Website speed testing tools measure how fast pages load and which parts slow them, in scheduled lab runs or from real visitors.
This guide ranks twelve by ninety-day setup effort, the bill once retention and traffic grow, how results leave the product, independence from a parent company, and fit for the owner acting on the numbers.
Vendors can pay for more visibility on this website, for example a higher place in a ranking or more room on a page. Commercial relationships can also affect which products we feature and how prominently. Every vendor link is marked the same, whatever the arrangement behind it. How that works.
In short
What website speed testing tools software does
Website speed testing tools load pages under lab or field conditions and report load times, Core Web Vitals and waterfalls, so teams find slow assets and watch regressions across releases.
Every website speed testing tool was judged on five points in a fixed order: setup effort, cost in practice, how data is exported, how easily you can leave, and the buyer it was designed for. One rule overrides the order: smaller vendors take the first three places when the category offers them, and the best-known brands start at four. Our how we work page explains the choice of these five and why we give no mark out of ten.
Lighthouse runs and real visitor data in one small monitor
Ranked #1 of 12 in Best Website Speed Testing Tools in 2026.
Free tierPublished pricingElsewhere
The ladder starts at a plan with one website and ends with a contact-us level. Essentials lists unlimited users, three or more websites, 30-day retention and 75,000 or more monthly traffic; Professional lists ten or more websites, 90 days and 500,000 or more.
Each paid card shows two prices, such as 37 and 34 dollars, without saying which is annual billing. Its data, the page adds, is resident in Canada, while naming no country of incorporation.
What stands out
Lighthouse plus RUM
Free tier
Per-site plans
Where it costs you
The entry plan keeps data for only 24 hours, so trends need a paid step.
Which of its two monthly prices is annual is left unlabelled, since the page does not say which of its two monthly prices is the annual one.
Right for
Small sites wanting lab tests and visitor data without sales calls or per-seat pricing
Wrong for
Large traffic estates needing long history on every page
Not stated by the vendorPlan ladder from a free tier to a custom level; each step lifts websites, retention and monthly traffic; trial button on the paid steps
Page speed tests from 30+ locations with weekly credit allowances
Ranked #2 of 12 in Best Website Speed Testing Tools in 2026.
Free tierPublished pricingElsewhere
Every limit is laid out in a grid. Weekly website test credits run 15, 300, 750, 1,500 and 3,500 across the five plans; monitoring slots run 1, 3, 5, 10 and 30; RUM page views run from 5K to 1 million a month.
Reports are kept from one month up to two years, and AI chat messages rise from 3 to 350. Because the prices were not captured, the grid shows scale but not cost.
What stands out
30+ test locations
Weekly credits
TTFB tests
Where it costs you
Credits reset weekly, so a burst of release-day testing can drain the allowance.
Free and Starter monitor only once every 24 hours, and no amounts show on the pricing capture.
Right for
Freelancers and agencies comparing speed from many locations before committing money
Wrong for
Teams that need named-company contracts and published prices
Not stated by the vendorFive-step ladder from Free to Ultimate sized by test credits, monitoring slots and RUM views; Enterprise handled by contacting sales
Open-source testing toolkit with HAR waterfalls and page-load video
Ranked #3 of 12 in Best Website Speed Testing Tools in 2026.
Open sourcePublished pricingElsewhere
A run yields Core Web Vitals, a video of the page loading, a HAR waterfall and advice from the Coach. The Docker image bundles Firefox, Chrome, Edge and XVFB, and npm is the alternative.
Results flow to Graphite or InfluxDB, and a live dashboard at dashboard.sitespeed.io shows the output. Nothing is billed, and nothing is hosted for you, so the cost shifts from licences to the hours spent operating test machines.
What stands out
MIT licence
Docker or npm
Graphite and InfluxDB
Where it costs you
No managed service is described, so someone must run and patch the machines.
No vendor or contract is named, so there is nobody to hold to a service level.
Right for
Engineers who want raw performance data in their own Grafana stack
Wrong for
Marketing teams wanting a login, a dashboard and email alerts
Not stated by the vendorOpen source under the MIT licence; free of charge with no sign-up, installed through Docker or npm
Lab tests and visitor sessions metered per month in dollars
Ranked #4 of 12 in Best Website Speed Testing Tools in 2026.
Pricing on requestAsia-Pacific
Three plans carry a test allowance rather than a page limit. Starter gives 5,000 RUM sessions and 5,000 synthetic tests monthly with 90 days of history; Team gives 10,000 sessions, 15,000 tests and a year; Company gives one million sessions, 50,000 tests and two years, at 1,500 dollars a year.
Sites and teams are unlimited beyond Starter. Only Starter and Team come with a trial, so the largest plan cannot be tried without a call.
What stands out
Synthetic plus RUM
15-day trial
Unlimited sites
Where it costs you
Starter has three seats and 5,000 sessions, which small agencies may outgrow quickly.
The top plan is reached only by booking a demo.
Right for
Product teams that track many sites and pay for volume, not pages
Wrong for
Buyers wanting an always-on entry tier or a self-serve top plan
AustraliaStarter and Team list monthly prices in USD with a 15-day trial; Company is billed yearly and sold through a demo
Hosted by DebugBear Ltd, trial button first, price list unread
Ranked #5 of 12 in Best Website Speed Testing Tools in 2026.
Pricing on requestElsewhere
The evidence here is thin and says so. Pages read show the legal entity DebugBear Ltd, a hosted service with no mention of self-hosting, and footer links to a Trust Center, Security and a DPA.
A trial button sits in the navigation, yet its length is missing. The pricing page gave back only menu and footer text. Treat it as a candidate for a trial, and request the plan list in writing before comparing.
What stands out
Hosted service
Trust Center
DPA available
Where it costs you
No plan or amount was readable on the pricing page, so cost cannot be compared.
How long the trial lasts is not given anywhere on the pages read.
Right for
Buyers who want security documents and a DPA visible before talking to anyone
Wrong for
Anyone who must see plan limits and prices before a trial
Not stated by the vendorNavigation leads to a trial of unstated length; the pricing capture held neither plan names nor amounts
Scheduled Lighthouse tests with budgets and multi-step flows
Ranked #6 of 12 in Best Website Speed Testing Tools in 2026.
Pricing on requestElsewhere
Five capabilities are named on the home page: scheduled Lighthouse tests, CrUX tracking, real user monitoring, multi-step tests and performance budgets. The offer is a risk-free trial of 14 days with no credit card required.
Nothing on the pages read says who runs the service or where, and the pricing page gave no plans. That makes the trial the only way to learn what the budgets and multi-step tests cost at your scale.
What stands out
Performance budgets
Multi-step tests
CrUX tracking
Where it costs you
No owner, entity or country is named on the pages read.
The plan list did not load, so the cost of the budgets feature is unknown.
Right for
Teams wanting budgets and multi-step flows checked on a schedule
Wrong for
Buyers who need a named legal entity before any trial
Not stated by the vendorA 14-day trial needs no card; the captured plan list returned no amounts
Core Web Vitals tracking from CrUX data and Lighthouse
Ranked #7 of 12 in Best Website Speed Testing Tools in 2026.
Pricing on requestElsewhere
One job is described on the home page, tracking Core Web Vitals, and splits it into Pages, Sites and Scan. Pages and Sites follow individual URLs and whole sites, while Scan reads a sitemap.
Data comes from the Chrome UX Report and Lighthouse together. The copyright runs 2017 to 2026 under Treo d.o.o., and the footer gives a VAT number. What the pricing page held was not returned, so this entry makes no claim on cost.
What stands out
CrUX plus Lighthouse
Sitemap scanning
Made in the EU
Where it costs you
Plan prices were not captured, so budgeting needs a direct quote.
The country is not named on the pages read, only an EU claim.
Right for
Search and front-end teams watching Core Web Vitals across many URLs
Wrong for
Teams wanting synthetic transaction scripts or uptime alerts
Not stated by the vendorPlan list not returned by the captured pricing page, so no tier or amount was recorded
Usage sliders for RUM and synthetic checks, monthly or annual
Ranked #8 of 12 in Best Website Speed Testing Tools in 2026.
Pricing on requestElsewhere
Prices start at 90 dollars a month for Starter and 576 for Growth, both in USD, with annual billing saving 20%. Usage is set with sliders that top out at 100 million RUM views and 500K checks a month.
Growth lists 13 months of retention, high priority support and priority synthetic testing. Enterprise is custom and lists single sign-on, private agents, regular check-ins and consulting. No trial appears on the page.
What stands out
Slider-built plans
13-month retention
Private agents
Where it costs you
No trial or entry tier is named, so the first payment is a commitment.
Growth costs several times Starter, and private agents wait for Enterprise.
Right for
Performance teams that outgrow a small tool and want retention and support
Wrong for
Solo site owners needing a zero-cost starting point
Not stated by the vendorStarter and Growth are priced in USD from slider settings; annual billing saves 20%; Enterprise is custom; no trial is named
Lighthouse runs on every commit with regression assertions
Ranked #9 of 12 in Best Website Speed Testing Tools in 2026.
Open sourcePricing on requestElsewhere
The repository describes its purpose as automating Lighthouse for every commit, viewing the changes and preventing regressions. The README calls it a suite of tools for continuously running, saving, retrieving and asserting against Lighthouse results.
The licence is Apache-2.0. Nothing on the page names a hosted offer, plans or support, so it is a component to wire into continuous integration, and its running costs fall to whoever sets it up.
What stands out
Apache-2.0 licence
Per-commit runs
Regression checks
Where it costs you
The README describes tooling for builds, not a dashboard for marketers.
No support plan or company service is described on the repository page.
Right for
Developers who want speed checks to fail a pull request
Wrong for
Non-technical owners wanting a page to paste a URL into
Not stated by the vendorApache-2.0 open source repository on GitHub; no plans or prices are listed
Credit-metered checks from 230+ checkpoints, speed as one use
Ranked #10 of 12 in Best Website Speed Testing Tools in 2026.
Pricing on requestElsewhere
Core and Pro differ by users, regions and polling. Core offers 10 users, US and EU checkpoints, two-minute uptime checks and 180 days of data; Pro offers 25 users, 230-plus checkpoints, one-minute checks and 365 days.
Credits do the metering: 6 for an uptime check, 50 for a browser check or transaction step, 15 for an API step. Enterprise adds SSO, SCIM, Terraform and OpenTelemetry with checks every 30 seconds.
What stands out
Credit metering
Checkpoint regions
Status pages
Where it costs you
The page gives no monitor count per plan, so credits must be modelled.
Live chat support begins at Pro, leaving Core with tickets alone.
Right for
Operations teams adding page-speed checks to uptime and API monitoring
Wrong for
Teams after a simple page report with Lighthouse scores
Not stated by the vendorCore and Pro list monthly and annual USD prices; usage is metered in credits per check; Enterprise custom; 30-day trial without a card
Synthetic checks and RUM sold as two separate tier ladders
Ranked #11 of 12 in Best Website Speed Testing Tools in 2026.
Pricing on requestElsewhere
Synthetic Monitoring is sized by uptime checks from 10 to 30,000, advanced checks from 1 to 2,500 and SMS alerts from 50 to 1,500. Real User Monitoring is sized by pageviews from 100,000 to 1 billion, with unlimited sites and users and 13 months of retention.
Enterprise lists a quote and 500K to over 5M pageviews a month. One trial button covers both, but neither amounts nor trial length appear.
What stands out
Two product ladders
Pageview tiers
SolarWinds-owned
Where it costs you
Amounts were missing, so the cost of running both ladders cannot be added up.
Synthetics and RUM are priced separately, and each trial length is not stated.
Right for
Teams that already want uptime checks and pageview-based RUM together
Wrong for
Buyers needing a single price they can read before contacting anyone
Not stated by the vendorTwo ladders, one sized by uptime and advanced checks and one by pageview tier; amounts missing from the captured page; Enterprise by quote
Single-URL analysis page provided by Google, no account step shown
Ranked #12 of 12 in Best Website Speed Testing Tools in 2026.
Pricing on requestElsewhere
The page opens with the headline 'Make your web pages fast on all devices', a URL box with an Analyze button beside it. The footer shows the Google Developers logo and Google's terms.
It does not say whether use is free, give quotas, or state a licence, and links a v5 documentation page. Scheduling, history and alerting are absent from the home page, history or alerting features, so it works as a first look.
What stands out
Single URL
Analyze button
Google-provided
Where it costs you
Usage limits are not stated on the home page or quotas to plan around.
History and alerts are not offered on that page.
Right for
Anyone wanting a one-off check of a single URL before choosing a tool
Wrong for
Teams needing scheduled runs, history or alerting across pages
Not stated by the vendorThe page names no price, quota or licence terms; it links documentation for the v5 API
How to choose website speed testing tools software
Website speed testing tools load pages under lab or field conditions and report load times, Core Web Vitals and waterfalls, so teams find slow assets and watch regressions across releases. Rather than compare features, settle the following points one after another.
01
Work out how much a sales call is worth to you
The split is 3 published against 9 quoted. If you want to compare numbers first, start with Request Metrics, Speedvitals, Sitespeed.io.
02
Look at where the vendor sits
Look at the contracting entity: these 12 vendors sit in 2 countries, split by region as Elsewhere 11, Asia-Pacific 1. The servers can be anywhere; the legal obligations follow the company.
03
Decide if open source matters
2 of these are open source: Sitespeed.io, Lighthouse CI. Hosting one yourself trades a subscription for maintenance.
Website speed testing tools: what the entry price really buys
A speed tool is rarely priced by pages. It is priced by what it keeps and how often it looks. Request Metrics gives its entry plan one website, 1,000 monthly traffic and 24 hours of retention, while Essentials lifts retention to 30 days and Professional to 90. Speedvitals counts test credits per week and monitors Free and Starter once a day.
Calibre meters 5,000 sessions and tests on Starter, and SpeedCurve builds a plan from sliders. Compare the unit each one counts, then check how long results are kept, because a regression you spot in March cannot be traced if the March data expired within a day, and a longer history costs more on every ladder.
Note retention in days or months next to each plan, not just the price
Convert your pages and release rhythm into each vendor's unit before comparing
Ask what happens to old results when you move up a plan
Lab runs against real visitors: Calibre, Request Metrics and Pingdom
Lab tests and real user monitoring answer different questions, and vendors bill them differently. A lab run repeats the same load under fixed conditions, so a change in the score points at your code. Real visitor data shows what customers on slow phones experienced, but needs traffic to mean anything.
Pingdom sells the two as separate ladders, with RUM tiered by pageviews from 100,000 upward. Calibre bundles both and meters sessions and tests apart, and Request Metrics combines Lighthouse with RUM in one monitor. Decide which question matters for the next quarter, because buying both from separate vendors doubles the setup, the dashboards to check and the invoices to reconcile.
Decide whether the next question is a code regression or a customer experience
Check that visitor-data tiers match your real monthly traffic
Ask whether lab and visitor figures share one dashboard
Sitespeed.io and Lighthouse CI: when the test lives in the pipeline
Open source options trade invoices for operating hours. Sitespeed.io ships as a Docker image or an npm package under the MIT licence, writes HAR waterfalls and videos, and can send metrics to Graphite or InfluxDB. Lighthouse CI, under Apache-2.0, aims to run Lighthouse on every commit and stop regressions.
Neither pages you at night, and neither names a company to call. Probe locations equal the machines you run. They suit engineers who already own a build server and a Grafana board. A marketing team with no one to maintain test machines will spend more hours than a monthly plan would cost, and the bill arrives as lost time.
Name the person who patches the test machines before choosing open source
List the locations you must test from and count the machines needed
Decide where alerts will go, because the tools described offer none
Trials and unread price lists at PageVitals, DebugBear, Treo and Uptrends
Several vendors in this market show a trial button and little else. PageVitals offers 14 days without a card, DebugBear shows a trial button of unstated length, and Treo's pricing page returned no plans. Pingdom lists tiers but no amounts on the captured page, and Uptrends meters credits, 6 for an uptime check and 50 for a browser check.
Calibre gives 15 days on two plans only. When a price is hidden, the trial becomes the quote. Write down the pages you plan to test, run them for the whole trial, and request the post-trial price in writing before any card is entered, so that the first invoice holds no surprise for finance or for the engineers who chose it.
Request the plan list by email if the pricing page shows no amounts
Run the full trial against your real pages, not the demo site
Get the post-trial price and cancellation steps in writing
What goes wrong most often when buying website speed testing tools software
Buying Pingdom's RUM tier by pageviews without checking that the Synthetic ladder is billed separately, and then finding two invoices.
Treating a PageSpeed Insights result as monitoring, when the page offers no schedule, history or alerting for the next release.
Choosing Sitespeed.io or Lighthouse CI to save licence money and then leaving nobody responsible for the machines that run them.
Starting on the Request Metrics entry plan and expecting trends, when 24-hour retention means last week's results are already gone.
07
Frequently asked questions
7 answers
What is the best website speed testing tools in 2026?
First of 12 on this page is Request Metrics. Request Metrics, operated by TrackJS LLC, pairs scheduled Lighthouse testing with real user monitoring and tracks Core Web Vitals from both.
The plan list is on the page, so the bill can be forecast. The entry plan covers one website, one seat and 1,000 monthly traffic, with data kept for 24 hours only, which makes it a trial bench rather than a record of how a site changed.
Which website speed testing tools publish their pricing?
The ones that do, with the model each states:
Request Metrics: Plan ladder from a free tier to a custom level; each step lifts websites, retention and monthly traffic; trial button on the paid steps.
Speedvitals: Five-step ladder from Free to Ultimate sized by test credits, monitoring slots and RUM views; Enterprise handled by contacting sales.
Sitespeed.io: Open source under the MIT licence; free of charge with no sign-up, installed through Docker or npm.
9 vendors do not print a price.
Is there a free website speed testing tool?
Request Metrics, Speedvitals come with a no-cost entry point, a free plan or a self-hosted edition.
Where are these website speed testing tool vendors established?
Across 2 countries (Elsewhere 11, Asia-Pacific 1):
Request Metrics: Not stated by the vendor.
Speedvitals: Not stated by the vendor.
Sitespeed.io: Not stated by the vendor.
Calibre: Australia.
DebugBear: Not stated by the vendor.
PageVitals: Not stated by the vendor.
Treo: Not stated by the vendor.
SpeedCurve: Not stated by the vendor.
Lighthouse CI: Not stated by the vendor.
Uptrends: Not stated by the vendor.
Pingdom: Not stated by the vendor.
PageSpeed Insights: Not stated by the vendor.
Which website speed testing tools are open source?
Sitespeed.io, Lighthouse CI.
What should you use instead of Request Metrics?
Speedvitals and Sitespeed.io are the next two on this page. Speedvitals is for Freelancers and agencies comparing speed from many locations before committing money; Sitespeed.io is for Engineers who want raw performance data in their own Grafana stack.
Who should not buy Request Metrics?
Large traffic estates needing long history on every page. The entry plan keeps data for only 24 hours, so trends need a paid step..
Makers of website speed testing tools products can ask us to look at theirs. Say what it does and who uses it. We choose what to add, and nothing can be bought; see software advice for what a listing means.
