
What is EDR, and is antivirus still enough?
What is EDR, and do you need it when every laptop already runs antivirus? The short answer: antivirus stops what it recognises, EDR records what happens and lets you act when something gets through anyway. The longer answer is about who reads the alerts, and that is where XDR and MDR come in.
Antivirus blocks what it recognises. EDR records what happens on a device, spots suspicious behaviour and lets you take a laptop off the network remotely.
The difference lies less in detection than in follow-up: EDR produces alerts, and an alert nobody reads protects nothing.
XDR connects more sources than the device; MDR is a service with people who watch and act. Start with who decides at three in the morning, not with a product name.
What is EDR?
EDR stands for endpoint detection and response. It is software on laptops, desktops and servers that does three things: it records what happens on the device (which programs start, which files they touch, which addresses they connect to), compares that behaviour with what attackers do, and lets you intervene when something goes wrong.
That last part is what matters. From one console you can isolate a laptop from the network while it can still talk to the administrator, stop a process, quarantine a file and trace back how something got in. Some products can roll back changes made by ransomware; do not treat that as a substitute for a backup.
What EDR is not: a product you install and forget. It produces alerts, and some of them are false positives. Switching it on without deciding who reads those alerts buys you an expensive log.
EDR vs antivirus: the difference in practice
Traditional antivirus compares files with the signatures of known malware and blocks whatever matches. Modern antivirus looks at behaviour too, so the line is blurrier than the brochures suggest. The real difference lies in three things:
- What is kept. Antivirus reports a block and forgets the rest. EDR keeps the chain: which document started a script, which fetched a program, which opened a connection. Without that chain you cannot tell after an incident what else was touched.
- What you can do. Antivirus blocks or allows. EDR lets you isolate a device, search whether the same file sits on other machines, and follow an attack back step by step.
- Who does the work. Antivirus largely runs itself. EDR needs someone to assess alerts and decide.
Attackers often no longer use a malicious file at all, but tools already present on every Windows machine, such as PowerShell or a remote management tool, with a stolen password. A signature has nothing to recognise there; behaviour does. That is why EDR exists, and also why it generates more alerts.
Do you need anything besides Microsoft Defender?
This is the question we hear most, and the answer starts with your licence. Every Windows machine has built-in antivirus. Whether you also have EDR depends on the subscription: Microsoft 365 Business Premium, for example, includes Defender for Business, an EDR aimed at smaller organisations, while many other plans only give you the built-in antivirus. Check before you buy anything; it is common for EDR to be paid for and never configured.
If you have it, the next question is whether it works as intended: are all devices onboarded, including the Macs and the servers, are the policies set centrally, and does an alert reach a person? A second product alongside it answers none of those three.
A different product can make sense: if you have many devices outside Windows, if your IT provider works on another platform and has people on hand for it, or if you deliberately do not want your security with the same vendor as your office software. Do not let two products scan actively at the same time, though; they get in each other’s way. In that case the built-in antivirus belongs in passive mode.
Who reads the alert at three in the morning?
An attack rarely starts on a Tuesday at ten. Agree what happens before the first alert arrives:
- Who receives the alert. A shared mailbox nobody reads does not count.
- Who may isolate a device. Without asking, even if it is the managing director’s laptop, or only after a phone call? The difference is often an hour, and in that hour ransomware encrypts a file share.
- What may happen automatically. Most EDR products can isolate a device or stop a process on an unambiguous alert. Decide per type of alert whether that is switched on.
- Who decides on the bigger steps. Resetting passwords, shutting down a server, calling the insurer: that is not for whoever reads the alert but for someone named in advance.
If the honest answer to the first question is “nobody outside office hours”, that is no reason to leave EDR off. It is the reason to take the MDR question further down seriously.
EDR vs XDR: what the X adds
XDR, extended detection and response, puts device alerts next to those from other sources: email, sign-ins and accounts, cloud applications and sometimes the network. An attack is rarely visible in one place. A phishing email, a sign-in from an unusual location and a laptop that then starts something odd are three separate alerts; XDR turns them into one incident.
In practice XDR is usually one vendor extending across its own products. That works well when email, identity and devices already sit with that vendor, and less well when your estate is five brands. So do not ask whether a product is called XDR; ask which of your sources it actually ingests.
The step makes sense once three things are true:
- EDR runs on every device and its alerts are followed up;
- you already have more than one source worth listening to, usually email and accounts;
- there is someone who can investigate a correlated incident, because XDR improves the overview, not the workload.
If the first point is missing, XDR gives you a nicer dashboard over the same gaps. XDR is also different from a SIEM, which collects logs from any source; that is covered in our piece on SIEM and SOAR.
MDR is a service, not a product
MDR, managed detection and response, is EDR or XDR with people attached: a team that assesses alerts day and night, filters out false positives and steps in during a real attack. For an organisation without its own security team, that is usually the step that pays off most, more than a heavier product.
Providers differ widely in what “response” means. Ask:
- Do they act or advise? Does the provider isolate a device, or do you get an email advising you to?
- On which devices? Servers too, Macs too, devices at a branch with its own network too?
- Who calls whom? A phone number and a list of names called in order, outside office hours as well.
- Where their work stops. After isolation a device has to be rebuilt, an account restored and a user back at work. Does the MDR provider do that or your IT support, and do the two know about each other?
- What you see. A monthly report of what was flagged, what was done and what is still open.
An order that works for an SME
If you are looking at this for the first time, work in this order. Each step only makes sense once the previous one is in place.
- The basics. Timely updates, multi-factor authentication on every account, no daily work with admin rights. Without these three, no detection product catches up.
- Antivirus, centrally managed. On every device, with one view showing where it is switched off.
- EDR with an agreement. On all laptops and servers, with a written answer to who reads alerts and who may isolate.
- MDR when there is nobody. When the answer to “who watches at night” is nobody, or one person who also takes holidays.
- XDR when there is more to connect. When email and accounts demonstrably play a part in the incidents you see.
Two questions help you choose between EDR with your own arrangement and MDR: how long can your business run without files and email, and how long does it take today before someone sees an alert that arrives at midnight? If the second is longer than the first, that is your gap.
Where our work stops
What we do. We deploy EDR on laptops and servers, set the policies and automatic actions, and follow up alerts within the agreed hours, as part of our managed IT or of managed security. During an incident we isolate devices, restore devices and accounts, and get people working again.
What we do not do. We do not make security software: how well a product detects is the manufacturer’s work, and no product stops everything. We do not carry out forensic investigations that must stand as evidence or that an insurer or regulator requires; a specialist investigation firm does that, and we supply the data from the EDR console. Nor are we an insurer or a lawyer: whether an incident must be reported is something you assess with them.
Questions we get about this
The five that come up most often.
What is EDR?
EDR stands for endpoint detection and response: software on laptops and servers that records what happens on a device, recognises suspicious behaviour and lets you act, for example by isolating a laptop from the network. Unlike antivirus, it also spots misuse of tools already on the device, and it keeps the chain of events so you can see afterwards what was affected.
Is EDR the same as antivirus?
No. Antivirus blocks what it recognises and reports it. EDR keeps a record of what happens on the device, recognises behaviour and gives you the means to intervene. Many EDR products include antivirus themselves, so EDR usually builds on antivirus rather than replacing it.
Is Microsoft Defender enough for a small business?
The antivirus built into Windows is a reasonable baseline. Whether you also have Microsoft’s EDR depends on your licence; Microsoft 365 Business Premium, for example, includes Defender for Business. Whether it is enough depends less on the product than on whether it is active on every device and whether someone follows up the alerts.
What is the difference between EDR, XDR and MDR?
EDR watches the device. XDR puts that next to other sources, such as email and accounts, and turns separate alerts into one incident. MDR is not a product but a service: people who assess EDR or XDR alerts around the clock and act on them. For a smaller organisation without its own security team, MDR is usually a bigger step forward than XDR.
Do you also carry out forensic investigations after an attack?
No. We isolate, restore and get people working again, and we supply the data from the EDR console. A forensic investigation that must stand as evidence, or that an insurer or regulator asks for, is done by a specialist investigation firm. Whether an incident must be reported is something to assess with your lawyer and your insurer.
Further reading
The layer around it, and who keeps watch.
Shall we check who reads your alerts?
Tell us which licences you have and what runs on your devices today. We will tell you whether you already have EDR without using it, who sees the alerts now, and what it takes to cover the night as well.
Practical IT knowledge in your inbox
New guides on management, security and the workplace, written by the people doing the work. No sales talk, and you can unsubscribe in one click.