"Up to $45 an hour." That figure ran on the Handy platform for furniture assembly jobs. It was a ceiling posted where the floor should be.

Most people who have looked at a gig listing or a pitch deck have seen some version of those two words. "Up to" sounds like a disclosure. It is a frame. And the frame does its work before anyone reads the fine print.

The research on this is worth reading. In early 2025, the Federal Trade Commission and the New York Attorney General filed a joint action against Handy Technologies, now doing business as Angi Services. The complaint was plain. More than 90 percent of workers on the platform made on average more than $20 an hour less than the $45 figure in the ads.

Lawn care listings ran higher. Handy quoted rates as high as $62 an hour. Fewer than one in ten workers ever reached it. The gap between the posted rate and the earned rate was not a rounding error. It was the whole story.

The settlement cost $2.95 million. The FTC mailed 62,893 checks, putting $2.7 million back in workers' hands. Each check was a receipt for the distance between the ad and the truth.

"Up to" did not fail the workers. It failed the system the workers tried to build around it.

Arise Virtual Solutions ran the same play at a larger scale. The company posted ads quoting "up to $18 an hour." Its own records showed the average was $12. Of every thousand workers who signed up, 999 never hit the posted rate.

This was not sloppy copy. Arise ran its own market tests. The inflated number drew more sign-ups than simply advertising "extra income." They saw the data. They kept the ads running. It was a tested system.

The structural problem is not that the top number is false. The problem is that the top number replaces the one that would tell you the truth: the median.

There is a name for this in the literature. Two psychologists, Amos Tversky and Daniel Kahneman, called it anchoring in 1974. The first number you see pulls your estimate toward it. You adjust from the anchor, but you never adjust far enough.

A team led by Li, Weigel, Ferraro, and Messer tested that claim again. Their replication, published in Economic Inquiry in early 2025, confirmed the pull is real. But the lab effect had been overstated by roughly a factor of seven. The bias does not need to be dramatic. It just needs to set the starting point.

The pattern does not stop at gig platforms. The same two words appear in partnership pitches, course sales pages, and affiliate offers. The product changes. The tactic does not.

The Handy settlement now requires the company to back up future pay claims with what a typical worker is likely to make. The legal system just named the replacement principle for you. Stop reading ceilings. Start asking for medians.

Once that is clear, three moves follow from it.

When a pitch leads with a figure, ask one question. "What does the median earner make?" Not the top tier. Not the best month. The middle. If the person across the table will not answer, that silence is the answer. One question. Ten seconds. It cuts through most inflated claims before you spend a dollar or an hour.

That one question asks you to sit with the pause after you say it. Most people rush to fill the gap. Do not.

The FTC found that Handy's highest rate was gated behind targets most workers could not reach. High job counts, high ratings, a tier system built to keep most people in the lower bands. The number on the screen described a small group at the top. It was sold as if it described everyone.

When you see "up to" on a course sales page, an affiliate pitch, or a franchise brochure, ask a second question. "How many people who started this reached that number?" A percentage tells you more than a promise ever will.

The FTC sent 1,100 penalty offense notices beyond Handy. The list went to franchises, multi-level marketing firms, coaching companies, and other gig platforms. A search for "FTC," the company name, and "earnings" takes thirty seconds. If a federal complaint exists, the gap between the ad and the check is already public. Read it before you sign a thing.

Running these three checks over a single quarter does something the advice never did:

At the end of 90 days, ask three things.

→ Which deal passed the median test and moved a real number in your business?

→ Which one looked like progress but left no trace once you checked the math?

→ Which friction showed up more than once when you asked for the real figure?

That is the difference between advice that sounds right and a system that proves itself.

$45 an hour still sits on the Handy app. The number has not changed. The question you ask before you act on it has.