Know a company with years of records? Refer it and earn up to $100K
Security operations

MDR Provider: What It Does and How to Compare Providers

What an MDR provider does, how to compare managed detection and response services, the questions to ask and what the contract should say about your data.

Short answerChecked

An MDR provider runs security monitoring and response for you from its own platform: its analysts watch your endpoints, network, cloud and logs around the clock, investigate alerts, hunt for threats and can contain an attack, for example by isolating a host, under rules you approve. Compare providers on coverage, response authority, speed, reporting and what the contract says about your data.

Staffing
24/7A mandatory MDR feature (Gartner)
Log retention baseline
6 months+For the most important logs (CISA and partners)
End of service
Delete or returnPersonal data, at the controller's choice (GDPR Art. 28)
A server rack with blue patch cables and a laptop showing a ticket queue
Key points
  • MDR means managed detection and response: a remote security operations center (SOC) that also acts, not only alerts.
  • Agree in writing which actions the provider may take without asking, such as isolating a laptop.
  • Compare on telemetry coverage, response authority, response times, reporting and data terms, not on a vendor ranking.
  • The contract should say where your logs live, how long they are kept, what the provider may share and how you get your data back.

What does an MDR provider do?

Managed detection and response (MDR) is security monitoring and response run by an outside team. Gartner "defines managed detection and response (MDR) services as those that provide customers with remotely delivered security operations center (SOC) functions" and adds that providers use "a predefined technology stack that commonly covers endpoints, networks, logs and cloud" (Gartner). A SOC is the team that watches for and responds to attacks.

The difference from plain monitoring is the response. Gartner lists as mandatory "24/7 staffing that recognizes customer-specific cyber-risk-based use cases" and "immediate remote mitigative response, investigation and containment activities (such as quarantining hosts), beyond alerting and notification, delivered and coordinated by service providers' staff and preapproved by end users" (Gartner).

NIST's incident response guide treats this as one normal way to staff incident handling, for example "outsourcing a security operations center [SOC] to a managed security services provider [MSSP]" (NIST SP 800-61r3). It also names the main benefit: "A service provider may detect malicious activity sooner than individual organizations would because it can correlate events across its customers."

StageWhat the MDR provider doesWhat you still own
CollectTakes telemetry from endpoint agents, network sensors, identity, email and cloud logsDeploying agents and granting access
DetectRuns detection rules and models on its platformTelling the provider what is normal for your business
InvestigateAnalysts triage alerts and hunt for threatsAnswering questions about users and systems
RespondContains threats within preapproved limits, such as isolating a hostApproving actions outside those limits
Recover and reviewReports what happened and recommends fixesRebuilding systems, fixing root causes

How is an MDR provider different from an MSSP or SOC as a service?

In short: a managed security service provider (MSSP) often manages security tools and forwards alerts, while an MDR provider is defined by investigating and acting on threats. SOC as a service is a broader label for renting a SOC, and offerings overlap. The full comparison is on MSSP vs MDR and SOC as a service; this page stays on choosing and contracting an MDR provider.

How do you compare MDR providers?

Ranked lists rarely match your systems. Compare providers against your own estate and your own limits. Ask each one the same questions in writing and score the answers.

Logs deserve their own questions. The Canadian Centre for Cyber Security's checklist for managed-service customers asks "What types of logs can the MSP provide?" and "Can the MSP support and comply with your data retention policy for logs?" (Canadian Centre for Cyber Security). CISA and its partners "recommend all organizations store their most important logs for at least six months", because it can take months to detect an incident (CISA).

What to compareQuestion to askA good answer shows
CoverageWhich of our endpoints, servers, cloud accounts, identity and email will you watch?A list matched to your estate, with gaps named
Response authorityWhich actions do you take without calling us?A written list you approve, such as isolate host or disable account
SpeedHow fast do you acknowledge, investigate and contain, by severity?Times in the contract, with how they are measured
PeopleWho works our alerts at 3 a.m., and how do we reach them?Named escalation path, 24/7 analysts
ReportingWhat does an incident ticket contain?Timeline, evidence, actions taken, what you must fix
Logs and dataWhere are our logs stored, for how long, and how do we export them?Location, retention period, export format
ExitWhat happens to our data when we leave?Return and deletion terms with dates

What does an MDR contract say about your data?

General information, not legal advice. Start with ownership. Managed-service guidance from Canada tells customers: "Your organization is the data owner and is legally responsible for data security," and to ask "who retains legal ownership of the data" when a contract ends or the provider is acquired (Canadian Centre for Cyber Security).

Where EU personal data is involved, the provider acts as a processor. GDPR requires the contract to say the processor "processes the personal data only on documented instructions from the controller" and, "at the choice of the controller, deletes or returns all the personal data to the controller after the end of the provision of services relating to processing" (GDPR Article 28). US contracts usually reach a similar place through confidentiality and data-use clauses.

NIST adds a point many buyers miss. The contract should set "restrictions on what the service provider can do, such as sharing sanitized incident information with other customers" (NIST SP 800-61r3). Cross-customer correlation is part of what you pay for, so decide what the provider may reuse from your incidents, in what form, and say it in writing.

  • Ownership: your logs, alerts, tickets and reports belong to you.
  • Use: the provider may use them to deliver the service; any other use, including sharing anonymized indicators, is listed.
  • Retention: how long each record type is kept, and where.
  • Export: format and timing for logs, tickets and reports on request and at exit.
  • Deletion: when and how the provider deletes your data after the contract ends, with confirmation.
  • Subcontractors: who else processes your data, and where.

Worked example: scoring three MDR providers

This is an illustration written for this page, not real providers or real scores. A 400-person software company runs on laptops, two cloud accounts and a single identity provider. It asks three MDR providers the questions above and scores each answer from 1 to 5, weighted by what matters most to it.

Criterion (weight)Provider AProvider BProvider C
Coverage of our estate (25)543
Response authority we can approve (20)354
Contracted response times (20)435
Data terms: retention, export, deletion, reuse (20)253
Reporting quality from a sample ticket (15)435
Weighted score (out of 5)3.654.053.90

Provider A had the widest coverage but kept logs for only 90 days and reserved the right to reuse incident data across customers without limits. Provider B scored lower on speed but put retention, export and deletion terms in the contract. The company chose B and negotiated a faster containment target. The weights are the company's own; another company would weigh them differently.

Common mistakes when choosing an MDR provider

Most problems appear during the first real incident or at the end of the contract.

  • Choosing from a ranked list without checking coverage of your own systems.
  • Never agreeing which response actions are preapproved, so the provider calls you at 3 a.m. for every containment.
  • Accepting response times that are not in the contract or have no measurement method.
  • Not asking where logs are stored or how long they are kept until an investigation needs them.
  • No exit terms, so getting logs and tickets back means a negotiation.
  • Assuming the provider fixes root causes. MDR contains and reports; your team still patches and rebuilds.
  • No runbooks on your side for the actions the provider asks you to take (see runbook vs playbook).

What records does an MDR service leave behind?

A year of MDR leaves a detailed record: alerts, analyst triage notes, investigation timelines, containment actions and incident reports. Kept together with your own change tickets and fixes, they form the full story of each incident. The incident timeline template shows how to join them by time and ticket ID, and audit log examples covers the logs underneath.

Are MDR alerts, tickets and incident records also a data asset?

It depends whose records they are. AI data buyers have asked Origin for the records of companies that defend and run their own infrastructure: incidents joined to the ticket, chat thread, config change and outcome that fixed them; ordered logs of what happened and what was done; and the SOC tickets, runbooks and postmortems of the same period. The request covers a company's own infrastructure, not a client's unless the client has agreed.

An MDR provider's case records come from client environments. Under GDPR the provider acts "only on documented instructions from the controller" (GDPR Article 28), and managed-service guidance names the customer as the data owner (Canadian Centre for Cyber Security). So the client, as owner, is the one who can decide; a provider can offer only its own internal incidents or client cases with written client agreement. A company that uses an MDR service holds its own incident records and can decide for itself. See cybersecurity incident and SOC case data and security incident data.

Origin is an independent data partner. The company decides what is shared and signs any license itself, and a buyer may pay Origin when a deal closes.

  • Client cases: stay out unless the client has agreed in writing.
  • Secrets and indicators that expose live systems: removed first.
  • Personal data: user names, emails and IP addresses removed or replaced with consistent pseudonyms.
  • Threat intelligence received from others under sharing limits: stays out.

Does your company hold records like these? See what your data is worth.

Common questions

What does an MDR provider do?

An MDR provider runs security monitoring and response for a customer from its own platform. Its analysts watch endpoints, networks, cloud and logs around the clock, investigate alerts, hunt for threats and take containment actions, such as isolating a compromised host, within limits the customer has approved. It then reports what happened and what the customer must fix.

How do I choose an MDR provider?

Write down your systems and your limits first. Ask each provider the same questions: what it will monitor, which actions it takes without asking, its contracted response times, what an incident ticket contains, where logs are kept and for how long, and how you get your data back at exit. Score the answers with weights that match your priorities.

Who owns the data an MDR provider collects?

Usually the customer. Managed-service guidance names the customer as the data owner, and under GDPR the provider is a processor that acts only on the customer's documented instructions and deletes or returns personal data when the service ends. Check the contract for ownership, permitted uses, retention, export and deletion terms.

Can an MDR provider share my incident data with other customers?

Only as the contract allows. Providers correlate events across customers, which is part of their value, and NIST says the contract should set restrictions such as on sharing sanitized incident information with other customers. Ask exactly what is reused, in what form, and whether your company could be identified from it.

How long should an MDR provider keep my logs?

Long enough to investigate incidents found late. CISA and partner agencies recommend keeping the most important logs for at least six months, because detection can take months. Set the retention period in the contract, make sure it matches your own legal and regulatory needs, and confirm how logs can be exported.

Does an MDR provider fix the root cause of an incident?

Usually not. An MDR provider detects, investigates and contains threats, and recommends fixes. Rebuilding systems, patching, changing configurations and closing the root cause normally stay with your team or another service provider, unless the contract adds incident response or remediation work.

Sources and further reading

What is your company’s data worth?

AI companies pay $20K to $5M for company records. Six questions, no files.