ETFs vs Individual Stocks: A Framework for Tech Investors Who Want Both

The honest answer on ETFs versus individual stocks is that most people should own more index funds than they think. Not because stock picking is impossible, but because the conditions where it actually works are rarer than the finance industry wants you to believe. The data on active management is pretty damning, and ignoring it is a mistake I see technically sophisticated investors make all the time. They understand the products. They follow the companies. And they consistently overestimate how much that translates into investment edge.

That said, there are real exceptions. If you work in tech and understand the actual dynamics of specific markets, you're better positioned than most to find them. The framework I use: start with the data, be honest about your edge, and structure any active bets so that being wrong doesn't hurt you too badly.

The Data on Active vs Passive: Know What You Are Up Against

The SPIVA scorecard tracks active fund performance against benchmarks twice a year. The results are consistent across time periods and geographies: over 15 years, roughly 90% of actively managed US large-cap funds underperform the S&P 500 after fees. Tech-focused funds do similarly poorly despite running more concentrated and higher-conviction books.

Two distortions make the data look even worse than headline numbers suggest. Survivorship bias: the funds you can analyze today are the ones that didn't close. Add back liquidated funds and underperformance gets more severe. Selection bias hits individual investors differently: we remember our winners and call them skill, then attribute the losers to bad luck or timing.

None of this argues for paralysis. Professional managers underperform primarily because of fees and transaction costs, not because markets are perfectly efficient. Individual investors with genuine information edges in specific sectors do generate alpha periodically. The key word is genuine. Being a heavy technology user doesn't make you a better investor in tech stocks. Having worked inside enterprise SaaS sales for five years might.

When Individual Stock Picking Makes Sense

There are specific conditions where individual stock selection is analytically defensible. Most investors satisfy zero of them and pick stocks anyway.

Genuine Information Edge

You've worked in a specific corner of tech, enterprise SaaS sales, semiconductor design, cloud infrastructure, and you understand customer buying behavior, competitive dynamics, or technical barriers better than a generalist analyst covering 50 companies across five sectors. A former enterprise software salesperson who knows why procurement teams are switching from Vendor A to Vendor B before it shows up in earnings reports has information that isn't priced in. That's a real edge. Liking a company's products is not.

Special Situations

Spinoffs, carve-outs, post-bankruptcy equities. Institutional mandates create structural selling pressure that's unrelated to valuation. When a tech conglomerate spins off a subsidiary, index funds and institutional holders who can't hold small-caps must sell. This creates temporary mispricing that patient individual investors can exploit. Joel Greenblatt documented consistent alpha in spinoffs; the mechanism still works because the structural sellers aren't discretionary.

Concentrated Thesis with Asymmetric Upside

You have a specific, high-confidence view on a single company: a regulatory decision, a product launch, a customer win the market hasn't credited. A concentrated position makes sense here, but as a small percentage of total portfolio. Not 20%. Two to five percent.

ETF Taxonomy for Tech Investors

Not all ETFs are equal. Understanding what you actually own matters more than most people think.

Broad market index ETFs like VTI, VOO, and SPY give you diversified exposure at near-zero cost. VTI's expense ratio is 0.03%. The S&P 500 is already over 30% tech by market cap, which means a "passive" investor already has meaningful tech exposure without owning a single sector fund.

Sector ETFs concentrate that exposure. QQQ is over 50% technology by weight, with heavy concentration in Apple, Microsoft, Nvidia, Alphabet, and Meta. SMH and SOXX concentrate further into semiconductors with different constituent weights. These are appropriate for investors with a specific view on semiconductor cycle positioning, not as core holdings.

Thematic ETFs are mostly bad. The performance record of funds like ARKK and BOTZ is poor: fees run 0.5 to 0.75%, turnover is high, and thematic narratives tend to peak exactly when inflows are highest, meaning the average dollar buys near the top of the narrative cycle. ARKK's 2021 run and subsequent 80% drawdown is the canonical example. The exception: a few low-fee thematic ETFs with mechanically defined indices and real underlying demand, not narrative-driven product launches. If you want sector exposure, individual stocks or low-fee sector ETFs beat narrative-driven thematic funds almost every time.

Building a Core-Satellite Portfolio

The core-satellite structure is the most practical framework for investors who want index-level diversification as a foundation with room to express specific views through individual positions.

The core, 70 to 80% of the portfolio, is low-cost broad market and international index ETFs. Rebalance annually or when allocation drift exceeds 5%. Expense ratio on the core should average under 0.10%.

The satellite, 20 to 30% of the portfolio, is individual stocks or higher-conviction sector positions. Each satellite needs a written thesis, defined conditions under which you'd sell, and a maximum position size. Cap any single satellite at 5% of total portfolio. That's the upper bound, not the target.

Rebalance mechanically. When satellite positions grow past 30% of total portfolio value due to appreciation, trim back toward target and add to the core. When they shrink below 15%, either add to existing convictions or find new ones. This prevents the common pattern of letting winners run until they dominate the portfolio without a thesis refresh.

Tax Efficiency: ETFs, Direct Indexing, and the Capital Gains Question

ETFs have a structural tax efficiency advantage over mutual funds. The in-kind creation and redemption mechanism lets them avoid distributing capital gains in most cases. Mutual funds must distribute gains annually, creating taxable events even for holders who didn't sell. In taxable accounts, this ETF advantage is worth roughly 0.5 to 1.5% annually in deferred taxes, which compounds.

Individual stocks in a taxable account have the highest potential tax efficiency: you control timing of gains realization entirely and can harvest losses against other gains at will. The wash sale rule means you wait 30 days before repurchasing a substantially identical security, but for individual stocks that's easy to work around (sell MSFT, buy GOOGL to maintain tech exposure while harvesting the loss). Tax-loss harvesting in a volatile satellite portfolio can generate 1 to 2% in annual tax alpha in down years.

Direct indexing, owning the individual constituents of an index rather than the fund itself, is now accessible to retail investors through platforms like Fidelity Managed FidFolios, Schwab Personalized Indexing, and Wealthfront. This gives you index-level diversification with individual-stock-level tax-loss harvesting. For taxable accounts above $250,000, direct indexing often beats equivalent ETF positions on an after-tax basis. Below that, transaction costs tip the balance back toward ETFs.

Evaluating Individual Tech Stocks: A Rigorous Framework

The questions that matter for tech stock evaluation are different from consumer staples or industrials. Here's what practitioners who've worked in both operating and investing roles actually look at.

• TAM and penetration: Is the total addressable market large enough to support the implied growth trajectory? What's realistic penetration in 5 years? Be skeptical of TAM slides that include every adjacent market the product has never touched.
• Competitive moat: What specifically prevents a well-funded competitor from replicating this product? Network effects, high switching costs, scale economies, and proprietary data are durable moats. "Better product" is not a moat.
• Unit economics: What's the contribution margin per customer? What's the payback period on customer acquisition cost? A SaaS company with 80% gross margins and an 18-month CAC payback has fundamentally different economics than one with 50% gross margins and a 36-month payback.
• Management quality signals: Track record of capital allocation (acquisitions at reasonable prices vs. trophy deals), insider ownership (founders with meaningful economic stakes behave differently from professional managers), and communication quality in earnings calls (specificity vs. vagueness on metrics and competitive dynamics).
• Valuation discipline: A great business at a terrible price is a mediocre investment. Price to sales and EV/EBITDA multiples must be contextualized against growth rates. Use the Rule of 40 as a first-pass quality screen.

Behavioral Traps That Destroy Tech Portfolios

The behavioral risks in tech investing are distinct because tech culture itself reinforces certain cognitive errors. FOMO is the biggest portfolio killer, and it hits especially hard in tech.

FOMO-driven buying is the primary value destroyer. Nvidia in late 2023 was a great company with a genuine AI infrastructure thesis. Nvidia in early 2024, up 200% from the prior year and priced for decades of perfect execution, was a different risk/reward proposition. The product thesis didn't change. The price did. Buying after 200% appreciation because you fear missing further gains isn't investing. It's momentum speculation, and momentum reverses.

Confirmation bias in tech is particularly acute because the sector produces compelling narratives. If you own a cloud security company and read every analyst report from its management team and bull-case analysts, you'll convince yourself the thesis is stronger than ever even as competitive dynamics deteriorate. Deliberately seek out the highest-quality bear case for every tech position you hold and engage with it seriously.

Position Sizing and When to Sell

Position sizing should reflect conviction and volatility, not emotional attachment. A simplified Kelly criterion approach: size a position as edge divided by odds, where edge is your estimated alpha and odds reflects volatility. In practice, full Kelly is too aggressive. Most practitioners use half-Kelly as a maximum. Cap individual positions at 5% of total portfolio and require a written thesis for anything above 3%.

Selling is harder than buying and the subject of more poor decisions. The correct sell triggers are: the original thesis is broken (competitive dynamics you identified haven't materialized, or a new competitor invalidates your moat assessment); the position has grown to represent too large a percentage of portfolio due to appreciation (mechanical trim regardless of thesis strength); you've identified a clearly superior opportunity and this is your lowest-conviction holding. Selling because a stock has declined is not a valid trigger unless the thesis is also broken. Selling because a stock has risen is not a valid trigger unless valuation has become disconnected from any realistic scenario.

The core-satellite framework succeeds not because it finds extraordinary investments, but because it prevents extraordinary mistakes like the 10% position in a narrative stock that falls 80% and takes years to recover from.

The most successful tech investors aren't the ones who identify every winner. They're the ones who limit concentration to positions they understand most deeply, maintain the core as genuine insurance against their own overconfidence, and apply the same analytical discipline to selling that they apply to buying. The framework here isn't exciting. But it's durable, and durability compounds.