Warren Buffett is one of the most successful investors in history, with a 60-year track record of building one of the greatest investing records. He achieved this not by chasing whatever was popular, but by finding businesses he never wanted to sell. In his 1988 shareholder letter, Buffett wrote, ‘Our favorite holding period is forever.’ This line was not just about one stock, but a philosophy that shaped every major decision Berkshire Hathaway made under his leadership.

Apple Stock and the Ecosystem

Apple is Berkshire’s largest holding, accounting for roughly 22% of the portfolio as of the first quarter of 2026. Buffett first bought the stock in 2016, which was unusual for him, given his historical reluctance to invest in technology companies. He has since called it one of the best businesses he has ever seen.

Ask most iPhone users why they won’t switch to Android, and the answer isn’t really about the phone. It’s about everything else. Three years of photos in iCloud. Apps they paid for. A watch that only fully works with an iPhone. Apple Music. Apple Pay habits. At some point, the cost of leaving gets high enough that most people stop thinking about it. That’s when Apple starts making real money, through services that cost almost nothing to deliver and carry margins hardware could never touch.

Apple’s installed base now exceeds 2.5 billion active devices globally. That is the foundation on which every new product and service launch sits. The company does not need to win new customers to grow. It needs to keep selling more to the ones it already has, and its customers tend to be among the more affluent and brand-loyal in the consumer market.

Coca-Cola Stock and the Power of Compounding

Buffett bought Coca-Cola in 1988 and has never sold a share. That decision has produced one of the most remarkable compounding outcomes in investing history.

Berkshire completed its purchase of 400 million shares by 1994, spending roughly $1.3 billion in total. Today, those shares generate approximately $848 million in annual dividends, and the effective yield on Berkshire’s original cost basis is now roughly 65%, according to The Motley Fool.

A separate calculation using the full $4.1 billion total cost puts the yield on cost closer to 20%, but either way, the math illustrates what holding a great dividend compounder for 38 years actually produces.

What makes Coca-Cola worth holding for that long is a combination of brand and business model. The brand is one of the most recognized on the planet. The business model is quietly brilliant. Coca-Cola sells syrup and concentrate to independent bottling partners who handle manufacturing, logistics, and much of the capital expenditure. That keeps Coca-Cola’s own capital requirements light while the cash keeps flowing.

Alphabet Stock and the AI Revolution

Alphabet is the newest of the three. Berkshire invested $10 billion in the company, and it has since become one of Berkshire’s top five holdings. Much of the addition happened under new CEO Greg Abel, who has shown more comfort with complex technology platforms than Buffett historically displayed, as long as those platforms exhibit the same wide-moat characteristics Buffett always prized.

Start with Search. When something happens anywhere in the world, billions of people type it into Google. That behavior is so deeply ingrained it barely registers as a choice anymore. Chrome and Android make it even stickier. Google is the default on most of the world’s devices before a user ever opens a browser. That kind of distribution is almost impossible to replicate, and it has been generating advertising revenue through recessions, pandemics, and rate cycles without missing a beat.

Five years ago, Alphabet was essentially a search and advertising business with a cloud unit that hadn’t proven itself yet.

That’s not what it is anymore. Google Cloud is now a real enterprise competitor to AWS and Azure, winning AI and computing contracts from companies that were Microsoft-only customers not long ago.

Alphabet also builds its own chips, Tensor Processing Units designed specifically for AI workloads. Most AI companies have to rent computing power from someone else. Alphabet builds its own. That’s a different kind of business.