What is 13F screening, in one paragraph?
13F screening means searching quarterly SEC filings for stocks that professional money managers are accumulating. Institutional investment managers overseeing more than $100 million in qualifying assets must disclose their long positions in U.S.-listed equities within 45 days of each quarter's end. By comparing one quarter's filing to the next, you can see which managers opened new positions, added to existing ones, trimmed, or sold out, and you can screen the whole dataset for the strongest accumulation patterns. It is evidence of what sophisticated investors did with real money, lagged by up to 45 days and limited to long U.S. equity positions. The lag and the limits are the price of the data; the guide below shows how to work within them.
What a 13F filing actually contains
Form 13F is filed quarterly by institutional investment managers, including hedge funds, mutual fund companies, pension funds, and investment advisers, that exercise discretion over more than $100 million in qualifying securities. Each filing lists the manager's long positions in U.S.-listed stocks and certain other securities: the security name, CUSIP, number of shares, and market value at quarter end. Thousands of managers file every quarter, creating a near-comprehensive map of professional long equity positioning.
The filing deadline is 45 days after quarter end, which creates the dataset's rhythm: early-to-mid February, May, August, and November bring waves of filings covering the prior quarter. StockEyes organizes this data into free per-ticker ownership pages (which managers hold the stock, position sizes, quarterly changes), a manager directory (each manager's reported portfolio), and signed-in Pro screeners: the 13F change screener filters by holder-count changes, share-count changes, and position values, and the cluster-buying screener finds managers accumulating across multiple tickers.
Why this data matters: professional managers research full-time with resources individuals lack, and their disclosed positions represent real capital committed after real diligence. Following the filings does not copy their edge, the best ideas may already be priced in, and you cannot see their full strategy, but it surfaces candidates that passed someone's serious filter, which is a better starting point than a random ticker.
What 13F data hides: the limits you must respect
Filings arrive up to 45 days after quarter end. A manager who bought in January and sold in February shows a March filing with no position; you see the quarter-end snapshot, never the round trip. Treat every 13F idea as "this was true six weeks ago" and check the price action since.
13F covers long positions in qualifying securities. Short positions, put options used as hedges, and pair trades are invisible. A manager "buying" a stock in 13F may be hedging a short elsewhere; the filing shows half the trade.
International stocks, bonds, private investments, and most derivatives are excluded. A global macro fund's 13F is a fragment of its book, not the book.
Managers can request the SEC's permission to omit positions temporarily while building them. The most interesting new positions are the ones most likely to be hidden this way, then revealed in amended filings later.
A filing shows the manager's positions, not which analyst picked them or why. Ten managers buying the same stock may share one thesis or ten different ones; the filing cannot tell you which.
Some managers adjust quarter-end portfolios for appearance, buying recent winners to show clients. Quarter-end snapshots overweight whatever looked good on the last day of the quarter.
None of these limits invalidates 13F research; they define how to use it. The data is a lagged, partial, long-only snapshot of professional positioning. Used as a hypothesis generator with these limits in mind, it is powerful. Used as a real-time signal, it will mislead you.
Reading quarterly changes: new, added, trimmed, sold
The position level tells you who owns a stock; the change tells you what they think now. Quarter-over-quarter comparison classifies every manager-stock pair into four moves, each with a different meaning:
New positions: the strongest single-manager signal
A manager initiating a position did fresh work and committed new capital this quarter. Weight new positions by size relative to the manager's portfolio: a new 5% position is a conviction statement; a new 0.1% position is a watchlist stub. Check whether the initiation followed a price decline (value recognition) or strength (momentum confirmation).
Increased positions: conviction or averaging
Adding to an existing position usually signals continued conviction, but verify against the price: adding after a 30% decline may be averaging down on a broken thesis rather than fresh enthusiasm. Large percentage increases on already-large positions are the most meaningful form of this move.
Trimmed positions: usually noise
Partial sales have many innocent explanations: rebalancing, risk management, raising cash for redemptions, or funding a better idea. A 10% trim of a large position means little; treat trims as neutral unless the pattern breaks (a manager who never sells suddenly halving a core holding).
Sold out: the move that deserves a question
Full liquidation ends the relationship and deserves investigation, but not automatic bearishness: mandates change, funds close strategies, and tax considerations intervene. Ask why, check whether other managers agree (cluster selling), and read the company's recent filings before concluding anything.
Cluster buying: when several managers act at once
A single manager's new position is one informed opinion. Several respected managers opening or adding to the same stock in the same quarter is a pattern: independent research processes converging on the same name with real capital. This is the highest-conviction pattern in 13F screening, and it is what StockEyes' cluster-buying screener is built to find.
- Count independent managers, not total shares. Five managers each adding a 2% position is stronger evidence than one manager adding 10%; independence is the signal, size is secondary.
- Check manager quality, not just quantity. A cluster of managers with strong long-term records means more than a cluster of unknown funds. The manager directory helps you learn whose filings deserve weight.
- Verify the timing context. Cluster buying after a price decline suggests collective value recognition; cluster buying into a run-up may be performance chasing. The 45-day lag means you must reconstruct what the managers saw, using the quarter's price history.
- Rule out index and mandate effects. If a stock joined a major index during the quarter, many managers bought it mechanically. Check for index inclusions, sector reclassifications, or ETF rebalances before interpreting the cluster as active conviction.
Crowding: when everyone already owns it
13F data cuts both ways: it finds accumulation, and it reveals crowding. A stock owned by a large fraction of professional managers, with holder counts at multi-year highs and few new buyers left, is crowded: the marginal buyer is exhausted, and any disappointment triggers a rush for a narrow exit. The most dangerous crowded stocks look the safest, because the ownership itself feels like validation.
Screen for crowding with the same tools you use to find accumulation: extreme holder counts relative to history, decelerating new-position formation despite rising prices, and ownership concentrated in momentum-style managers rather than long-term holders. Crowding does not predict timing, crowded stocks can stay crowded for years, but it changes the risk/reward: the upside of consensus is limited while the downside of consensus breaking is not.
The symmetric insight: the best 13F ideas are often stocks where accumulation is beginning, holder counts are rising from a low base, and the crowd has not arrived yet. Early accumulation plus a reasonable valuation is the pattern worth the most research time.
The 13F screening workflow
- Start with the change screener. On StockEyes' 13F change screener, filter for the largest quarter-over-quarter increases in holder count and shares held, focusing on new positions above a meaningful portfolio weight. This is the raw idea list.
- Look for clusters. Cross-check the top results on the cluster-buying screener: names appearing on both lists, with multiple independent managers accumulating, go to the top of the research queue.
- Adjust for the lag. For each candidate, check what the stock did between quarter end and today. A stock up 40% since the filing quarter is a different proposition than the one managers bought; the edge decays with the price move.
- Confirm with insiders. Check Form 4 activity: managers accumulating while insiders buy with their own money is the strongest combined signal in filing-based research. Managers accumulating while insiders sell aggressively is a divergence to resolve.
- Research the business. 13F data tells you who is buying, never why. Read the 10-K, understand the thesis the managers might hold, and write down what would change your mind. The 10-K lesson is the companion here.
- Check for crowding on the way in. Before acting, verify holder counts are not at historic extremes and the accumulation is early, not late. The same data that found the idea should clear it of crowding.
Run this workflow each earnings season, when the filing waves arrive in February, May, August, and November. Between waves, the dataset is static; there is no edge in re-screening unchanged filings weekly.
Questions about institutional ownership screening
What is a 13F filing?
Form 13F is the quarterly SEC filing in which institutional investment managers with over $100 million in qualifying assets disclose their long positions in U.S.-listed equities. It is due 45 days after each quarter end and lists holdings by security, share count, and market value.
How delayed is 13F data?
Up to 45 days by regulation: filings for the quarter ending March 31 arrive by mid-May. Treat every 13F observation as a snapshot from six weeks ago, and always check what the stock has done since the quarter ended before acting on it.
Does 13F show short positions?
No. 13F covers long positions in qualifying securities only. Shorts, most derivatives, international holdings, and bonds are invisible, so a filing shows part of a manager’s book, never the whole strategy.
What is cluster buying?
Cluster buying is when multiple independent professional managers open or add to positions in the same stock in the same quarter. It is more informative than any single manager’s trade because independent research processes converged on the same name with real capital.
Can I just copy what the best managers buy?
Not directly, and the lag is only the first problem: you cannot see their entry prices, position sizing logic, hedges, or exit timing, and by the time you see the filing the opportunity may be priced in. Use 13F data as a research starting point, then do the business analysis yourself.
Is 13F data free?
The filings themselves are public on the SEC’s EDGAR database. StockEyes organizes them into free per-ticker institutional-ownership pages, with the 13F change screener, cluster-buying screener, and manager detail on the signed-in Pro tier.
StockEyes provides informational research organization and transparent calculations, not personalized investment advice, valuation conclusions, trade instructions, or predictions. Data described here reflects the sources and dates noted in the article; verify current pricing, features, and data coverage on each provider’s site before subscribing. Consult a qualified professional for advice about your own situation.