Idea generation is treated as inspiration and it is a process. Professional investors do not wait to notice something; they run a small number of repeatable funnels that produce candidates, then apply a fast filter that kills most of them. Building your own version of that, even as a student, changes what you can say in an interview from an opinion into a practice.
The question every idea has to answer
Before any of the sources below are useful, be clear about what you are looking for. An idea is not a good company. An idea is a company where you believe something the market does not, and where something will happen to prove you right.
That means every candidate needs three things: a view, a reason the view is not already in the price, and a reason it will become visible. Sources of ideas are really sources of situations where those three can coexist, which is why they cluster around neglect, complexity and forced selling.
The funnels professionals actually use
| Source | Why it works | Typical find |
|---|---|---|
| Quantitative screens | Cheap, systematic, covers the whole market | Statistically cheap names needing a reason |
| Spin-offs and demergers | Forced selling by holders who cannot own the new entity | Neglected mid-caps with no coverage |
| Insider buying | Purchases by executives with their own money | Small and mid-caps at inflection |
| Post-bankruptcy equity | New shares nobody follows, held by former creditors | Deeply uncovered situations |
| 52-week low and high lists | Sentiment extremes in both directions | Value traps and momentum, both worth checking |
| Disclosed fund holdings | What good investors already own, with a delay | Starting points, never conclusions |
| Industry primary sources | Trade press, conferences, regulatory filings, job postings | Changes before they reach earnings |
| Competitor and supplier commentary | A company describing someone else market | Read-across nobody has modelled |
| Index changes and forced flows | Mechanical buying and selling for non-fundamental reasons | Short-lived dislocations |
| Complexity and conglomerates | Sum-of-the-parts hidden by disclosure | Structural mispricing |
Screens, done well
A screen is a way of shortening the list, never a reason to buy. The two failure modes are screening on price alone, which produces a list of companies that deserve to be cheap, and screening on so many criteria that nothing survives.
Better screens combine a valuation measure with a quality measure and a change measure. Low enterprise value to EBIT with a return on capital above the cost of capital and an improving margin trend produces a far more interesting list than any of the three alone. Then read the list rather than sorting it: the point of the screen was to buy yourself the time to read thirty annual reports instead of three thousand.
Why spin-offs and forced selling keep working
When a company separates a division, the shares in the new entity land in portfolios that never chose them. Index funds tracking a benchmark the spin-off is not in must sell. Large funds receive a position too small to matter and sell it for that reason alone. Analysts covering the parent do not pick up the child.
The result is a real company with a functioning business and a shareholder register full of people who want out, and no research coverage to tell anyone otherwise. The same logic explains post-bankruptcy equity, deletions from an index and small stakes distributed in a demerger. None of this is a guarantee. It is a reliable way of finding situations where the price is being set by something other than an opinion about value.
The fast filter
A funnel is only useful with a kill step. Spend twenty minutes per candidate before deciding whether it earns twenty hours.
- Can I explain how this business makes money in two sentences? If not, move on unless complexity is the thesis.
- Is there an identifiable reason it is mispriced? Neglect, complexity, a temporary problem, forced selling, or a genuine difference of opinion. If the answer is that the market has not noticed, be suspicious.
- Does the balance sheet let me be wrong for a while? Leverage shortens your permitted time horizon whatever your conviction.
- Is there a catalyst or a clock? A cheap stock with no reason to re-rate can stay cheap for a decade.
- Can I get an edge here? Sector knowledge, patience, a willingness to hold something ugly. If there is no edge, the price is probably right.
Building a process you can describe in an interview
Interviewers ask where your ideas come from, and the question is really about whether you have a repeatable method or a lucky anecdote. Have an answer with a structure and a number in it.
Something like: a monthly screen on two sectors you follow, a standing list of spin-offs and index deletions, and a habit of reading the trade press for one industry. That produces a handful of candidates a month, of which one or two get real work. That answer tells a fund you will still be finding ideas in year three.
A note on other people ideas
Disclosed holdings, conference presentations and published letters are legitimate starting points and terrible finishing points. The data is delayed, the position may already be closed, and you will not know why it was bought.
Use them the way a journalist uses a tip: as a reason to look, never as evidence. And never pitch a widely known position without saying so and explaining what you think everyone else has wrong, because the person interviewing you almost certainly holds a view on it already.
Frequently asked questions
How do professional investors find stock ideas?
Through a few repeatable funnels rather than inspiration: quantitative screens, spin-offs and forced selling, insider buying, post-bankruptcy equity, industry primary sources and complex structures with hidden value. Each produces candidates, and a fast filter kills most of them within twenty minutes.
Why are spin-offs a good source of ideas?
Because the new shares land with holders who did not choose them. Index funds must sell, large funds receive positions too small to matter, and no analyst covers the new entity. The price gets set by people who want out rather than by an opinion about value.
What makes a good screen?
Combining a valuation measure with a quality measure and a change measure, rather than screening on cheapness alone. Cheap-only screens return companies that deserve to be cheap. The screen exists to buy you time to read thirty annual reports, not to rank a list.
Can I pitch an idea I found in a fund filing?
As a starting point, yes, and say so. The data is delayed and the position may be closed, so it is a reason to look rather than evidence. Never pitch a well-known position without stating what you think consensus has wrong, since your interviewer probably already holds a view.
Related guides
Want this applied to your recruiting?
Reading is the easy part. For practitioner feedback tailored to your situation, work 1:1 with Suro.