A conversion rate optimization specialist (CRO specialist) finds out why visitors don't buy, sign up or subscribe, turns that into testable ideas, and runs A/B tests to prove which changes actually raise revenue.
The job is part research, part statistics and part persuasion: most ideas fail, so the value lies in picking better bets and reading results honestly.
In the US, Indeed puts average pay for the role at $111,259 a year, with a range of about $66,000 to $188,000 (updated October 2026).
This page covers what the job involves, what it pays, and how to get into it.
Every claim about what testing work looks like is backed by real results from our database of 1,724 A/B tests, each credited to the team that ran it.
What a CRO specialist does
The core loop is the same whether the employer is a Shopify store, a SaaS company or an agency: research, hypothesis, test, analyze, document.
Our guide to the CRO process walks through each step in detail.
- Research. Analytics funnels, heatmaps, session recordings, surveys and user interviews to find where and why people drop off. See customer research for CRO.
- Hypotheses. A specific claim such as "shoppers abandon because shipping cost appears too late," tied to a metric.
- Test design. Choosing the primary metric, estimating sample size and run time, and briefing designers and developers.
- Analysis. Checking significance, segment effects and guardrail metrics before calling a result.
- Communication. Explaining losses as clearly as wins, and keeping a record so the team doesn't retest old ideas.
Most of the job is managing losses
New specialists are often surprised by how rarely ideas win.
In a Microsoft paper on online experimentation, Ronny Kohavi and colleagues reported that only about 1/3 of ideas improved the metrics they were designed to improve, with the rest flat or negative.
Agencies that publish everything show the same pattern.
Clean Commit publishes its full test library, and of the 173 Clean Commit tests in our database, 16 won, 35 lost and 122 were flat.
A good specialist treats that as normal and builds a program that learns from every result.
Context beats rules of thumb
A CRO specialist can't run a playbook of "best practices," because the same change can win in one place and lose in another.
Dave Diederen added a countdown timer to checkout and saw +3.16% revenue per visitor, with wins repeated across four separate tests.
Yet Dave's team, after testing countdown timers across 20+ brands, found they often hurt outside the US: when they added a countdown timer for a non-US (European) market, revenue per visitor fell 35.61% and conversion rate fell 17.35%.
Intuition is just as unreliable. Store owners often assume popups annoy shoppers, but when Dave Diederen removed the email popup on a client store, revenue per visitor dropped 20.51% and conversion rate fell 12.18%.
The test was then repeated on a second store and lost again.
Picking the right metric
Much of the skill is choosing what counts as a win.
Clean Commit tested a wider desktop collection page: click-through rate rose a significant 4.0%, but conversion rate fell 5.9% and orders fell 8.0%.
A specialist who reported clicks would have shipped a loss.
The same trap applies to pricing. In an Elevate A/B Testing case, Biddlebee cut a bottle's price from $29.95 to $24.95.
Conversion rose from 2.38% to 3.24%, but profit per visitor fell from $0.28 to $0.23, so the brand kept the original price.
Reading results honestly
Specialists also need to say "we don't know" when the data is thin.
A Clean Commit test of a ship-date message on product pages showed -6.9% conversion rate over 21,816 visitors and 24 days, but it wasn't statistically significant.
Knowing the difference between a trend and a result is covered in our guide to statistical significance.
Big redesigns carry big risk
Large changes are where judgment matters most. In the same Clean Commit library, replacing a homepage hero with a sales carousel cut conversions by 46.2%.
A specialist's job is to break bold ideas into pieces that can be tested, so a loss teaches the team something specific instead of just "the redesign didn't work."
That's also why specialists push for testing over launching.
Without a control group, a 46% drop could easily be blamed on seasonality or ad spend, and the redesign would stay live.
@DaveDiederen@DaveDiederen@DaveDiederen
Clean Commit
Elevate_abClean CommitSkills employers look for
| Skill area | What it looks like in practice |
|---|---|
| Analytics | Building funnels and segments in GA4 or a product analytics tool; spotting tracking errors |
| Statistics | Sample size, significance, confidence intervals, and knowing when a test can't detect the effect you care about |
| Qualitative research | Writing unbiased survey questions, running usability tests, reading session recordings |
| UX and copy | Turning research into concrete design and copy changes |
| Testing tools | Setting up experiments in a platform such as Optimizely, VWO, Convert or a Shopify testing app; basic HTML, CSS and JavaScript help |
| Communication | Writing test briefs and readouts that a founder or VP can act on |
Job titles you'll see
The same work appears under many names: CRO specialist, CRO analyst, conversion optimization manager, experimentation manager, growth analyst and web optimization specialist.
In-house roles tend to focus on one product.
Agency roles cover several clients at once, which means more tests and more variety.
A typical week mixes deep work and coordination: reviewing live tests for sample ratio problems or tracking bugs, analyzing finished tests, writing briefs for the next ones, and meeting with design, development and marketing.
Junior specialists usually spend more time on setup and QA; senior ones spend more on prioritization and stakeholder buy-in.
Conversion rate optimization salary
There is no official government pay series for CRO specialists, so salary figures come from job-site estimates.
They vary by source, title and country, so treat each one as a range rather than a precise number.
| Source | Role / scope | Figure | Date |
|---|---|---|---|
| Indeed | Web optimization specialist, US (Indeed's page for "conversion rate optimization specialist" redirects here) | Average $111,259; range about $65,892 to $187,861; 160 salaries | Updated October 28, 2026 |
| IT Jobs Watch | UK job ads citing conversion rate optimisation | Median £42,500; 10th percentile £31,200; 90th percentile £78,000 | 6 months to November 2, 2026 |
| US Bureau of Labor Statistics | Market research analysts (closest official category, not CRO-specific) | Median $78,760 per year | May 2025 |
A few patterns hold across sources. Pay rises with seniority and with how close the role sits to revenue.
Senior specialists who can run a whole experimentation program, or who can code their own tests, sit toward the top of the range.
The UK figure is a warning about volatility.
IT Jobs Watch shows the median down 22.73% from £55,000 in the same period of 2025, so a single year's number can mislead.
Freelancers and consultants price differently, usually by retainer or project.
Rates depend on traffic, the number of tests per month and whether design and development are included.
How to become a CRO specialist
There's no required degree or license. Most people move in from an adjacent role, such as analytics, UX design, paid media, ecommerce management, copywriting or front-end development, and build the missing skills on the job.
1. Learn analytics first
Every test starts with data, so get comfortable with an analytics tool.
Google's Analytics Academy on Skillshop offers free courses and a Google Analytics certification.
Our GA4 for CRO guide covers the reports specialists use most.
2. Learn enough statistics to avoid false wins
You don't need a statistics degree, but you do need to understand sample size, significance, statistical power and why stopping a test early inflates false positives.
Practice with our sample size calculator before you design your first test.
3. Study real test results, not just best-practice lists
Reading what others actually tested trains your judgment faster than any checklist.
Browse the LiftArchive test database by page type or element, and pay special attention to the losses.
Our CRO books list covers the experimentation classics.
4. Run your own tests
Employers want proof that you've done the work. Options include:
- Volunteering to run tests for your current employer's site or emails.
- Testing on a side project, a nonprofit's site or a friend's store.
- Writing up a full research audit of a public website, even without a test.
Low traffic is the usual barrier. If a site can't reach a useful sample size, test bigger changes, test higher-traffic pages, or focus your portfolio on research and test design.
5. Build a portfolio of write-ups
For each project, record the problem, the research, the hypothesis, the result and what you'd do next.
Include tests that lost or came out flat: hiring managers know most tests don't win and look for honest analysis.
Keeping a searchable record of past tests is a habit that carries straight into the job.
6. Prepare for interviews
Expect questions about a test you ran, how you'd prioritize ideas, and how you'd explain a losing result to a stakeholder.
Our list of CRO interview questions includes sample answers.
If you'd rather hire than become a specialist, our directory of CRO experts lists practitioners whose real test results appear in the database.