Investing & Dividends Mostly accurate, with one big caveat

By the Income Reality Check Editorial Team · Updated 2026-08-19 · 7 min read

This article is general information, not financial, tax, or investment advice. Income claims and platform fees change. Talk with a licensed professional before making financial decisions based on anything you read here.

Oracle, AppLovin and Brian’s “5 stocks on sale”: what the receipts really mean

Verdict: Mostly accurate, with one big caveat. The numbers are real and mostly check out — but a paid gold-stock promotion sits in the middle of the list, and “signed revenue” is not the same thing as a safe bet.

“This Sale Won’t Last – 5 Stocks Worth Buying” has pulled in more than 220,000 views for the channel BWB – Business With Brian. The pitch: a market correction knocked down five high-quality companies, and Brian is buying them because they trade on “receipts” — signed, contracted revenue — instead of Wall Street guesses. Unlike most videos we review here, the core financial claims are largely verifiable. That’s the good news. The caveat is what’s tucked between the picks.

What the video actually claims

Brian opens by describing himself as a former Target and Amazon corporate employee who retired at 46. He then walks through five stocks the market “got bored with” at the wrong moment: Oracle, Innodata, Sterling Infrastructure, MasTec, and AppLovin.

The through-line is his “receipt” idea. Most stocks trade on estimates of future sales, he argues, so they fall hard when the mood sours. A handful trade on booked, contracted revenue that a customer is legally obligated to pay. His headline example is Oracle, which he says sits on $638 billion in signed contracted revenue while trading 57% below its high — most of it from a single $300 billion, five-year deal to supply computing power to OpenAI. He runs similar math on the others: Innodata’s 58% quarterly revenue growth, Sterling’s $5.5 billion-plus backlog, MasTec’s $21 billion record backlog, and AppLovin’s roughly 80% operating margins.

He’s also candid about risk, repeatedly arguing against himself — Oracle’s cash flow has gone negative, Innodata trades at 46 times earnings with insiders selling, MasTec carries real debt. He closes with the standard “I’m not a financial adviser, this is educational” disclaimer.

Do the numbers hold up?

Mostly, yes — and that’s genuinely unusual for this genre.

Oracle really did report $638 billion in remaining performance obligations (RPO) at the end of fiscal Q4 2026, up 363% year over year, according to the company’s own results and CNBC’s coverage. The $300 billion OpenAI deal is real; Bloomberg reported that Bank of America analysts estimate more than half of Oracle’s RPO leans on OpenAI alone — a customer that loses money and hasn’t gone public. Brian’s own “the risk is real” caveat matches the reporting. He’s not hiding the ball there.

AppLovin’s story checks out too. Short-seller firms including Muddy Waters and Fuzzy Panda published reports in early 2025 accusing its AXON ad engine of improper data harvesting; the SEC opened an inquiry that October, per CNBC. And on the same August 2026 day the company reported a soft quarter, its CFO confirmed the SEC had closed the probe with no action. Morningstar’s analysis tracked the same arc. Brian’s framing — best possible news and worst-received number in one report — is a fair read.

Innodata’s 58% Q2 2026 growth and 12th straight quarter of gains are accurate. On MasTec, the $21 billion figure is right (reported at roughly $21.4 billion). Sterling is where he stretches slightly: reporting around that period put signed-plus-combined backlog closer to $5.2 billion, with “future phases” pushing visibility toward $6.5 billion — so his “$5.5 billion-plus” is in the neighborhood, not exact. Small thing, but worth flagging.

The caveat the framing glides past

Here’s the part that deserves a hard look. Sandwiched between Innodata and Sterling, Brian pivots to Howard Marks and Oaktree Capital, then to a company called Mayfair Gold — and says, almost in passing, that this is “the portion disseminated on behalf of Mayfair Gold Corporation.”

That single phrase changes the segment entirely. It means Mayfair is a paid promotion, not a pick he found by running his own screen. The SEC warns investors directly about exactly this setup: seemingly independent commentary that is actually paid stock promotion, where a writer or presenter is compensated to talk up a stock. The agency’s guidance is blunt — look for who paid, how much, and in what form, and be skeptical when those details are thin.

In the video, the Mayfair segment is delivered in the same confident, receipt-driven tone as the four genuine analyses around it. A viewer half-listening would have no reason to file it differently. That’s the mechanism the SEC flags: paid placement that borrows the credibility of the surrounding independent content. The disclosure exists — buried in one clause and a “link in the description” — but the presentation does the opposite of setting it apart.

What “signed revenue” does and doesn’t guarantee

Brian’s receipt metaphor is useful, and mostly honest. But it can carry more weight than it should.

A signed contract is booked revenue, not banked profit. Oracle is a live example: to deliver on that $638 billion, it has to build data centers first, and that construction spending flipped its free cash flow to roughly negative $24 billion — a fact Brian states plainly. The receipt only pays out if the customer pays, if the build finishes on budget, and if the contract doesn’t get renegotiated. When more than half of your order book depends on one unprofitable, privately held buyer, the receipt and the risk are the same piece of paper.

Would you buy a house because the builder showed you a stack of signed purchase agreements? You’d probably ask whether the buyers can actually close. That’s the question the “receipts are real” framing quietly answers for you.

Who this is (and isn’t) for

These are five large, real, heavily analyzed companies — not penny stocks or a course funnel. If you already invest in individual equities, understand that AI-infrastructure names are volatile (AppLovin fell more than 50% from its high; Innodata carries 14% short interest), and can hold a position through a bad quarter without panic-selling, the video is a reasonable starting point for your own research. Brian’s self-critical tone is a point in its favor.

It’s a poor fit if you’re a beginner looking for a “sale” that’s guaranteed to bounce, if you’d treat one YouTube segment as due diligence, or if you can’t tell the paid Mayfair pitch from the four unpaid ones. And nothing here is individualized advice — including this article.

What to remember

The backlog and contract numbers in this video are real, verifiable, and mostly stated fairly, with risks acknowledged rather than hidden. That puts it well above the average “buy these now” clip. The caveat is twofold: one of the five segments is a paid promotion presented in the same voice as the rest, and “signed revenue” is a reason to look closer, not a guarantee of profit. Treat the four genuine picks as a research to-do list, treat the gold segment as an advertisement, and price both accordingly.

For more on how these “stocks on sale” videos are built, see our looks at 16 stocks to buy now (July 2026) and Dips don’t last: 8 stocks I’m buying.

Views and figures were accurate at the time of review and may have changed since publication.