Usability Testing: How Much You Need and What It Costs

By Nick Babich

Key Takeaways

• Budget per round: in my model, one moderated remote round with five participants costs $1,250 to $2,200 in cash when your team runs it, and $5,750 to $10,200 when an outside UX partner runs it end to end.
• Five per segment, not per product: five participants are a sensible default for one group of users doing one set of tasks. Every segment with different tasks needs its own five.
• Three small rounds beat one big one: 15 sessions split into three rounds, with fixes in between, find more problems than 15 sessions run at once for the same money.
• Researcher time is the biggest line: planning, moderation and analysis make up about 70 to 75 percent of an outsourced round in the worked example below.
• Scale up only for numbers: benchmarks, SUS scores and A/B comparisons need 20 or more participants per segment. Finding problems does not.
• Read the quote line by line: check whether recruiting fees and incentives are included or billed later as expenses.

Usability testing is cheap per session and easy to overspend on per program. In the budget model I built for this guide, one moderated remote round with five participants costs about $1,250 to $2,200 in cash when your own team runs it, and $5,750 to $10,200 when an outside UX partner handles recruiting, moderation and analysis. A three-round program for a B2B product with two user segments, 30 sessions in total, comes to $27,150 to $39,000 outsourced. Every line item and rate is listed below, so you can swap in your own numbers.

The useful questions are how many people, how many rounds and what the invoice will look like. If you are an early-stage team weighing a proposal from one of the startup-focused UX design teams we have reviewed, the ranges here show what a fair testing line looks like and where quotes tend to leave costs out.

Sample size comes first below, because participants and rounds drive every other figure. When you compare UX design partners that include usability testing in their work, ask each one to price the same plan, so you compare like with like.

What One Round of Usability Testing Costs

A round is one batch of sessions on one version of the design, followed by analysis and a readout. The ranges below are for a single round with one user segment: five sessions plus one backup participant. They come from my own budget model, which uses these rates: $75 to $150 per participant for recruiting and screening, a $100 to $150 incentive for a 60-minute session with a professional user, $200 to $400 for tools, and a blended outside researcher rate of $150 to $200 an hour. These are estimates built from line items, not quotes.

The infographic below sums up how much testing you need and what each setup costs.

• In-house unmoderated remote: $300 to $800 in cash. You pay pay-as-you-go platform credits and small incentives for 15- to 20-minute task sessions. Annual enterprise platform plans cost far more. Plan on 8 to 12 hours of staff time to write tasks and watch the recordings.
• In-house moderated remote: $1,250 to $2,200 in cash for recruiting, incentives and tools, plus 30 to 40 hours of your team's time.
• Outsourced moderated remote: $5,750 to $10,200. The same cash costs, plus 30 to 40 researcher hours at $150 to $200 an hour ($4,500 to $8,000) for planning, moderation, analysis and the readout.
• Outsourced in person: $9,750 to $17,200. Start from the outsourced remote range and add a usability lab for two days at $1,500 to $2,500 a day, plus $1,000 to $2,000 for travel and recording equipment. Higher in-person incentives are not included.

A five-session round looks expensive per session once a researcher is involved, and that is expected. Planning and analysis take almost as long for five sessions as for ten, so adding a second segment to the same round lowers the cost per session.

How Many Participants and Rounds You Actually Need

Teams often overspend by recruiting 20 people for one study, expecting four times the findings of five. That logic fits surveys, not qualitative tests, where problems start repeating after the first few sessions.

The Five-User Rule and Its Limits

The usual rule of thumb is five participants per round. It holds up as an average, but an average hides risk. When Laura Faulkner tested the five-user assumption with 60 participants, drawing random groups from the full set, some groups of five found 99 percent of the problems and others found only 55 percent. With 10 users, the weakest group still found 80 percent, and with 20 users it found 95 percent.

So five is a good default when another round will follow, because what one group misses gets a second chance to surface. It is a weak default for a one-off test of a flow where a missed problem is expensive, such as payments.

The rule also applies per segment, not per product. If your software serves hospital administrators and floor nurses, their tasks and mental models differ, and five administrators tell you nothing about how nurses will cope. You need five of each. Add a segment only when the tasks really differ, not just the job titles.

Why Iterative Testing Beats a Single Study

A common budgeting mistake is spending the whole budget on one large test at the end of a project, when the design is too far along to change cheaply. Three rounds of five users cost about the same in incentives and recruiting as one round of 15, and they find more. In round one, participants trip over the obvious problems, such as navigation, labels and missing steps. You fix those. In round two, the next five get far enough into the flow to hit the deeper issues the first problems were hiding, like confusing logic or data that does not match what users expect. You fix those too. Round three checks the fixes and catches the remaining friction before launch.

When to Scale Up for Quantitative Metrics

Larger samples matter when the goal shifts from finding problems to measuring performance: a task completion rate you will report to leadership, a System Usability Scale (SUS) score, time on task, or a comparison of your product against a competitor or an older version. With five people, a single failure moves your completion rate by 20 points, so the number tells you very little. For a benchmark I plan at least 20 participants per segment, and more when you need to show that one version beats another. These studies usually run unmoderated, so the cost per session drops even as the count rises.

How Much Testing You Need by Situation

Use this table to size a testing program before any money is discussed. Start from the situation, then multiply by the number of segments.

If you are not sure the product is ready for testing at all, start smaller. An expert review that can come before or instead of testing clears the obvious issues, so paid sessions go to the questions only real users can answer.

The Line Items in a Usability Testing Budget

Whether you run tests internally or hire outside help, every usability testing budget is made of the same four blocks. You can estimate any round with one formula:

Cost per round = [Participants x (Incentive + Recruiting fee)] + (Researcher hours x Hourly rate) + Tool fees

Count participants per segment, add one backup per segment, then multiply the round total by the number of rounds. Here is what goes into each variable.

Participant Incentives

People need to be paid for their time. In my model, a consumer testing a retail app might accept a $50 to $75 gift card for a 60-minute session. Professional B2B users usually need $100 to $150 in cash. Specialist roles, such as enterprise security leads or physicians, can need $250 to $400 per session before they agree to join. Short unmoderated tasks pay $10 to $30.

Recruiting and Screening Fees

Finding the right people takes effort. A panel provider or recruiter charges a fee on top of the incentive, usually $75 to $150 per participant depending on how strict the screening criteria are. Always recruit and pay at least one backup per segment to cover no-shows. Recruiting from your own customer list removes the fee, but it costs staff time, and existing customers already know the product, so they are the wrong pool when you are testing how new users get started.

Researcher Time

Staff time is the most expensive line in any outsourced round. For a moderated round with 10 sessions, a researcher spends 10 to 15 hours designing the tasks, writing the discussion guide and running a pilot session. Moderating the sessions takes about 15 hours, including buffers and notes between calls. The heaviest lift comes after the sessions end: analyzing the data, clipping highlights, ranking findings by severity and building the readout takes 20 to 30 hours. That is 45 to 60 hours per round. A five-session round still needs 30 to 40 hours, because planning and reporting do not shrink much with fewer sessions.

Tool Fees

Video calls with recording, transcription, scheduling and, for unmoderated work, a testing platform. In my model they add $200 to $400 per round; outside teams often fold them into the fee.

What Drives Costs Up or Down

Quotes for the same study can differ twofold or more. These variables explain most of the gap.

• Audience availability: testing a consumer fitness app is cheap because almost anyone can take part. Testing a dashboard used only by freight dispatchers is expensive, because the recruiter has to find them and they expect more for their time.
• Number of segments: each segment with different tasks adds five or more sessions, plus its own screener and analysis.
• Method: remote sessions over video are the most cost-effective. In-person work adds lab rental, camera setups, travel and on-site time.
• Accessibility: testing with people who use screen readers, magnifiers or switch devices needs specialized recruiting and a moderator trained to run sessions without interfering with the participant's setup. That expertise costs more.
• Session length: a 90-minute session for a complex workflow raises both incentives and moderation hours by about half compared with a 60-minute one.
• Depth of analysis: a $4,000 round usually means raw notes or an automated summary. A $9,000 round should include edited highlight clips, severity-ranked findings and specific design recommendations your team can act on.

In-House Versus an Outside UX Partner

Running tests in-house looks cheaper on paper because teams treat their own salaries as sunk costs, so an internal round only shows incentives, recruiting and tools.

The hidden cost is focus. Asking a product manager to recruit users, schedule sessions, moderate calls and synthesize the findings pulls them away from their main job for one to two weeks per round. It also adds bias: the person who designed the flow is the least likely to notice when they lead a participant.

An outside partner sells speed, method and distance from the design. A good researcher spots leading questions, does not rescue struggling participants and reports harsh feedback plainly. Many teams mix the two: the partner plans and runs the first round and trains the team, and the team runs later rounds with the same guide.

If testing is one part of a wider research phase with interviews or surveys, read how to scope a wider user research study before you ask for quotes, so the testing line is not priced in isolation.

When a quote arrives, check that it states:

• Participants: how many per segment, how many backups, and who recruits them.
• Incentives: whether they are included in the fee or billed later as a pass-through expense.
• Rounds: how many, and which version of the design each round tests.
• Sessions: length, moderated or unmoderated, and remote or in person.
• Deliverables: highlight clips, severity-ranked findings, recommendations and a live readout.
• Recordings: who owns them, how participant consent is collected and how long data is kept.

A low base fee with recruiting and incentives billed separately in the fine print is the usual reason a cheaper-looking quote ends up costing more.

Worked Estimate: Three Rounds for a B2B Onboarding Redesign

To show how the variables interact, here is an estimate I modeled with the formula above. It is my own estimate, not a quote and not a project we ran: every number comes from the rates stated in this article, and you can rerun it with yours.

A B2B SaaS company is redesigning its onboarding flow. It has two segments with different tasks: account admins, who set up the workspace and invite the team, and end users, who complete their first piece of work. Following the five-user default, each round has 10 sessions, and the team recruits 12 people (one backup per segment) and pays all 12 incentives. It plans three rounds: the current flow, the first prototype and the final high-fidelity design, for 30 sessions in total. Sessions are 60 minutes, remote and moderated, at a blended researcher rate of $150 an hour.

The outsourced program comes to $27,150 to $39,000, or $9,050 to $13,000 per round. Researcher time is $6,750 to $9,000 of each round, about 70 to 75 percent of the total, which is why the hourly rate matters more than anything else in the quote. At a $200 rate, the same plan costs $11,300 to $16,000 per round and $33,900 to $48,000 overall.

Run in-house, the cash cost falls to $2,300 to $4,000 per round, or $6,900 to $12,000 for the program. The trade-off is 45 to 60 hours of someone's time per round, 135 to 180 hours in total. If the team also recruits from its own customer list, cash drops to $1,400 to $2,200 per round, but the onboarding rounds then miss the people who matter most: users who have never seen the product.

I kept researcher hours flat across the three rounds to keep the math easy to check. In practice, rounds two and three reuse the discussion guide, so planning time often shrinks and the high end of the range is a safe ceiling. To see how this testing line sits next to discovery, design and handoff in a full project, read the full UX budget across research, design and testing.

Frequently Asked Questions

How many users do you need for a usability test?

Five participants per distinct user segment is a sensible default for qualitative testing, as long as another round follows. Groups of five vary a lot in what they find, so for a one-off test of a high-risk flow, plan 8 to 10. For benchmarks and SUS scores, plan 20 or more per segment.

How much does one round of usability testing cost?

In my model, a moderated remote round with five participants costs $1,250 to $2,200 in cash when your team runs it, plus 30 to 40 hours of staff time. Outsourced end to end, the same round costs $5,750 to $10,200, depending mostly on the researcher's hourly rate.

Is unmoderated testing enough?

Unmoderated testing works well for simple UI changes, preference tests, copy checks and quantitative benchmarks. For complex workflows it falls short: a moderator can ask a hesitating participant why, while a recording leaves you guessing.

How can you cut testing costs without losing quality?

Run remote sessions instead of lab sessions, test one segment per round when the tasks overlap, and keep sessions to 60 minutes. Recruiting existing customers removes panel fees for tests of features they already use.

Conclusion

A usability testing budget is simple once you price it per round: participants times incentive and recruiting, plus researcher hours, plus tools. Start with five participants per segment, plan several small rounds instead of one large study, and save large samples for the moments you need numbers. Whoever runs the sessions, ask for the same plan to be priced line by line, so you know what you are paying for before the first participant joins.