# Where Can I Find Good Stock Ideas?

> A neutral map of where stock ideas come from: analysts, fund filings, insiders, screens, your own expertise and the news. For each, what it gives you, how late it arrives and what the research says.

Source: https://www.exent.ai/resources/where-to-find-stock-ideas · Reviewed by David, Senior Analyst, Exent · Published Oct 29, 2026 · Updated Oct 29, 2026

## Key takeaways

- Good sources share two traits: a measurable track record and little delay. Analysts ranked by past accuracy, insider purchases and well-built screens qualify; tips, headlines and social media mostly do not.
- The average analyst is close to a coin flip: 53.0% of price targets were reached in Exent’s 7-year backtest of 988,127 ratings. Ranking analysts by their own records is what makes ratings useful.
- Fund holdings (13F) arrive up to 45 days after each quarter ends; insider trades (Form 4) within two business days. Research found insider purchases more informative than sales.
- Every source only produces candidates. An idea still needs a thesis, a valuation check and a portfolio-fit check before it is a decision.

## What are the main sources of stock ideas?

There are six common sources. They differ in how much evidence backs them and how late the information reaches you.

| Source | What you get | Delay | What the evidence says |
| --- | --- | --- | --- |
| Wall Street analysts | Ratings, price targets, research | Same day for public ratings | Average call near a coin flip; records vary widely by analyst |
| 13F fund filings | Holdings of large managers | Up to 45 days after quarter end | Managers’ highest-conviction holdings did better than the rest |
| Insider trades (Form 4) | Executives buying or selling | Within two business days | Purchases informative, mostly in smaller firms; sales not |
| Screens | Stocks matching rules you set | Real time | Factor effects exist but shrank after publication |
| Your own expertise | Industries you know from work or life | None | No direct study; helps judge an idea, not find a cheap one |
| News, social media, newsletters | Whatever is getting attention | Usually priced in | Individuals tend to buy attention-grabbing stocks |

## Are Wall Street analysts a good source of stock ideas?

Some are, most are average. In Exent’s 7-year backtest of 988,127 ratings by the 1,191 ranked analysts (statistics as ofNov 16, 2026), the average analyst reached 53.0% of price targets. Analysts with a 5-star rating in that scoring reached a median 59.0% of their targets, against 53.9% for 4-star analysts. Stars are computed from those same calls, so this shows that records differ, not that they will persist.

Academic work points the same way. Using consensus ratings from 1986–96, [Barber, Lehavy, McNichols and Trueman (2001)](https://ideas.repec.org/a/bla/jfinan/v56y2001i2p531-563.html) found that buying the best-rated stocks and shorting the worst-rated earned more than 4% a year in abnormal gross returns, but only with daily rebalancing; after trading costs the net return was not reliably above zero. Ratings carry information; acting on all of them is expensive. More in [analyst accuracy](https://www.exent.ai/resources/analyst-accuracy) and [how to follow Wall Street analysts](https://www.exent.ai/resources/how-to-follow-wall-street-analysts).

## Should I copy what big funds are buying?

Use fund filings to generate ideas, not to copy trades. Managers with $100 million or more in US-listed stocks file Form 13F, due [45 days after each quarter ends](https://www.sec.gov/rules-regulations/staff-guidance/division-investment-management-frequently-asked-questions/frequently-asked-questions-about-form-13f), according to the SEC. By the time you see a position, it may already be sold. A 13F also shows only long US equity positions, not shorts, hedges or the reasoning.

The most useful signal in the research is conviction. The [“Best Ideas” studies by Cohen, Polk and Silli](https://eprints.lse.ac.uk/24471) found that the stock each fund manager overweighted most did better than the market and than the manager’s other holdings, while most of the other holdings showed no significant outperformance. See [13F filings](https://www.exent.ai/learn/13f-filing) and [copying 13F portfolios](https://www.exent.ai/resources/copying-13f-strategy).

## Is insider buying a good signal?

Purchases more than sales. Officers, directors and 10% owners must report trades on Form 4 within [two business days](https://www.investor.gov/additional-resources/news-alerts/alerts-bulletins/investor-bulletin-insider-transactions-forms-3-4-5). Studying all US-listed companies from 1975 to 1995, [Lakonishok and Lee (2001)](https://ideas.repec.org/a/oup/rfinst/v14y2001i1p79-111.html) found that insider purchases predicted returns, mainly in smaller firms, while insider selling had no predictive ability. Insiders sell for many reasons (taxes, diversification, a house); they buy for one. More in [insider buying](https://www.exent.ai/learn/insider-buying) and [Form 4](https://www.exent.ai/learn/form-4).

## Do stock screens find good ideas?

Screens are good at narrowing thousands of stocks to a list worth reading, and some of their building blocks have long evidence behind them. [Jegadeesh and Titman (1993)](https://ideas.repec.org/a/bla/jfinan/v48y1993i1p65-91.html) found that buying recent winners and selling recent losers earned significant returns over three to twelve months. But published effects weaken once investors know about them: [McLean and Pontiff (2016)](https://doi.org/10.1111/jofi.12365) found the returns of 97 published predictors were 58% lower after publication. A screen is a filter, not a verdict. Our guide to [stock screener settings](https://www.exent.ai/resources/stock-screener-settings) explains each filter.

## What about news, social media and newsletters?

They are the easiest source and the weakest. Studying brokerage records, [Barber and Odean (2008)](https://ideas.repec.org/a/oup/rfinst/v21y2008i2p785-818.html) found individual investors were net buyers of attention-grabbing stocks: those in the news, with unusually heavy trading or extreme one-day moves. By the time a stock is everywhere, the news is usually in the price. Newsletters vary; check whether a service publishes a full, audited record including its losers. See do newsletters beat the market.

## How do I turn a list of ideas into decisions?

Treat every source as a funnel into the same process. Base rates are against single stocks: in Exent’s data, 36.5% of today’s largest US stocks beat the S&P 500 over the 10 years to Sep 30, 2026.

- Keep a watchlist with the source of each idea, so you can later see which sources worked for you.
- Write a one-sentence thesis and what would prove it wrong.
- Check the business, the valuation and how the stock behaved in past crashes.
- Check portfolio fit: correlation with what you own, the new concentration, and the crash exposure it adds.
- Size the position so its worst historical fall is a loss you could hold through.

The next steps are in [how to evaluate a stock idea](https://www.exent.ai/resources/evaluate-a-stock-idea) and [does this stock fit my portfolio](https://www.exent.ai/resources/does-this-stock-fit-my-portfolio).

## How to source ideas with Exent

1. Gems are stocks where several top‑rated analysts (4★ and above in Exent’s scoring) agree: our hunt for the next big winners. They are research, not personal advice.
2. Insider, congress and 13F “smart money” views and a screener with plain‑English filters cover the other sources on this page.
3. Before acting, add a candidate in the Portfolio Analysis what‑if simulator to see how it changes correlation clusters, concentration and crash exposure.

Exent is our product: [Exent Gems](https://www.exent.ai/radar)

## Frequently asked questions

### Where do professional investors get stock ideas?

From their own industry research, company meetings, screens, sell-side analyst reports and other managers’ filings. Most professionals generate many candidates and reject the majority after deeper work.

### Is following insider buying a good strategy?

Research on US stocks from 1975 to 1995 found insider purchases predicted returns, mainly in smaller companies, while insider sales did not. It is a useful signal to investigate, not a reason to buy on its own.

### How late is 13F data?

Form 13F is due within 45 days after each calendar quarter ends, so a position can be up to about four and a half months old when you see it, and it may already have been sold.

### Are free stock tips on social media reliable?

Generally no. They rarely come with a verifiable track record, and research shows individual investors tend to buy stocks that are already getting attention, when much of the news is in the price.

### What is the best source of stock ideas?

There is no single best source. The most useful ones have a measurable record and little delay, such as analysts ranked by their own past accuracy and insider purchases, and every idea still needs its own evaluation.

---

Educational content, not investment advice. Past performance does not guarantee future results.
