What is ServiceNow Discovery?

ServiceNow Discovery is a capability that automatically scans your network and populates the CMDB with what is actually running. Every server, every application, every network device, every cloud resource. It finds them, identifies their relationships, and keeps the records up to date as your environment changes.

Before Discovery, CMDB records were created and updated by people. People are slow, inconsistent, and busy with other things. Within weeks of a manual update, the CMDB drifts from reality. A server gets decommissioned without a ticket. A new application gets deployed without documentation. A cloud instance spins up automatically and nobody records it.

Discovery does not replace human judgment. It replaces human data entry. The difference matters: your team still makes decisions about what to do with infrastructure. Discovery makes sure those decisions are based on accurate information.

How does ServiceNow Discovery work?

Discovery uses a component called the MID Server as the bridge between your infrastructure and ServiceNow. The MID Server is a lightweight Java application that runs inside your network. It receives instructions from ServiceNow, scans your environment, and sends the results back.

The Discovery process runs in four stages.

01	Scanning Discovery sends probes across your network to identify IP addresses that are active. It uses port scanning, SNMP, WMI and SSH depending on the type of device and the credentials available. Cloud environments are scanned via API rather than network probe.
02	Classification Each active device is classified by type: Windows server, Linux server, network switch, load balancer, database, application server. ServiceNow uses patterns to identify what a device is and which additional probes to run against it.
03	Identification Discovery collects detailed information about each device: hostname, IP address, operating system, installed software, running processes and the relationships between them. This is the data that becomes CMDB records.
04	CMDB population Discovered items are created or updated in the CMDB. ServiceNow uses identification and reconciliation rules to decide whether a discovered item matches an existing record or is something new. Duplicate records are avoided through these rules.

What is a ServiceNow MID Server and why does Discovery depend on it?

The MID Server sits inside your network because ServiceNow cannot reach your internal infrastructure directly from the cloud. It acts as a secure, authenticated bridge: ServiceNow tells the MID Server what to scan, the MID Server accesses your devices using stored credentials, and returns the results.

Where the MID Server runs On a dedicated Windows or Linux server inside your network. It needs network access to the devices Discovery will scan and outbound HTTPS access to your ServiceNow instance.

How many MID Servers you need One MID Server can typically handle a mid-size environment. Large or distributed environments need multiple MID Servers, one per network segment or data centre, to ensure coverage and performance.

Credentials and security Discovery requires read-only credentials for each device type it scans: Windows admin credentials for WMI, SSH credentials for Linux, SNMP community strings for network devices. These are stored encrypted in ServiceNow and never leave the MID Server in plain text.

What does ServiceNow Discovery find?

Discovery covers all major infrastructure types. The exact scope depends on which patterns and credentials are configured, but a standard implementation discovers:

Servers and operating systems Windows, Linux, Unix. Hostname, IP, OS version, installed software, running processes, hardware specs.	Network devices Switches, routers, load balancers, firewalls. Model, firmware, interfaces, VLANs and connections between devices.
Applications and databases Web servers, application servers, databases. Running instances, version, configuration, connections to other services.	Cloud resources AWS, Azure, Google Cloud. Virtual machines, containers, storage, networking. Discovered via cloud APIs without requiring a MID Server in the cloud environment.

Want to see what Discovery would find in your environment? We run Discovery implementation projects for mid-size and enterprise organisations across Europe. We configure MID Servers, define scan schedules, set up credentials and validate CMDB accuracy after the first Discovery run. Talk to a Discovery specialist →

How to implement ServiceNow Discovery: a step-by-step guide

Discovery implementation is a structured process. Here is how we approach it with clients.

Step 1	Define scope and credentials Before Discovery runs a single scan, you need to define what it will scan and what credentials it will use. Map your IP ranges, identify your infrastructure types and gather the read-only credentials for each device category. The quality of this step determines how much Discovery will find.
Step 2	Install and configure MID Servers Install MID Servers in each network segment that Discovery needs to reach. Configure them in ServiceNow, validate connectivity, and test that they can reach the devices in scope. MID Server placement significantly affects Discovery coverage and performance.
Step 3	Run a pilot Discovery Run Discovery against a limited IP range first. Review what was found, what was missed and what errors appeared. This pilot identifies credential gaps, network access issues and any patterns that need customisation before you run Discovery across your full environment.
Step 4	Configure identification and reconciliation rules Identification rules tell ServiceNow how to match a discovered item to an existing CMDB record. Reconciliation rules determine which data source wins when Discovery conflicts with data from another source. Getting these rules right prevents duplicate records and data quality problems.
Step 5	Run full Discovery and validate CMDB Run Discovery across your full environment. Review the results: compare discovered records against known infrastructure, identify gaps and verify relationship data. CMDB accuracy should improve significantly after the first full run.
Step 6	Set up ongoing scan schedules and governance Discovery is not a one-time run. Configure scheduled scans that keep your CMDB current as infrastructure changes. Define ownership for Discovery results: who reviews errors, who handles discovered items that cannot be automatically classified, and who manages MID Server health.

What are Service Graph Connectors?

Service Graph Connectors are pre-built integrations that bring data from third-party tools directly into the ServiceNow CMDB. Where Discovery scans your network to find infrastructure, Service Graph Connectors pull structured data from sources that already have it.

Examples of Service Graph Connectors AWS, Azure, Google Cloud, VMware vSphere, Microsoft SCCM, Jamf, CrowdStrike, Qualys, Dynatrace, Splunk. Each connector maps data from the external tool to the correct CMDB classes and attributes using ServiceNow's standard data model.

When to use connectors vs Discovery Use Discovery for infrastructure that does not have another authoritative data source. Use Service Graph Connectors when a tool already has accurate, structured data about a class of assets. Most organisations use both: Discovery for general infrastructure, connectors for cloud and endpoint management tools.

ServiceNow Discovery implementation: questions we get asked

How long does a ServiceNow Discovery implementation take? A standard Discovery implementation takes 6 to 10 weeks from scope definition to full production scan. The largest variable is credential availability and network access: environments with well-documented credentials and clean network segmentation move faster. Complex multi-site environments with many device types take longer.

Will Discovery overwrite existing CMDB data? That depends on how reconciliation rules are configured. By default, Discovery data takes precedence for attributes it discovers. You can configure rules to protect manually entered data for specific attributes. This is a key configuration decision that should be made before the first full scan.

Does Discovery work in cloud-only environments? Yes. For public cloud environments (AWS, Azure, GCP), Discovery uses cloud APIs rather than network scanning. No MID Server needs to be deployed in the cloud environment itself. For private cloud or hybrid environments, a MID Server in each network segment provides coverage.

What CMDB accuracy can we expect after implementing Discovery? In our experience, organisations that implement Discovery correctly move from CMDB accuracy of 40 to 60% to above 85% within the first three months. The remaining gap is typically in areas Discovery cannot reach due to credential or network access limitations, or in asset classes that require a Service Graph Connector rather than network scanning.

Can Discovery replace our existing asset management tool? For IT infrastructure assets, yes. Discovery combined with ServiceNow ITAM (IT Asset Management) covers the full asset lifecycle from discovery through retirement. If you currently maintain a separate CMDB or asset management tool, Discovery is the path to consolidating that data into ServiceNow and keeping it current automatically.

What changes after Discovery is running

The change is not immediate and dramatic. It is gradual and structural.

In the first weeks, your team sees infrastructure they did not know existed. Servers that were decommissioned on paper but still running. Applications with no owner. Cloud instances spinning costs that nobody was tracking.

Over the following months, incident resolution gets faster because the CMDB is accurate enough to trust. Upgrade projects become less risky because you know what you are upgrading. Service Mapping becomes possible because Discovery has already built the infrastructure layer it depends on.

The teams that get the most out of Discovery are the ones that treat it as infrastructure, not a project. It runs continuously, feeds every other ITOM capability, and quietly keeps your CMDB honest while your team focuses on everything else.

Key takeaways •  ServiceNow Discovery automatically scans your network and populates the CMDB with what is actually running. No manual data entry. •  Discovery uses a MID Server inside your network as the bridge between your infrastructure and ServiceNow. •  A standard implementation takes 6 to 10 weeks. The largest variable is credential availability and network access. •  Organisations move from 40 to 60% CMDB accuracy to above 85% within the first three months of Discovery running correctly. •  Service Graph Connectors complement Discovery by bringing structured data from cloud platforms, endpoint tools and monitoring systems directly into the CMDB.

Ready to see what is actually running in your environment? We implement ServiceNow Discovery for organisations across Europe. MID Server installation, credential configuration, scan schedule setup and CMDB validation included. Start with a platform health check to understand your current CMDB accuracy before Discovery runs. Get a Discovery implementation →

HappyMan Solutions
Your system works. Your team is happy. That's the job.