; consistent practice beats raw horsepower.
Mechanics, inputs, and outputs you should expect
These tests enforce focus. Time caps run 60 to 120 minutes. Tools are Excel only, with the internet and often VBA blocked. Inputs typically include two to three years of simplified historicals and a short business description, sometimes with a transaction angle. Required outputs include a three to five year forecast, a simple debt and interest schedule if relevant, one or two sensitivities, and a short commentary that states your view. Hard-code only in input cells, discourage uncontrolled circularity, and ensure all line items reconcile to cash.
In practice, you win marks by stabilizing the core first. Build the three-statement model so it balances and cash ties, then add debt, valuation, and sensitivity work only as the clock permits.
Grading criteria and how to allocate your time
Most teams use a simple rubric. Accuracy and reconciliation carry the most weight. Structure and auditability come next. Pace and judgment matter because build sequence drives stability. Communication closes the loop by turning your numbers into a conclusion.
- Accuracy first: Balance sheet balances, cash ties, and depreciation and working capital flow correctly. Misses here are fatal.
- Structure and auditability: Transparent inputs, consistent signs, modular schedules, and no buried constants are table stakes.
- Pace and judgment: Lock the three statements before building debt, tax, and sensitivities; sequence reduces rework.
- Communication: Lead with drivers and numbers, not prose. Tables, labels, and a clear summary win points.
Use a simple 60-30-10 split as a rule of thumb. Spend 60 percent of the time getting the core three statements stable, 30 percent on debt, valuation, or credit outputs, and 10 percent on sensitivities and formatting that improve readability.
Core technical topics that repeat in cases
Focus preparation on topics that appear in almost every test. These are the bankable moves that carry marks under pressure.
- Three-statement mechanics: Link net income to retained earnings, depreciation to PP&E, and working capital to operating cash flow.
- Working capital: Use DSO, DIO, and DPO; handle negative working capital sensibly in retail or software maintenance.
- PP&E and capex: Build a simple roll-forward; forecast capex as a percent of sales and align depreciation rates to reconcile.
- Debt and interest: Model a term loan and optional revolver; interest on average debt and a cash sweep only if specified.
- Taxes: Use a 25 percent UK main rate unless told otherwise; adjust interest deductibility only if asked.
- Valuation: A straightforward DCF with WACC and terminal value or trading comps using EV/EBITDA or P/E.
- Sensitivities: One or two variable tables on growth, margins, WACC, or exit multiple; keep formats clean.
- Credit metrics: Leverage, interest coverage, fixed charge coverage, and DSCR with a view on covenant headroom.
- Short-form LBO: Sources and uses, entry EV/EBITDA, net debt, target IRR/MOIC, and a simple cash sweep.
When you build cash flows, many tests expect an indirect cash flow statement. Set it up early to surface reconciliation issues fast.
Accounting conventions that must be right
Keep accounting simple but correct unless the case calls out complications. IFRS 16 capitalizes leases, with depreciation and interest recorded, but ignore if the packet says so. IFRS 15 recognizes revenue on transfer of control; use given figures unless told to adjust. Stock-based comp is non-cash, so it reduces net income but adds back to operating cash flow. Non-controlling interests and associates are rare; adjust EV and EBITDA definitions if they appear. Keep deferred taxes basic; apply rate changes prospectively unless provided with a schedule.
What changes by division and why it matters
Coverage and M&A tests balance three-statement mechanics with valuation, sometimes ending with one industry question. Leveraged finance and private credit ask for more debt, covenants, rates, and sweeps, plus downside and liquidity commentary. PE insight days often run a paper LBO or simplified Excel LBO and emphasize the investment case and sensitivity to leverage and multiple. Markets teams rely less on builds and more on data manipulation and P&L attribution.
Skim, plan, and extract the right drivers
Use two to three minutes to confirm the deliverables, time limit, and constraints, then make a minimal checklist. Keep historicals in a separate sheet and reference rather than retype to avoid data entry errors. From the business description, extract two to three drivers and convert them to quantitative levers. If peers, growth, pricing, or loan terms are provided, use them, but do not invent detail beyond the packet.
A build order that fits the clock
Sequence your work to stabilize fast and avoid rebuilding under time pressure. Start with an inputs panel holding growth, margins, capex as a percent of sales, working capital days, tax, and interest assumptions. Then build the income statement skeleton from revenue to net income with clear subtotals. Populate the balance sheet with working capital via days, a PP&E roll-forward, debt balances, and retained earnings. Build cash flow with ties to the above, add capex and financing flows, and reconcile change in cash to the balance sheet. Only after cash ties, add a debt schedule with interest on average balances, amortization, and a sweep if relevant. Then add valuation and sensitivities that earn marks. For deeper reading on debt scheduling, see this guide.
Excel hygiene that reduces errors
Format for accuracy and auditability. Use consistent signs and avoid hidden flips. Keep inputs in one place and highlight them. Avoid volatile functions and arrays under time pressure; rely on SUM, IF, MIN/MAX, and LOOKUP. Use clear time periods and units, and add a top-left model summary. Insert simple error checks so you know when something breaks: balance sheet balances, change in cash ties, and no negative debt unless a revolver allows it. Drill the keystrokes you will actually use with targeted Excel shortcuts.
Pick assumptions that are realistic and defensible
Align revenue growth to the narrative, staying modest without explicit data. Let capex and depreciation move together, usually with capex as a percent of sales and a simple depreciation rate that reconciles. Set working capital days that fit the sector, lighter in services than manufacturing. For UK corporate tax, use a 25 percent base for non-small profits and avoid credits or loss use unless instructed.
Credit-specific traps and LBO pitfalls to avoid
- Revolver and cash minimum: Implement a sweep with a floor; draw only to avoid negative cash and within limits. For mechanics, review a cash sweep primer.
- Floating rates: Use a flat reference plus margin unless a forward curve is provided; show a sensitivity on rates.
- Covenants: Compute exactly as defined; apply adjustments only as specified.
- Amortization vs bullet: Show the schedule and the bullet; keep sweep effects transparent.
- Sources and uses: Reconcile and include fees only if specified, based on gross sources.
- Equity value to EV: Bridge EV using equity value plus net debt and any other instructed adjustments.
- IRR timing: Keep time-weighted cash flows correct; do not switch to quarterly unless asked. For structure, see an LBO modeling framework.
Communicate your answer like a banker
Lead your debrief with three numbers: revenue CAGR, the EBITDA margin path, and year five free cash flow. Then state leverage sustainability or valuation support in one sentence. Explain two or three core assumptions and a single sensitivity grid that changes the answer. Name one risk and one mitigant tied to the case, such as a working capital unwind risk and staggered maturity mitigant. If time permits, include a one-page summary with tidy charts.
Practice plan that mirrors the test
Repeat realistic reps rather than memorize theory. Spend 30 to 45 minutes daily for four weeks on keyboard fluency in a locked environment. Build five three-statement models from scratch and target under 75 minutes with no errors. Complete five valuation-only builds and two short-form LBOs in 60 minutes each, including a five-sentence summary. Run post-mortems with a checklist when the balance sheet does not tie within 10 minutes, then fix build order or formula habits. Finally, print to PDF to check formatting and totals line wraps.
To sharpen edge skills that set you apart, build a simple DCF in under 20 minutes after the core ties and prepare a compact sensitivity table. If you want a refresher on when to use scenario analysis instead of single-variable tables, compare sensitivity vs scenario analysis.
Fresh tactics that improve speed under pressure
- 80-15-5 rule: Spend 80 percent of your time on forecast and reconciliation, 15 percent on the single output that earns the most marks, and 5 percent on formatting that improves readability.
- Triage scratchpad: Keep a top-left box with three checks (BS balances, cash ties, no negative debt) and three highlights (revenue CAGR, EBITDA margin year five, FCF year five) that you update as you go.
- Link once: Use named ranges or a single inputs tab and never hard-code in formulas; this avoids silent breaks during last minute changes. If comp tables appear, reference them instead of retyping to reduce error risk.
- One-chart rule: Add one chart that explains the answer, such as FCF ramp or net leverage decline, rather than formatting five that add little.
Test-day time management
Use the first five minutes to read, list deliverables, pick drivers, and decide what you will cut if time compresses. Use the next 40 to 60 minutes to stabilize the three statements, tie cash, and add error checks. Only then build debt. Spend the final 20 to 30 minutes on valuation or credit metrics, one or two sensitivities, and a crisp summary. Format only what aids reading.
What graders notice immediately
They look for a logical file structure with clear tabs, highlighted inputs, and a cover area with a summary. They expect precise definitions for EBITDA, net debt, and WACC on-sheet. They reward assumptions that do not show big margin jumps without a reason tied to mix or operating leverage. If your file tells this clear story on first open, you are already ahead.
Conclusion
A clean, balanced model with correct links and a pointed commentary beats a feature-rich but unstable file every time. Candidates who pass show judgment under time pressure: they simplify where needed, quantify what matters, and state constraints plainly. That is the analyst a team can trust with a live client deliverable on day one.