Scheduled monitoring in DeskSEO: checks, score history and alerts

A one-off audit tells you how a site looked on one day. Most SEO problems, though, arrive later and without warning: a release that adds a noindex to every page, a plugin update that breaks canonicals, a hosting change that makes the server slow, a certificate that expires. Each of these can cost weeks of traffic if nobody notices. Scheduled monitoring exists to shrink that delay from weeks to a day.

This guide explains how DeskSEO's monitoring works and how to use it well.

How scheduled checks work

Every domain you add under Your sites is checked against the same 20 criteria as the free SEO checker, on a schedule set by your plan:

Plan	Domains	Scheduled checks	On-demand checks per day
Free	1	Weekly	3
Starter	3	Daily	20
Pro	10	Daily	50
Agency	30	Daily	150

Each check measures the home page in depth (status, redirects, speed, tags, structured data, indexability), the site-level setup (http to https, www handling, robots.txt, sitemap) and a sample of internal links from the home page. Every result is stored, so you can go back to any earlier check and see exactly what it found. Current prices are on the pricing page.

What triggers an alert

DeskSEO compares each scheduled check with the previous one and emails you when:

• The score drops by five points or more. Smaller movements are usually noise, for example a slightly slower response on a busy server.
• A new critical issue or warning appears that was not there before, such as a noindex, a robots.txt block, a broken canonical, a broken sitemap or a redirect problem.
• The home page becomes unreachable: DNS, connection or certificate failures. When it comes back, the next check records that it is reachable again.

Each check is compared with the one immediately before it, whether that was scheduled or on-demand. Emails are sent after scheduled checks; differences found by an on-demand check are recorded on the site page without an email, since you are looking at the result already. The alert email names the site, what changed and links to the check. The site page also keeps a list of recent alerts, and the overview of all your sites shows how many alerts each had in the last 30 days.

Alerts are on by default. You can switch them off in Account, under profile and alerts. Checks continue either way.

Reading the score history

The site page shows a chart of the score over the last checks, with a table below listing each check's score and the number of critical issues, warnings and minor issues. Each point on the chart opens that check. Patterns worth recognising:

Pattern	Usual meaning
Flat line	Nothing changed; good
Sudden drop, stays down	A release or configuration change introduced a problem; compare the two checks
Sudden drop, recovers next check	A temporary outage or slow server at check time
Small zigzag of one or two points	Response time varying with server load
Step up	A fix took effect

When the score drops, open the check before the drop and the one after, and compare the issue lists. The new issue is usually obvious. The score dropped guide walks through what to do in the first fifteen minutes.

On-demand checks after a fix

After fixing an issue, press Check now on the site page rather than waiting for the next scheduled check. The result confirms the fix and resets the comparison for future alerts. On-demand checks count toward a daily limit by plan; scheduled checks do not.

How long history is kept

The score and issue counts of every check stay in the history, so long-term trends remain visible. The full detail of each check (every measurement and the robots.txt text at the time) is kept for several months and then trimmed to save space; if you need a permanent record of a particular check, for example to document a client incident, save the result page when it happens.

Why five points

The threshold is deliberate. Response time is measured from the checker on each run and naturally varies by a few hundred milliseconds depending on server load, which can move the score by one to three points between checks. Alerting on every movement would train people to ignore the emails. Five points or more, or any new critical issue or warning, is the level at which something has almost always genuinely changed. Problems that build up slowly, such as response time creeping upward over months, show in the score history chart rather than in alerts; reviewing the chart once a month catches them.

Monitoring is not a full audit

DeskSEO's checks are deliberately focused. They catch the site-wide failures that cost the most traffic, quickly and cheaply, every day. They do not crawl every page, so a problem limited to one deep template (for example, product pages only) may not appear. Pair monitoring with:

• A periodic full audit with a crawler; see the audit checklist.
• Search Console's Page indexing report, which covers every URL Google knows about.
• Connected Search Console data in DeskSEO, which shows clicks and positions per page and query, so a drop limited to one section is visible; see connecting Search Console.

If you want an important inner page checked as well, the free checker accepts full page addresses; run it on one URL per main template after releases.

When a plan ends or changes

If a paid plan expires, the account returns to the Free plan. Nothing is deleted: domains, history and alerts stay. Scheduled checks continue for the domains the plan allows, in the order they were added, and the others pause until the plan is renewed or upgraded. Renewing or buying a plan from your balance extends it from where it stands, so there is no need to wait for the old period to end. Plans are paid from a prepaid balance in US dollars; the agencies guide explains how the balance works.

Making alerts useful for a team

• Route alerts to someone who can act. Alerts go to the account's email address; if that inbox belongs to a manager who forwards things weekly, the speed advantage is lost. Use an address that the person responsible for the site actually reads, or set up a forwarding rule.
• Agree what happens on an alert. A short rule such as "check the site within the working day; if critical, roll back the last release" saves debate when it matters.
• Note releases. Keeping a simple log of deploys and plugin updates makes it easy to match a drop in the score history to the change that caused it.

A worked example

A hypothetical online shop updates its SEO plugin on a Friday afternoon. The update changes a default setting so that category pages, including the home page template, get a canonical pointing to the first paginated page with a parameter. On Saturday morning, the scheduled check finds a canonical that points to another URL, flags it as a new warning, and the score drops by seven points. The alert email arrives before anyone is at work; the developer sees it on Monday morning, compares the two checks, reverts the setting and presses Check now. The score returns to its previous level. Without monitoring, the problem would have surfaced weeks later as a slow decline in category traffic, with no obvious date to connect it to.

Checklist

1. Add every domain you are responsible for, within your plan's limit.
2. Make sure alerts are enabled and go to someone who acts on them.
3. After each fix, run Check now to confirm it.
4. Review score history monthly alongside traffic data.
5. Run the free checker on one URL per main template after major releases.