How to Diversify Around a Core Stock Without Selling It
August 3, 2026
Altsets
Research by Altsets Research
Keep a high-conviction position while choosing additional holdings that reduce repeated customer, supplier, and network dependencies instead of relying only on sector or return correlation.
Data used:Altsets Supply Chain Intelligence: 90k+ entities, 400k+ relationships, 20+ years of history.
Key findings
- A new holding can have a different ticker and still repeat the core stock's economic dependencies, so candidate selection should compare customer, supplier, and second-order overlap before calling the position diversifying.
- The mapped KLA-Micron relationship shows how adding a stock directly connected to a core holding can be a valid investment while still providing a different kind of diversification than an economically separate candidate.
Diversification does not always require selling the stock you have the highest conviction in. Supply-chain data can help an investor keep a core position while choosing additional holdings that do not repeat the same customer, supplier, or network dependencies.
That is a different problem from asking whether the existing portfolio is diversified. The decision here starts with one stock the investor intends to keep and asks what the rest of the portfolio should look like around it.
The core stock creates a dependency footprint
Every company sits inside a network of customers, suppliers, products, and second-order relationships. Once a stock becomes a core holding, those relationships become part of the portfolio's existing economic exposure.
Micron illustrates the idea. The supplied network contains a KLA to Micron relationship associated with 4.59% of KLA revenue. An investor who already owns Micron and then adds KLA is not simply adding a different ticker. The new position introduces another stock with a direct economic relationship to the core holding.
That can still be a good investment. It is simply not the same kind of diversification as adding a company whose business sits on a different economic path.
A complementary stock should change the portfolio's dependency map
The safest candidate is not automatically the company with the lowest historical correlation to the core stock. Historical returns can change when a common economic shock appears.
A dependency-aware complement can instead be evaluated by asking whether it shares the core stock's major customers, suppliers, geography, product cycle, or second-order bottlenecks. The fewer important paths that overlap, the more genuinely different the new position may be.
Diversification can happen inside the same broad theme
An investor can remain bullish on semiconductors, AI infrastructure, electric vehicles, or another theme while spreading exposure across different economic roles. One position can depend mainly on customer demand, another on equipment spending, another on infrastructure deployment, and another on a different customer base.
That is more precise than simply buying several stocks with the same thematic label. A theme ETF can contain many names while still concentrating on the same underlying demand cycle or supplier bottleneck.
The relationship direction changes what counts as overlap
A candidate that sells to the core holding creates one kind of connection. A candidate that buys from the same supplier creates another. A candidate that shares the same end customer creates a third.
Those paths should not be blended into one universal dependency score. The investor should know whether the overlap comes from common demand, common supply, or a direct commercial relationship between the holdings because each can fail in a different way.
The goal is not zero overlap
A perfectly disconnected portfolio is unrealistic and may sacrifice the investor's strongest ideas. The objective is to understand which overlaps are intentional and which are accidental.
A high-conviction semiconductor investor may willingly accept some shared AI demand. The safer version of that portfolio can still avoid stacking every position around the same customer, the same manufacturing node, and the same equipment cycle simultaneously.
Position size can compensate when the best candidate overlaps
Sometimes the best expected-return opportunity is economically connected to the core holding. Dependency awareness does not require rejecting it.
Instead, the investor can reduce the candidate weight, offset the exposure elsewhere, or set a portfolio-level limit on one shared customer or supplier. The custom-index risk-limit guide shows how relationship constraints can coexist with an otherwise conventional portfolio construction process.
A core-stock strategy can become more deliberate over time
As relationships change, the set of good complements can change too. A stock that once diversified the core position can later become more connected through a new customer, supplier, product cycle, or acquisition.
Point-in-time relationship data lets the investor review the portfolio using the network that exists now instead of assuming the dependency structure is permanent. That makes diversification an ongoing research process rather than a one-time sector-allocation decision.
This is where an investing agent can be useful without taking over the portfolio
An agent can compare candidate stocks against the core holding's dependency footprint before the investor adds them. The task is narrow: identify shared customers, shared suppliers, direct relationships, and important second-order paths, then explain where the candidate adds or repeats exposure.
The agent does not need to choose the stock. It can make the tradeoff visible so valuation, expected return, tax considerations, and investor conviction can remain separate decisions.
The portfolio concentration guide focuses on diagnosing hidden concentration in an existing portfolio. This article solves the forward-looking version of the problem: choosing what to own next while preserving a core position.
For broader dependency-aware diversification, read the supply-chain diversification guide. The Altsets methodology explains the relationship metrics and evidence limits used in the analysis.