What It Takes to Set Up Website Monitoring Properly

The monitoring tools most teams reach for are cheap, and sometimes free. The money goes into setting them up properly, because that takes someone who knows what to measure, how often, from where, and when to wake a person up. That person is an SRE, and SREs are expensive and in demand. Most small businesses cannot hire one, so they run no monitoring at all or run it badly.

There is a second gap. Datadog and Grafana watch whether a site is up and fast. Neither, as far as their public product pages show, checks whether the site's privacy policy and cookie behaviour stand up to GDPR. That is a separate category of tool, and it is billed separately.

This piece walks through the setup decisions, what they cost in practice, and what the GDPR tools charge, using public pricing pages read in October 2026.

The decisions nobody warns you about

Pointing a monitoring tool at a website takes a minute. Making it useful means answering questions that only experience helps with:

• How often to check. Every minute catches short outages but costs more and adds noise. Every 15 minutes is cheap but can miss a five-minute outage.
• From where. One location cannot tell a site outage from a network problem near the checker. Several locations can, but each extra location multiplies the number of checks you pay for.
• What counts as down. A single failed request is often a blip. Alerting on the first failure creates false alarms, so a sound setup waits for consecutive failures before it raises an alert. We cover this in our post on avoiding alert fatigue.
• Who gets alerted, and how. Email, SMS and chat have different costs and response times, and an alert nobody reads is the same as no alert.
• What good looks like. An uptime target and a response-time target (an SLO) turn raw numbers into a verdict, but choosing targets that are honest for your business takes judgement.
• How long to keep the data. Retention is a plan limit on most tools, and it decides whether you can look back at last month's incident.
• Which performance measures matter. Page speed tools report many numbers, and only a few, such as LCP and CLS, map to what visitors actually feel. Our Core Web Vitals guide explains which.

Each answer is easy to get wrong, and a wrong answer shows up as a missed outage, a flood of false alarms, or a surprise bill.

What setup looks like in Datadog and Grafana

Both are excellent platforms, and both price synthetic checks by usage. A plain HTTP check, which only asks whether a site responds, is cheap on both. A browser check, which loads the page in a real browser, costs much more per run.

The per-run prices are low. The part that surprises people is the multiplier. Run a check from three locations instead of one and the number of executions you are billed for can multiply, so the checks you add for reliability are the same ones that raise the bill. In my own experience, one website checked from three regions on Grafana cost about $50 a month.

The bigger cost is not on the pricing page. You choose the checks, write the alert rules, build the dashboards, set the retention, and keep all of it working as the site changes. Grafana's browser checks do report Core Web Vitals, but you have to know to set them up, and then read what they say.

The gap neither one fills: GDPR

A site can be up, fast and still be breaking the rules. A privacy policy that omits a legal requirement, a cookie banner with no reject button, or an advertising tracker that fires before a visitor consents are all GDPR and PECR problems, and none of them shows up as downtime. Our compliance monitoring guide covers the most common gaps.

The public pages for Datadog and Grafana describe tools for watching your own systems. Grafana's pricing page lists telemetry, observability, testing, visualization and incident management products, and none is a scan of a website's privacy policy or cookie behaviour. Datadog's nearest product, Sensitive Data Scanner, scans your logs, APM traces and RUM events for sensitive data. The listing we read makes no mention of cookie consent or privacy policy checks. That is a different job from auditing what a public website tells and does to its visitors.

We have only read the public pages, so this is a statement about what they list, not a claim about every enterprise add-on. It is also why GDPR scanning exists as its own product category.

How GDPR scan providers compare

The established cookie and consent vendors price per domain. Two of the best known are CookieYes and Cookiebot. Both are consent management platforms first: they run the cookie banner on your site, and scanning for cookies and trackers is part of that service.

CookieYes prices are in pounds and Cookiebot prices are in dollars, as each shows them, so they are not directly comparable.

Two points stand out. First, these tools charge per domain, so cost grows with every site you add. Second, what they list is cookie and tracker detection plus the banner itself. Their pricing pages do not describe scoring a privacy policy against GDPR requirements, and a site owner still has to interpret the results and fix what is wrong.

What SREmonitor does differently

SREmonitor is built for the business that has no SRE. You give it a URL and it applies the decisions above for you, so there are no agents to install and no alert rules to write.

• Uptime and response time: each site is checked every five minutes, with no check rules to write.
• Performance: a real browser loads the page on a regular schedule and reports LCP, CLS, FCP and TTFB, with a performance score you can read without knowing what those terms mean.
• SLO view: the monitor page shows uptime, performance and compliance as service-level indicators against targets, so the numbers arrive as a verdict.
• GDPR checks: the scanner reads your privacy policy against ten GDPR criteria and returns a score with a plain-English list of what is missing and how to fix it.
• Cookie compliance: testing the cookie banner and trackers is coming soon. We are not claiming it until it ships.

It is a flat price, not a meter. Pro covers up to 20 sites at £29.99 a month, currently £23.99 at the launch price, and the free plan covers 3 sites. See the pricing page for details. The comparison with the other tools is not that SREmonitor does more than a full observability platform, because it does not. It is that it does the part a small business actually needs, and the setup work is already done.

Sources

Pages read in October 2026. Prices change often, so check each vendor's page before relying on a figure.

• Datadog pricing: Synthetic Monitoring API and browser test rates.
• Grafana Cloud pricing: plan fees and the Synthetics API and Browser Testing calculators.
• Grafana k6 browser metrics and Synthetic Monitoring browser checks: Web Vitals reported by browser checks.
• Datadog Sensitive Data Scanner listing: what the product scans.
• CookieYes pricing and Cookiebot pricing: plan prices and scan limits.

The Grafana figure of about $50 a month comes from our own past experience and is not from a public page.