Skip to main content

Performance Monitoring

Most websites get slower gradually, and nobody notices until it hurts

We monitor how fast your key pages load from the outside and for real visitors, alert on meaningful changes and investigate the cause, so performance work is not undone by the next update.

All checks passingAlerting on

Business impact

Speed is not a one-off achievement

Every plugin update, new marketing tag, larger image or traffic increase can shift performance. Without monitoring, regressions are discovered by customers, or not at all.

Regressions caught early

Alerts tied to specific pages and metrics mean a slowdown introduced on Tuesday is investigated that week, not months later.

Evidence for decisions

Trend data shows whether hosting, platform or design changes made things better or worse.

Accountability with suppliers

Performance history helps when discussing issues with hosting providers, agencies or third-party script vendors.

Common issues

Problems we regularly solve

Uptime checks mistaken for performance monitoring

A site can return a 200 status and still take many seconds to become usable. Uptime alone does not show that.

Single-location testing

Tests run from one data centre close to the server miss the experience of visitors elsewhere or on mobile networks.

No real-user data

Synthetic tests cannot show interaction delays, device differences or how many visitors are actually affected.

Alert fatigue

Thresholds set too tightly trigger alerts on normal variation until people stop paying attention to them.

Slow server responses go unnoticed

Gradual increases in TTFB from database growth or worker saturation are invisible without a trend line.

Third-party slowdowns

An external script or API that becomes slow can delay your pages even when nothing on your own server has changed.

What's covered

What's included in website performance monitoring

Monitoring combines two complementary views. Synthetic monitoring tests the same pages in the same conditions on a schedule; real-user monitoring (RUM) records what actual visitors experience.

Synthetic page tests

Scheduled tests of key pages such as the home page, top landing pages, product pages and checkout steps, from selected locations and device profiles.

Real-user monitoring

Lightweight collection of load times and Core Web Vitals from real visitors, segmented by page type, device and country where your privacy setup allows.

Server response tracking

Time to first byte and error rate trends, correlated with server resource usage where we also manage the hosting.

Transaction checks

Scripted journeys such as search, add to basket or form submission, to catch slowdowns and failures in multi-step flows.

Alert thresholds

Thresholds agreed with you, based on your baseline and the Core Web Vitals good thresholds, requiring sustained breaches before alerting.

Investigation and reporting

Alerts are investigated by an engineer, with root cause and recommended action included in the monthly report.

Maintenance activities

What we do, and how often

Exact frequencies depend on your plan and platform. This is a typical schedule.

Continuous

  • Synthetic tests of key pages on a fixed schedule
  • Real-user metric collection where enabled
  • Alerting on sustained breaches of agreed thresholds

Weekly

  • Review of performance trends for priority pages
  • Check for new third-party requests and page weight changes

Monthly

  • Performance section in the monthly maintenance report
  • Comparison against baseline and previous month
  • Recommendations for any regressions found

Quarterly

  • Review of monitored pages and thresholds
  • Update of monitoring to reflect new site sections or journeys

Our methodology

How we work

01

Select what matters

We agree which pages and journeys are business-critical, rather than monitoring the home page alone.

02

Establish a baseline

A period of data collection sets normal ranges for each page and metric, including expected variation.

03

Set thresholds

Alerts are configured for sustained deviations from the baseline and for breaches of the Core Web Vitals thresholds, with severity defined in advance.

04

Investigate alerts

When an alert fires, an engineer checks recent changes, waterfalls and server data to identify the cause.

Business benefits

What you gain

Fewer surprises

You hear about slowdowns from us, with a likely cause, rather than from customers or a monthly ranking report.

Protection for optimisation work

Investment in website speed optimisation is protected by ongoing checks.

Meaningful alerts

Thresholds based on your own baseline are designed to reduce noise so that alerts are taken seriously.

Clear trend reporting

Monthly reporting shows how performance has moved and why, in language non-technical stakeholders can follow.

Technologies supported

  • WebPageTest
  • Lighthouse CI
  • Google Search Console
  • Chrome UX Report (CrUX)
  • web-vitals JavaScript library
  • Uptime Kuma
  • Grafana
  • Prometheus
  • New Relic
  • Cloudflare Web Analytics

Running something not listed? Ask us – we'll tell you honestly if it's a fit.

Who it's for

Who needs website performance monitoring?

Businesses that have invested in speed

Organisations that have completed optimisation work and want to keep the results.

Sites that change frequently

Marketing-led sites with regular campaigns, landing pages and new tracking tools.

Online stores and booking sites

Businesses where slow product, search or booking pages have a direct revenue cost.

SaaS marketing sites and apps

Teams that need performance visibility across both the public site and the logged-in application.

Industry guidance: E-commerce · SaaS Companies · Travel & Hospitality

FAQs

Website Performance Monitoring: common questions

What is the difference between synthetic monitoring and real-user monitoring?
Synthetic monitoring runs scripted tests from fixed locations under controlled conditions, which is good for spotting changes. Real-user monitoring measures actual visitors on their own devices and networks, which shows true experience. We recommend using both.
Does real-user monitoring affect privacy or cookie consent?
It can, depending on the tool and what it collects. We can use approaches that avoid personal data and work with your consent setup, but your own privacy obligations, such as under GDPR, remain your responsibility to confirm.
Will monitoring slow down our website?
Synthetic tests run externally and add no code to your pages. Real-user monitoring adds a small script, which we load in a way designed to have minimal impact.
Who responds when an alert fires?
An engineer reviews the alert, checks recent changes and investigates. Response targets for performance alerts are defined in your SLA and depend on your maintenance plan.
How is this different from uptime monitoring?
Uptime monitoring tells you whether the site responds. Performance monitoring tells you how quickly it becomes usable, for which pages and for which visitors.

Know when your website slows down

Talk to us about which pages and journeys to monitor, and how performance alerts fit into your maintenance plan.