Visual Testing Pricing: What to Compare

To compare visual testing prices, estimate your monthly workload in each vendor’s actual billing unit, then compare the cost and coverage at that volume. A page, browser, and viewport combination may count as one screenshot in one service, while another service bills page or component checkpoints. Those units are not interchangeable.

Start with this planning estimate:

monthly visual checks × builds per month × browsers × responsive widths

Then adjust for reruns and retries, and compare included usage, overage, capture model, review workflow, browser coverage, and contract terms. Treat public prices as dated plan details and verify them on the vendor’s live pricing page before purchase.

1. Start with the usage unit

The headline price is meaningful only when you know what the vendor counts. A “visual check” can expand into multiple billable captures as you add browsers, viewport widths, pages, or components.

BrowserStack’s Percy documentation defines a screenshot as a rendering of a page or component in an individual browser and responsive width combination. For example, two pages tested in two browsers at three widths produce 12 screenshots under that model. See Percy’s plan and billing documentation.

Billing unit	What to establish	Why it matters
Screenshot	Does each page or component, browser, and width combination count separately?	Coverage multiplies usage.
Snapshot	What creates a snapshot, and do browser or viewport variants multiply it?	Snapshot allowances may not match raw screenshot counts.
Page checkpoint	How is a page checkpoint counted, and what scope does it cover?	It may not be equivalent to a component checkpoint.
Component checkpoint	What does the allowance cover per component and run?	Component and page allowances can differ substantially.

Do not convert one vendor’s unit into another by assuming a one-to-one relationship. Use the vendor’s definition and your intended test configuration.

2. Estimate your monthly volume

1. Count pages, stories, or components that produce visual checks.
2. Estimate how many builds run each month, including pull request and scheduled builds.
3. List target browsers and responsive widths.
4. Multiply those factors as an initial estimate, where applicable.
5. Add a realistic allowance for retries, failed CI runs that repeat captures, and intentional reruns.
6. Translate the result into each vendor’s billing unit and compare it with included usage.

Example planning model: 40 stories × 30 builds × 2 browsers × 3 widths = 7,200 estimated capture combinations per month. This is a planning calculation, not a universal vendor billing formula. A service may define its units differently, apply optimizations, or count only certain capture types.

Record the assumptions beside the estimate. If the build count is uncertain, calculate a low, expected, and high scenario rather than selecting a tier from a single optimistic number.

3. Compare the published plans carefully

The figures below come from the research available for this guide. They are not a synchronized quote sheet. Vendor prices and plan names can change, so check the linked live pages before making a decision.

Service	Published or reported plan detail	Usage unit and qualification
BrowserStack Percy	Desktop: $199/month billed annually for 10,000 screenshots/month. Desktop & Mobile: $599/month billed annually for 25,000 screenshots/month. Official docs also describe a free plan with 5,000 monthly screenshots.	Screenshot count is based on page or component, browser, and responsive width combinations. Verify current plan details on the official pricing page and billing documentation.
Applitools	Starter: $667/month, paid annually; higher tiers are described as customizable.	The listed Starter allowance is 100,000 component checkpoints or 1,000 page checkpoints. These checkpoint units should not be compared directly with screenshot or snapshot counts. See Applitools pricing.
Argos	Reported at $100/month for 35,000 screenshots; reported overage of $0.004/screenshot and $0.0015 for Storybook screenshots.	These are figures reported in an Argos-authored comparison checked in July 2026, not independently verified current terms. Confirm directly with the vendor.
Chromatic	Reported Starter at $179/month for 35,000 snapshots, Pro at $399/month, and $0.008 per extra snapshot.	These figures were reported in a July 2026 comparison. Snapshot volume may multiply by browsers and viewports. Verify current pricing and counting rules with Chromatic.

For Percy, the official pricing page is the source of truth when older comparison articles disagree. In particular, older secondary pages may show a different entry price. The Argos and Chromatic amounts above are dated secondary evidence, so do not treat them as current quotes.

4. Calculate the likely monthly bill

For each candidate, use the same workload scenarios and apply that vendor’s actual unit definition:

estimated bill = plan price + applicable overage for usage above the included allowance

Check these details before calculating:

• Whether the allowance resets monthly and whether unused volume rolls over.
• Whether extra usage is billed automatically, blocked, or requires a plan change.
• The per-unit overage rate and whether different capture types have different rates.
• Whether the vendor bills for retries, parallel runs, or captures from failed builds.
• Whether annual billing is required for the displayed monthly equivalent.
• Whether concurrency or parallel-run limits could lengthen CI time or require a higher tier.

If overage is not clearly documented, ask the vendor how it handles a month that exceeds the allowance. Do not assume excess usage is free or automatically capped.

5. Compare coverage and workflow, not only volume

A lower price can represent a different testing scope. Check whether the plan supports the browsers, devices, viewport sizes, components, and pages your team needs. Also compare how captures run and how reviewers handle changes.

Area	Questions to ask
Capture model	Does rendering happen locally or in a vendor cloud? Which browsers and environments are available?
Coverage	Can you test the required browser and device combinations, responsive widths, pages, and components?
Review workflow	Can reviewers inspect visual changes, approve them, and connect the review to the pull request or build?
CI integration	How are checks started, reported, retried, and associated with a commit or build?
Volume controls	Can you limit scheduled runs, narrow capture scope, or manage reruns to keep usage predictable?
Team and contract	What are the seat, retention, support, security, and cancellation terms? Which require a higher tier?

Verify the terms that affect your own procurement process directly with the vendor. Public plan tables may not answer every question about concurrency, overage, or contract changes.

6. Keep the estimate useful over time

Visual test volume changes as teams add pages, stories, browsers, and pull request builds. Revisit the estimate when coverage expands and compare actual usage with the assumptions in your plan.

• Track the number of builds and visual checks that actually run.
• Identify frequently repeated captures and avoid rerunning unchanged scope where the service permits.
• Keep browser and viewport coverage intentional; add combinations when they address a real compatibility need.
• Recheck pricing and unit definitions before renewal, since plan terms are volatile.
• Keep a high-volume scenario for release periods or growth in test coverage.

Do not reduce meaningful coverage solely to fit a headline allowance. Instead, estimate what the needed coverage costs and decide whether its review workflow and risk reduction justify the spend for your team.

7. Where ScreenshotNeo fits

ScreenshotNeo is a website screenshot API and MCP server for developers. It can help when your workflow needs clean page captures for visual checks, reports, or agent-driven inspection. Its billing model is per clean screenshot: bot checks and CAPTCHAs, blank pages, timeouts, failed loads, and cache hits cost nothing, and response headers identify the page verdict and billing status. That is a different unit and use case from the visual regression plan allowances compared above, so compare it against the capture work you actually need.

For a page-level capture, one request returns an image or PDF. See the ScreenshotNeo API documentation for options and response details.

Or skip the browser setup

Use a single GET request to capture a page as WebP:

curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webp

ScreenshotNeo removes cookie and consent banners, newsletter popups, and chat widgets before the shot. Bot checks, blank pages, and failed loads are never billed. An MCP server lets AI agents take screenshots, and the free plan includes 1,000 screenshots per month with no card; paid plans start at $5 for 3,000. See the API docs for parameters and formats.

Sign up free for 1,000 screenshots a month, with no card required.

FAQ

How much does visual testing cost per month?

It depends on the vendor’s usage unit, your monthly build and coverage volume, and whether you exceed the plan allowance. Estimate your workload first, then apply current plan and overage terms.

What counts as a snapshot in visual testing pricing?

Definitions vary by service. A snapshot may be tied to a page, component, browser, viewport, or combination. Read the vendor’s billing definition rather than assuming it equals one screenshot.

Is free visual testing good enough?

A free allowance may suit a small workload or an initial evaluation. Check its monthly limit, supported workflow, coverage, and what happens when usage exceeds the allowance before relying on it for a team’s regular CI runs.

Can I compare component checkpoints with screenshots?

Not directly. They describe different billing units and possibly different scopes. Compare the number and type of checks your team needs under each vendor’s own definition.