Security

WordPress security, before anything goes wrong.

Most WordPress compromises are not targeted. They are automated traffic trying known weaknesses against every site it can reach, which means the defence is mostly about not having those weaknesses.

Filtering before the CMS Site isolation Patched on schedule
How sites actually get compromised

Almost always something known and unpatched.

The picture people have is of somebody choosing their site and attacking it. The reality is scripts working through enormous lists of addresses, trying a handful of known plugin vulnerabilities and a dictionary of passwords. Your site is not picked; it is simply reachable.

That is good news, because it means the effective defences are unglamorous. Patch promptly, keep the attack surface small, put filtering in front so the automated noise never reaches the application, and make sure one compromised site cannot reach another.

A security plugin helps but acts late — it runs once the request has already reached WordPress and started executing code. Stopping that traffic earlier is both cheaper and more effective.

Able Web Services Ltd.UK web development
200+PROJECTS DELIVERED
140+CLIENTS SUPPORTED
15+YEARS ESTABLISHED
UKBASED & HOSTED

What we do

  • Bespoke websites and custom CMS development
  • Ecommerce, custom checkout and subscriptions
  • Web applications, portals and internal tools
  • API, CRM and business systems integration
  • Retained support, maintenance and managed hosting
  • White-label development for agencies

Working UK-wide and internationally — directly with businesses, and white label behind design and marketing agencies.

Discuss a project
The layers

Where we stop things.

At the web server

Known-bad requests and probes refused before they cost you anything. The bulk of automated traffic never gets further than this.

Before the CMS loads

Requests inspected ahead of WordPress starting, so a vulnerable plugin is never reached by the thing trying to exploit it.

Rate limiting

Per-visitor limits with escalating blocks for repeat offenders, which is what stops password guessing being worth anybody's time.

Site isolation

Each site with its own system user and processes. One compromised site on the platform cannot read or reach another.

Scheduled patching

Core, plugins and themes updated in windows and checked, rather than when something forces the issue.

Scanning on a schedule

Malware scanning on the files, so a compromise is found by us rather than by a customer or by Google.

Straight answers

What this does not do.

It is not a guarantee

Nobody can promise a site will not be compromised. What these layers do is remove the easy routes, which is what almost all automated attacks rely on.

It will not fix bad credentials

A shared admin password, or an old account belonging to somebody who left, defeats most of this. Access control is as important as anything technical.

Plugins still matter

The fewer you run, the smaller the surface. Reducing an overgrown plugin list is often the single most useful security work available.

Related services

The wider picture.

Related work

Projects of a similar shape.

A lot of our work is delivered white label or under agreement, so these describe the problem and the approach rather than naming the client.

Related

Also worth reading.

Questions

What people ask about WordPress security.

Is WordPress inherently insecure?
No. WordPress core is well maintained and patched quickly. Nearly every compromise we see comes from an outdated plugin, a weak password or a hosting account rather than from core itself.
Do we still need a security plugin?
They add something, particularly scanning and logging. But they act after the request has reached WordPress, which is late. We would rather spend the effort on filtering earlier and patching promptly, and treat the plugin as the last layer rather than the first.
How would we know if we had been compromised?
Often you would not, which is the problem. The common signs are reached late: a browser warning, a drop in traffic, or a customer getting in touch. Scheduled scanning exists so somebody finds it before that.
Can you review our current setup?
Yes. A review covers plugin and core versions against known vulnerabilities, user accounts, what is exposed publicly, and the backup position. You get the findings and a clear view of where you stand. A full written assessment is a short piece of paid work rather than something we give away.
Next step

We will review what you have first.

Versions, accounts, what is exposed and whether the backups work. That tells you where you actually stand, and it is yours to act on with us or anybody else.