Dual-Class Shares Ate 40% of My Investment and I Didn't See It Coming

the pitch that sucked me in

In March 2025, I snagged 2,000 shares of VeloCloud Systems at $67 per share thru my Charles Schwab account, putting $134,000 into a single position. The company made edge computing software for telecommunications providers and had just announced a major partnership with Microsoft Azure that sent the stock up 12% in a single day. Revenue was growing 45% year over year, and the product reviews from Gartner and Forrester were glowing. The IPO prospectus cited dual-class share structure in a footnote on page 187, sandwiched between a description of the company's lease obligations and a table of executive compensation. Class A shares, which I snagged on the open market, had one vote each. Class B shares, held by the founder and early venture investors, had 20 votes each. I didn't read the footnote. I was focused on the revenue growth numbers and the Azure partnership press release.

why dual-class exists

Technology companies love dual-class share structures cuz founders can raise billions in public markets without giving up operational control of the company they built. Mark Zuckerberg did it at Meta. Larry Page did it at Google. Travis Kalanick did it at Uber before he was forced out. The argument in favor is that visionary founders need protection from short-term shareholder pressure and activist investors who might force cost-cutting or premature monetization. That argument sounds reasonable in a Silicon Valley boardroom. It stops sounding reasonable the moment a founder uses those extra votes to do somethin that destroys shareholder value, cuz there is absolutely nothin you as a Class A shareholder can do about it. The founder of VeloCloud, a man named Ryan Tsai who I'd rarely met and knew nothing about beyond his LinkedIn profile, controlled 72% of the voting power while owning only 19% of the economic interest in the company.

the secondary offering that changed everything

In September 2025, VeloCloud announced a secondary offering of 8 million Class A shares at $58 per share, a 14% discount to the then-current trading price of $67.50. The company's press release said the capital would fund international expansion into Europe and Southeast Asia, which sounded reasonable on its face. Then I clocked who was actually selling the shares in the offering. Ryan Tsai was selling 2 million of his personal Class B shares, converting them to Class A fore the sale. He pocketed $116 million in cash. The total share count of Class A stock outstanding increased from 53 million to 61 million, a 15% dilution. My ownership stake, which I reckoned was fixed at 2,000 shares out of 53 million, was now worth a slight percentage of a larger pie. Tsai's personal selling didn't just dilute my economic interest. It signaled that the founder believed the stock was overpriced at $67. The market read it exactly the same way I did, only faster. The stock dropped to $52 within a week of the offering pricing.

how the math worked against me

Before the September offering, there were 53 million Class A shares outstanding, and I owned 2,000 of em. After the offering, there were 61 million. My percentage ownership of the company fell from 0.00377% to 0.00328%. That sounds microscopic, and it is in isolation. But multiply that dilution across thousands of retail investors an institutional holders who snagged in the IPO or on the open market, and the cumulative effect is enormous. Tsai's selling represented a transfer of wealth from public shareholders to the founder personally. The stock dropped from $67 to $52 and maintained falling. By January 2026, VeloCloud was trading at $41. My 2,000 shares were worth $82,000, down from $134,000. thats a loss of $52,000, or 39%, in less than a year, in a company whose revenue was still growing at 30%+.

the acquisition nobody asked for

In February 2026, Ryan Tsai announced that VeloCloud was acquiring a slight artificial intelligence startup called NeuronBase for $340 million in cash and stock. NeuronBase had $12 million in trailing revenue, no profits, 45 employees, and a product that had been on the market for less than eight months. The acquisition was approved by the board of directors, which Tsai controlled entirely thru his super-voting Class B shares. There was no shareholder vote on the acquisition because VeloCloud's corporate charter, written by Tsai's lawyers fore the IPO, allowed the board to approve acquisitions valued at less than $500 million without seeking shareholder approval. Class A holders like me had zero ability to block it, comment on it, or even delay it. I called VeloCloud's investor relations department and spoke with a woman named Diane who told me, politely, that the dual-class structure allowed the board to act in what it determined to be the best long-term interest of the company. I hung up and stared at my laptop screen for a long time.

the PMI analogy that hit home

I was discussing the VeloCloud situation with a friend who works as a loan officer at a regional bank in Phoenix. She said the situation reminded her of PMI, or private mortgage insurance, which homeowners pay when they put down less than 20% on a home purchase. PMI protects the lender, not the borrower. In the same way, dual-class share structures are a kind of insurance policy for founders, protecting em from the consequences of bad decisions that shareholders might otherwise punish thru voting. But unlike that insurance, which eventually gets removed once you build enough equity in your home, dual-class shares rarely go away. The founder keeps the 20-to-1 voting advantage forever, even as they sell down their economic stake. The routine shareholder keeps paying the price in diluted influence, misallocated capital, an acquisition decisions driven by founder ego rather than shareholder return.

what I did next

I sold my 2,000 shares at $43 per share in April 2026, booking a total loss of $48,000 on my original $134,000 investment. That's 36% of my money gone in 13 months. I reinvested the remaining $86,000 in a Vanguard Sa Vanguard S&P 500 index fundP 500 fund where at least nobody with 20-to-1 voting rights can approve a $340 million acquisition of an unprofitable startup without asking me first. The painful irony is that VeloCloud's core business was actually performing reasonably well. Revenue hit $420 million in fiscal 2025, up from $290 million the prior year. The product was good. The customers were mostly happy. The problem wasnt the business at all. The problem was the governance structure that allowed one person to make capital allocation decisions with zero accountability to the people who actually owned the company.

my new rule for buying stocks

I now check the proxy statement, expressly Schedule 14A, before buying any individual stock. If the company has a dual-class structure where one person or a modest group of insiders controls more than 50% of the total voting power, I pass immediately and move on to the next idea. I dont care how fast the revenue is growing. I don't care how compelling the product seems. I don't care what the analyst price targets say. I've been burned once, and the financial scar tissue is permanent. There are currently 47 companies in the Russell 3000 index with extreme dual-class structures where insiders control more than 60% of total voting power. I maintain a list of all 47 on my phone. I will rarely own a single share of any of them.