Roster
Scheduled Oct 2, 2026Every asset and every licence, in one database you own.
Replaces
- Lansweeper · Lansweeper NV — $199 per org per month (Starter (2,000 assets, billed annually)), about $2,388/yr at 400 seats
- ServiceNow IT Asset Management · ServiceNow, Inc. — no published list price
Rented, per year
$2,388
Lansweeper list price at 400 seats.
Owned, per year
$4,920
AWS infrastructure only. Azure is $5,340.
Build
8 weeks
2 engineers. To parity on the ten core features, including the on-site collector, cloud and IdP inventory sources, and normalisation rules covering the top 200 software titles in the customer's estate.
Difficulty
3/5
Well-understood shape. Low delivery risk.
Reference organisation: 400 employees, ~2,400 discovered assets across 6 sites, 180 normalised software titles, 60 licence agreements, 5 years of asset history retained, single region, prod + non-prod. 3-year Compute Savings Plan on Fargate / 3-year reservation on Container Apps. Aurora minimum 0.5 ACU with non-prod stopped outside business hours. On-site collectors run on existing customer hardware and are not costed here. Excludes Azure Arc management add-ons and any per-server Update Manager charges. These figures are infrastructure only and exclude the build.
What it is
Roster discovers what you have, reconciles it against what you bought, and tells you what you are paying for and not using. It covers hardware and cloud instances, installed software, SaaS subscriptions, licence entitlements, contracts, renewal dates and assignment to people and cost centres. It is aimed at organisations of a few hundred to a few thousand assets where the asset database has become the thing three other systems disagree with, and where the tool bought to find licence waste has itself become an unpriced renewal.
Why this one stops making sense
- ServiceNow does not publish a price for ITAM, and it is licensed on top of a platform subscription. You are negotiating two contracts against a vendor who can see how much of your estate already depends on them.
- An asset register is a join across four systems you already own: the IdP, the MDM, the cloud APIs and the finance ledger. It is a data problem, not a product problem.
- The reconciliation logic — what counts as an installation, how a licence metric maps to a user, which cost centre owns a device — is specific to your contracts, and in a SaaS tool it lives in configuration you cannot diff or test.
- Asset data has a long tail of one-off questions: which laptops are out of warranty in Munich, which Adobe seats have not launched in 90 days. With Postgres you write SQL. With a SaaS tool you wait for a report builder.
- Discovery credentials are domain-wide read at minimum. Handing those to a hosted scanner is a real risk decision that a self-hosted collector removes.
What ships
- Agentless network discovery by SNMP, WMI, SSH and mDNS, plus cloud inventory from AWS, Azure and your IdP
- Software installation inventory with title normalisation and version rollup
- Licence entitlement register: metrics, quantities, true-up dates, and proof-of-purchase attachments
- Reconciliation engine producing an effective licence position per publisher, per metric
- SaaS subscription discovery from SSO sign-in logs and expense feeds
- Usage-based reclaim: flags seats not used in a configurable window, with a one-click revoke through the IdP
- Assignment model linking asset to person, cost centre, location and contract
- Contract and warranty register with renewal alerting at 120, 60 and 30 days
- Full asset lifecycle: procured, in stock, deployed, in repair, retired, disposed, with a disposal certificate record
- Historical state as a slowly-changing dimension, so you can answer what the estate looked like on any past date
Data model
Stack
- frontend
- Next.js 15 App Router + Tailwind + TanStack Query + TanStack Table for large grids
- backend
- NestJS on Node 22 for the API; a Go collector binary for network scanning
- database
- PostgreSQL 16, asset history as an SCD-2 table with tstzrange validity and a GiST exclusion constraint
- cache
- Redis 7 for scan coordination and grid aggregates
- queue
- BullMQ for discovery fan-out, reconciliation runs and renewal digests
- auth
- OIDC against the customer IdP; SCIM for person records; collectors authenticate with mTLS
- search
- PostgreSQL tsvector plus pg_trgm for fuzzy software title matching
- ai
- Optional: software title normalisation suggestions and contract clause extraction through a pluggable LLM gateway
Parity, honestly
What we match, and what we do not.
The rows marked No are the important ones. If one of them is the reason you bought Lansweeper, keep buying it — and we will tell you that on the first call rather than the fifth month.
| Capability | Lansweeper | Ours | Note |
|---|---|---|---|
| Agentless network discovery (SNMP / WMI / SSH) | Yes | Yes | — |
| Cloud resource inventory (AWS, Azure) | Yes | Yes | — |
| Installed software inventory | Yes | Yes | — |
| Licence entitlement vs. consumption reconciliation | Yes | Yes | Ours is SQL you can read, against contracts you enter once. |
| SaaS spend and unused-seat detection | Partial | Yes | Driven from your IdP sign-in logs, which is more reliable than a browser extension. |
| Contract and warranty renewal alerting | Yes | Yes | — |
| Point-in-time historical estate reporting | Partial | Yes | Every asset row is versioned, so "what did we own on 31 March" is a query, not an archive request. |
| Arbitrary reporting over the full dataset | Partial | Yes | Read-only Postgres role for your BI tool. No report builder in the way. |
| Vendor-maintained software recognition catalogue | Yes | No | This is the biggest honest gap in Roster. ServiceNow and Flexera maintain normalisation libraries covering hundreds of thousands of software titles and their licence metrics, built over two decades. Roster ships normalisation rules for the 150 to 250 titles that matter in your estate and a rule editor for the rest. If your licensing exposure is a genuine long tail across thousands of publishers, that catalogue is worth paying for. |
| Certified discovery patterns and dependency service mapping | Yes | No | ServiceNow Discovery ships thousands of vendor-certified patterns and maps application dependencies into a CMDB. Roster inventories assets; it does not build a service dependency graph. If your driver is CMDB-backed change impact analysis, that is a ServiceNow problem, not a Roster one. |
| Device fingerprint library for unusual network hardware | Yes | Partial | Lansweeper's OUI and fingerprint database is twenty years deep. Roster identifies the common estate and marks the rest as unclassified with its raw SNMP response attached, so you can classify it once and keep the rule. |
| Publisher audit defence support | Partial | No | Neither product defends you. But the vendors sell advisory services alongside; we hand over an evidence-grade dataset and, if you want it, an FDE for the audit window. |
| ITSM ticketing built in | Yes | No | Roster integrates with your service desk over its API. It does not become one. |
Reference architecture
Built twice, on purpose.
Every application in Techtons ships with a production architecture for AWS and one for Azure, using the same diagram grammar so you can read them side by side. Your cloud, your account, your bill.
Amazon Web Services
- Compute
- ECS Fargate (api, reconciler, worker) behind an ALB; discovery collectors run outside AWS
- Data
- Aurora Serverless v2 PostgreSQL
- Storage and edge
- S3 for contract PDFs, purchase evidence and raw discovery payloads
- Identity
- Cognito federated to the customer IdP; cloud inventory via read-only cross-account roles
- Observability
- OpenTelemetry to CloudWatch; a stale-collector alarm per site
- Infrastructure as code
- Terraform, with the collector packaged as a signed container image and a systemd unit
Services
The decision worth arguing about
The non-obvious decision concerns where discovery actually runs, and it is the decision most asset-management architectures get quietly wrong. The obvious answer on AWS is Systems Manager Inventory: the SSM Agent is already on your EC2 fleet, inventory collection is free, and it gives installed-software data without writing a scanner. We do use it — for everything that is an SSM managed instance. But a Fargate task cannot see the corporate LAN, and the assets that generate licence exposure and audit findings are disproportionately the ones nobody manages: an unpatched Windows box under a desk, a switch, a label printer, a lab machine. So Roster ships a small Go collector that runs on customer hardware inside each broadcast domain, scans locally, and pushes signed batches out over mTLS to an ingest endpoint. The trade-off is that we have reintroduced an on-premise footprint into an otherwise fully managed deployment — one container per site to patch, monitor and certificate-rotate — and we would rather say that clearly than pretend a cloud-only asset tool can discover a network it cannot reach. The mitigation is that the collector is stateless, holds no credentials beyond a short-lived client certificate issued through the IoT Core credentials provider, and a site going dark raises an alarm rather than silently shrinking the asset count.
Microsoft Azure
- Compute
- Azure Container Apps (api, reconciler, worker); collectors identical, running on site
- Data
- Azure Database for PostgreSQL Flexible Server
- Storage and edge
- Blob Storage for contracts and raw discovery payloads, cool tier after 60 days
- Identity
- Microsoft Entra ID; cloud inventory via a managed identity with Reader across subscriptions
- Observability
- Azure Monitor and Application Insights
- Infrastructure as code
- Bicep, plus an Arc onboarding script for the server estate
Services
The decision worth arguing about
Azure changes the discovery calculus in one specific way: Azure Arc. Onboarding a non-Azure server to Arc is free for inventory and policy, and it gives you a Resource Graph row for a machine sitting in a datacentre in Lyon exactly as if it were an Azure VM. That is genuinely better than the AWS equivalent and it removes a meaningful share of the server-side scanning we would otherwise do. The non-obvious part is knowing where Arc stops. Arc is machine-centric: it inventories servers, not switches, printers, badge readers, lab instruments or anything without an operating system you can install an agent on, and several of the Arc-enabled server management features — Update Manager for machines outside Azure, for instance — are billed per server per month rather than being part of the free tier. So the shape we deploy is a hybrid: Arc for servers, our own on-site collector for everything else, and a reconciliation step that dedupes a machine appearing in both by hardware UUID and MAC. The trade is a second inventory source with its own failure mode — a machine can be Arc-connected and scanner-invisible or the reverse — which is why the data model records DiscoverySource per asset version rather than flattening to one truth. When two sources disagree, Roster shows you both.
Sources
Every price on this page, with the page we read it from and the date we read it.
| Product | Plan | List price | Checked | Source |
|---|---|---|---|---|
| Lansweeper | Starter (2,000 assets, billed annually) | $199 / org/month | 2026-09-05 | www.lansweeper.com/pricing/ Lansweeper publishes Starter at $199/month billed annually including 2,000 assets, and Pro at $379/month for the same asset count expandable to 9,000. Above 10,000 assets the Enterprise tier is quoted, not listed. |
| ServiceNow IT Asset Management | — | Not published | 2026-09-05 | www.servicenow.com/lpgp/pricing-it-asset-man ServiceNow publishes no list price for IT Asset Management. Its own pricing page offers only "custom quotes" and a route to a sales representative. ITAM is additionally licensed on top of a Now Platform subscription, so any number quoted for it in isolation is incomplete. |
Should you still be paying for Lansweeper?
Two weeks, fixed price. We audit what you actually use, map it against Roster, cost the replacement on your own AWS or Azure account, and give you a delivery plan. If the honest answer is to keep the licence, that is what the report will say.