Freelance designer hourly rates: what each survey is actually measuring

A UX designer looking up what to charge will find, within about a minute, that the answer is $40.62 an hour (PayScale, May 2026) and also $100 an hour (a 2025 survey of web designers). Both figures are real. Both were published in the last eighteen months. Neither is wrong about what it measured. The difference between them is entirely a question of who was asked, and that question is the one every rate article skips.

Here are the six hourly figures we could verify, with the population attached to each, because the population is the number.

The figures, with their populations

| Source | Figure | Who was measured | |---|---|---| | State of Web Designer Pricing, 2025 (Web Designer Academy) | median $92.75/hr, average $100/hr, revenue tiers $76–$131/hr | 208 web designers, about 60% in the US, drawn from a pricing-coaching programme's audience | | PayScale, May 2026 | $40.62/hr average for the UX design skill | 929 US respondents reporting a base rate for the skill; not freelance-specific | | PayScale, March 2026 | $29.27/hr median for illustrators; range $17.45–$73.98 | 156 US respondents; entry level about 22% below the average, late career about 65% above | | Ruul, April 2026 | junior $20–35, mid $35–60, senior $60–150+/hr | a compiled ladder for freelance designers, corroborated against PayScale and ZipRecruiter; medium confidence | | YunoJuno 2026 Rates Report | roughly $63/hr designers, $69/hr creatives, $75/hr UX, converted from UK day rates | 182,000+ freelance bookings on one UK platform | | Broad aggregators (ZipRecruiter and similar), 2026 | near $35/hr for "designer" | job-listing and salary-report data, mixing employed and freelance, all seniorities |

All six are in the confirmed section of our facts dossier, each with its year and its warning label.

Why the same title comes out three times apart

Four things move the number, and none of them is the quality of the designer.

Self-selection. A survey run by a pricing-coaching business is answered by people who sought out pricing coaching, which means people already thinking about charging more, often already charging more. The Web Designer Academy figure is honest about its sample and it still skews high for exactly this reason. The dossier's own note says to cite it as "surveyed web designers," never as "the market."

Employed versus freelance. PayScale's skill rates are dominated by people reporting a base rate for a skill they use in a job. An hourly figure derived from a salary carries none of the freelancer's unbilled time, tax, or gaps between projects, so it reads low against a freelance rate that has to cover all three.

Country. YunoJuno's data is UK day rates, converted to dollars for convenience. Day rates on a booking platform bundle a different set of expectations (a day booked is a day paid) from an hourly rate quoted to a small business.

Seniority collapsed into one number. An aggregator that averages a first-year designer with a twenty-year art director produces a figure that describes nobody. Ruul's ladder is the only source here that keeps the levels apart, and it is also the one the dossier marks medium confidence, because it is compiled rather than surveyed.

Put those together and the four figures, from $29.27 to $92.75 across the 2025 and 2026 sources, stop being contradictions. They are four different rooms.

What these numbers are for

A floor check, and nothing else. Your floor is the hourly rate below which your year does not add up: your income goal plus your costs, divided by the hours you can genuinely bill. It comes from your own arithmetic, not from a survey, and the floor calculator does it in two minutes.

The survey figures are how you check that your floor is not absurd in either direction. Say your arithmetic comes out at $22 an hour and you are a senior designer in a US city: something in the arithmetic is wrong, usually the billable-hours line. Say it comes out at $180 when the senior band in the one compiled ladder tops out around $150 (Ruul, April 2026): you should at least know that before a client says it.

What the surveys are not for is setting a project price. A logo is not priced by multiplying a survey median by an estimate; it is priced by what the mark has to do for the client, checked against the floor so it never quietly works out to $25 an hour. The method is a separate and longer subject, and the value-based pricing guide is where it lives.

The ones we will not print

We looked for freelance-specific hourly data broken down by seniority and discipline from the big talent marketplaces, and found bands that could not be traced to a survey anyone would name. They are on the refuted list, not on this page. The same goes for every "average logo costs" table. A number with no population is not a number.

One more caveat, since this page is about caveats: all six figures above are a snapshot. PayScale's are dated to the month. Surveys get re-run. The freshness date at the top of the page is when we last checked, and if it is more than a year old when you read this, treat the whole table as history.