Playwright Load Testing

Run your Playwright Test scripts at scale with thousands of real browsers. Use the same scripts for load testing, stress testing, and 24/7 site monitoring.

Bring your own Playwright tests. Launch thousands of bots in minutes from 32 cloud regions.

Your team likely already has a Playwright test suite. It already works with your most important user flows, login sequences, checkout journeys, and key features.

Loadster takes those same scripts and runs them at load testing scale. Launch thousands of concurrent bots, each driving a headless Chrome browser, executing your Playwright script against your real site.

No need to write separate load testing scripts, and no need to pick up another scripting language. You don't even have to maintain separate test suites. The Playwright scripts your QA or dev team already maintains are the same ones that will power your load testing, stress testing, and 24/7 site monitoring.

Because every bot automates a real browser, your test results will reflect what real users experience. Full-page rendering, client-side JavaScript, network timing, everything. Real browser testing is especially important for modern JavaScript-heavy web applications where protocol-level load testing would miss half the story.

Most Playwright tests run without any changes. If you need to parameterize data, Loadster makes it easy.

Hammering your site with identical requests from every bot isn't always a useful load test. Caching and data locking might skew the results. Loadster's optional @loadster/bot helper library lets your Playwright scripts pull unique values from a dataset on each iteration, so every bot behaves like a different real user, with Loadster's platform handling the orchestration, scaling, and parameterization.

import { test } from '@playwright/test';
import bot from '@loadster/bot';

test('search and browse', async ({ page }) => {
  const [username, password] = bot.getVariable('credentials');
  const searchTerm = bot.getVariable('searchTerm');

  await page.goto('https://example.com/login');
  await page.fill('#username', username);
  await page.fill('#password', password);
  await page.click('button[type="submit"]');

  await page.waitForURL('**/dashboard');
  await page.fill('#search', searchTerm);
  await page.press('#search', 'Enter');

  await page.waitForSelector('.results');
});

Dataset variables flow through from Loadster into the script automatically, so each bot in the load test gets its own login, search term, or any other column of data you need. The helper also exposes bot context like the current iteration number, bot identifier, and elapsed time for fine-grained scripting.

Real browser metrics from every bot. Response times, Core Web Vitals, and detailed traces.

Because Loadster runs real browsers for Playwright tests, you get real browser metrics from every bot — not just server response times, but the full client-side experience.

Response time graph under Playwright load

Response times per navigation

Every page.goto(), click, and navigation is measured. Watch how response times change as the load ramps up, aggregated by URL and broken down by average and percentile.

Core Web Vitals during a Playwright load test

Core Web Vitals captured automatically

Time to First Byte (TTFB), First Contentful Paint (FCP), and Largest Contentful Paint (LCP) are collected from every page navigation. These metrics often degrade well before your site starts returning errors, giving you an early warning that the user experience is suffering.

Error rates during a Playwright load test

Script errors and assertion failures

Standard Playwright assertions, timeouts, and navigation failures are all tracked throughout the test. When an assertion fails at 500 concurrent users but passed at 100, you've found a bottleneck worth investigating.

Playwright trace viewer in Loadster

Playwright traces for failed runs

When a Playwright script fails under load, Loadster captures the full Playwright trace — the same one you'd view locally with npx playwright show-trace. Step through the recording inline to see exactly where things went wrong.

Load Test with Playwright

Load testing and monitoring, all with the same scripts. Reuse your Playwright scripts for 24/7 site monitoring.

The Playwright scripts you use for load testing also work as a 24/7 monitor. Instead of running thousands of bots at once, monitors run a single bot on a regular schedule. This tests your site every minute, every five minutes, or whatever cadence makes sense for you.

If a monitor detects a failure — like a timeout, a broken assertion, a page that didn't load — Loadster alerts you via email, SMS, phone call, or your incident management integration of choice. The same Playwright trace that helps you debug a failed load test also helps you diagnose a monitor alert.

Loadster Site & API Monitoring »

Playwright isn't your only option. You can also test with Browser Bots and Protocol Bots.

If your team doesn't have a Playwright test suite, no worries! Loadster supports other options.

Browser Bots run scripts recorded with Loadster's Chrome extension. They're great for teams who prefer recording and playback with a graphical editor over writing code. Browser Bots also run in real headless Chrome, and use Playwright behind the scenes.

Protocol Bots run lightweight HTTP scripts perfect for API load testing or high-throughput testing of simple websites. They're considerably cheaper per bot than browser-based testing.

Review the Loadster Manual for an in-depth look at scripting, or the Load Testing Guide for the high-level goals and techniques of load testing.

Better test results, less effort. Move faster with browser-based script recording, fully managed cloud infrastructure, live test metrics, and automatic reports.

1

Record test scripts

Record Playwright load test scripts right in your browser with the Loadster Recorder extension for Chrome or Firefox. Just navigate your site like a real user, while Loadster automatically records every action into a reusable test script.

2

Edit and replay scripts

Play scripts as often as you like in Loadster's script editor. Bots execute your script and give you immediate real-time feedback. In the editor you can add, edit, and replace steps from the recording, or you can even build a script from scratch.

3

Configure test scenarios

A scenario is a reusable configuration for a Playwright load test. You'll specify the number of bots in each group, the AWS or GCP regions bots will run from, which of your scripts to execute, and the schedule for ramping the load up and down.

4

Launch your Playwright load test

When you launch a Playwright load test from a scenario, Loadster quickly spins up cloud resources to power your bots. Tests usually start in about 2-3 minutes, using fully managed infrastructure through Loadster's AWS and GCP cloud providers.

5

Observe real-time results

While a Playwright load test runs, your bots report real-time metrics to Loadster, and Loadster surfaces the latest details and diagnostics within seconds. Keep tabs on response times, throughput, errors, and detailed traces so you can see how your site is performing under load.

6

Generate test reports

After a Playwright load test finishes, Loadster automatically creates a detailed report so you can review your site's performance and scalability. This includes detailed graphs showing response times, errors, network and transaction throughput, and more. The report also includes detailed traces with request and response bodies or screenshots to help with troubleshooting.

7

Repeat and iterate

Playwright load testing is an iterative process, and it's rare that everything goes perfectly on the very first try. After all, the point of the test is to discover scalability bottlenecks so you can fix them instead of crashing in production. With Loadster you can quickly re-test a previous scenario, or make changes to your scripts or scenarios and run again at any time.

Load test with Playwright from 32 cloud regions across AWS and GCP clouds. Monitor from 8 dedicated locations on 5 continents.

Loadster scales with your testing and your budget. Choose between flexible pay-as-you-go and subscription pricing.

Monthly Fuel Plans

Starting at $78 a month

Subscribe to a monthly plan to automatically add fuel every month at a discount. Unused fuel rolls over as long as you keep the plan.

Plan Pricing

Pay As You Go

Starting at $97 for 1000 units

Purchase as much Loadster Fuel as you need for your load testing and monitoring. No commitment, just buy it when you need it.

Fuel Pricing

Loadster has helped improve the performance, scalability, and uptime of thousands of sites. Proud to be chosen by many of the best teams.

  • Nginx
  • Wells Fargo
  • Amazon
  • Queensland Government
  • Accenture
  • Yale University
  • HBO
  • Mint Mobile
  • KPMG
  • Perficient
  • Dell
  • Pearson

Move fast
Don’t crash

Ready to cut your risk of downtime? Load test to prevent crashes. Monitor to react quickly.