What a portfolio is actually for

Nobody hiring for an analyst or associate seat reads a portfolio to check whether you can follow a template. They're checking whether you can take a real, messy situation — an actual company, actual numbers, actual uncertainty — and produce something that makes a clear, defensible call. A portfolio that's seven perfectly-formatted models with no opinion in any of them is worse than three rougher ones that each land on something specific: "this deal only works if the exit multiple expands," "this company's working capital is quietly eating its cash," "here's the one number in this deck I'd push back on."

The project types below are the standard set for a reason — each one demonstrates a distinct skill, and together they cover most of what an actual analyst does in the first two years of the job. Pick real, named companies (public ones, so the numbers are checkable) over invented ones; a reviewer trusts a model of Tesco more than a model of "TechCo."

1. An LBO model — private equity

The single most PE-specific project you can build. It proves you understand that a leveraged buyout isn't a valuation exercise — it's a returns exercise: how much you pay, how much debt you use, and what you sell for are the only three things that actually matter, and everything else is detail. A strong writeup doesn't stop at "the IRR is 22%" — it says why, and shows the sensitivity to entry multiple, leverage, and exit multiple that a real IC discussion would actually have.

Build one with the LBO Model Calculator → — entry price, debt/equity structure, a cash-sweep paydown schedule, MOIC/IRR, and the full entry × exit sensitivity grid, computed live.

2. A 3-statement model — financial modelling fundamentals

This is the project that proves you understand how a company's financials actually fit together, not just that you can build a discounted-cash-flow shortcut. The tell that separates a real 3-statement model from a template someone half-understood: does the balance sheet actually balance in every projected year, and can you explain why it balances — not just that a formula makes it balance.

Build one with the 3-Statement Model Calculator → — a linked income statement, balance sheet, and cash flow statement, with working capital driven by real days-based ratios and a live balance check.

3. DCF and comps valuation — equity research

The two standard valuation methods, and the point of doing both on the same company is the comparison: when a DCF and a comps-based valuation disagree by a lot, that disagreement is the actual finding — it usually means the market is pricing in something (growth, risk, a catalyst) that your DCF's assumptions don't capture, or vice versa. Pick two comparable public companies in the same sector (the classic pairing — one "quality compounder," one "cheaper but riskier" — makes for the most interesting writeup).

DCF Valuation Model → and Relative Valuation / Comps Calculator → — plus the WACC calculator → for a properly built discount rate instead of a guessed one.

4. An investment memo — investment research

The format every buy-side seat actually uses to make a decision: not a 40-slide deck, a tight, structured writeup that states the thesis, the risks, and the open questions. The skill being tested isn't writing — it's compression: can you take everything you know about a company and cut it down to what actually matters for the decision.

AI Investment Memo Generator → — turns a company's basic details into a structured memo skeleton (opportunity, market thesis, key risks, diligence questions) you then sharpen with your own view. If you're evaluating a live idea on StakeCrew, the same tool is built into every pitch page.

5. Portfolio construction & quant work — asset management

A less common but genuinely differentiated addition if you're recruiting anywhere near portfolio management: shows you understand risk and diversification quantitatively, not just as a slide-deck concept. An efficient frontier or a backtested allocation with a real max-drawdown number says more than another single-company writeup.

Efficient Frontier Optimizer →, Portfolio Backtest Calculator →, Monte Carlo Portfolio Simulator →, and the asset correlation matrix →.

6. Commercial / data analytics — strategy & operations

SQL and a BI tool (Power BI, Tableau) turning a real or realistic transaction dataset into a management dashboard — this one's less about finance theory and more about proving you can work with messy real-world data and turn it into something a non-technical stakeholder would actually use. There's no StakeCrew tool for this one; the honest advice is to find a genuinely public dataset (Kaggle's e-commerce and retail sets are the standard starting point) rather than fabricate one, since a reviewer can usually tell.

7. An audit or financial-reporting review — accounting

The least common addition, and the one that stands out precisely because of that: picking a public company's actual annual report and identifying where a real audit team would spend its time (revenue recognition judgment calls, goodwill impairment risk, related-party transactions) shows a level of financial-statement literacy most candidates never demonstrate. No shortcut here either — this one is close reading of a real 10-K or annual report, which is the actual skill being tested.

What separates a portfolio that lands interviews from one that doesn't

Every tool linked above is free, requires no signup to try, and computes live in your browser — nothing you enter leaves the page. See the full set on the Tools page.