The EAM Buyer's Guide: Evaluating Platforms for Linear and Discrete Portfolios

Who this guide is for

This guide is for asset managers, operations directors, IT leaders, and procurement teams evaluating Enterprise Asset Management platforms. It is written specifically for organisations that manage mixed portfolios — a combination of linear assets (pipelines, rail, water networks, roads) and discrete assets (manufacturing equipment, facilities, fleet).

If you manage only discrete assets, most major EAM platforms will serve you adequately and the selection decision comes down to integration fit, vendor support, and price. If you manage only linear assets, a specialist platform built around Linear Referencing Systems and GIS may be the right answer.

If you manage both — which is the default for most utilities, transport authorities, and infrastructure operators — this guide is specifically for you. The two-system problem is real, the cost is significant, and most vendor shortlists are built without a framework that surfaces it.

The two-system problem

Traditional EAM platforms were designed for discrete assets. A pump, a transformer, a vehicle — things with a nameplate, a serial number, and a fixed location. The data model is a hierarchy. The work order is attached to an asset ID.

Linear assets — a pipeline, a rail corridor, a water main — do not have a fixed location within themselves. A 40km pipeline is one asset that has different material, condition, pressure zone, soil type, and criticality at every kilometre. The right data model is a network with a Linear Referencing System, not a hierarchy with an ID.

Most platforms that started as discrete EAMs added linear capability later. The seam shows: LRS is a bolt-on module, spatial queries require an export to GIS, and reporting aggregates at asset level when regulations require segment-level summaries.

The consequence for mixed-portfolio operators: two systems, two data models, two sets of IDs, two reporting pipelines, and a recurring reconciliation exercise just to produce a unified asset register. That cost — in licence fees, integration maintenance, and data quality erosion — is the problem this guide helps you evaluate before you sign.

Phase 1: Build the right shortlist

Define your asset portfolio before talking to vendors

Before any vendor conversation, document:

• What percentage of your assets are linear vs discrete?
• Which regulatory regimes apply? (PHMSA IMP, AWWA AMP, FRA Track Class, NERC TPL, ISO 55000)
• What is your current GIS stack? (Esri ArcGIS, QGIS, other)
• What ERP do you run? (SAP, Oracle, Microsoft D365, other)
• What is your field crew's current mobile workflow?
• How many users need the system, and in what roles?

This documentation serves two purposes: it grounds your requirements before vendors shape them, and it gives vendors what they need to demo against your real context rather than a generic sandbox.

Shortlist criteria for mixed-portfolio operators

For a platform to belong on the shortlist of a mixed-portfolio operator, it must pass these gates:

Gate	What to verify
Native LRS	Can a single asset carry a route, a length, and a stationing system without child-record workarounds?
Native GIS integration	Is the GIS a peer system with two-way sync, or a read-only map view?
Discrete hierarchy	Does the platform support ISO 14224 or equivalent equipment hierarchy?
Shared asset register	Are linear and discrete assets in the same data model with shared IDs, not federated across systems?
AI/agent capability	Does the AI take actions (write to the system of record) or only answer questions?
EU data residency	For GDPR-regulated operations: where does the data live?

A platform that fails any of these gates for your portfolio type does not belong on a mixed-portfolio shortlist, regardless of market position.

Phase 2: Structure your vendor evaluations

The demo script that separates real capability from positioning

Generic demos show polished scenarios. Your goal is to see the system behave on your actual asset types. Before each demo, provide the vendor with:

1. A sample of your actual asset hierarchy (anonymised if needed)
2. A linear asset example: a pipeline segment, a road section, a water main with known break history
3. A discrete asset example: a pump station, a treatment plant, a depot
4. A representative work order from your current system

Then run these scenarios:

Scenario 1: Linear segment defect to work order. A field technician finds a corrosion indication on a pipeline at a specific chainage. How does the system capture it? Does the work order carry the precise stationing? Is the relevant segment's risk score updated? How long does it take?

Scenario 2: Discrete asset PM schedule. Show the preventive maintenance schedule for a pump station across the next 90 days. How are crew availability and equipment criticality balanced? Can the system explain why a PM was scheduled for a particular date?

Scenario 3: Cross-asset impact analysis. A compressor at a network node is flagged for maintenance shutdown. Which downstream pipeline segments are affected? What is the isolation strategy? This question directly tests whether the platform can reason across asset types in the same context — or whether it requires a switch to a different screen, a different module, or a different system.

Scenario 4: Agent action, not copilot answer. Ask the AI to create or modify a work order. Watch whether it writes to the system of record or produces a text output for a human to transcribe.

Twelve questions every vendor must answer

1. Is linear asset management a native capability or a licensed module added to a discrete EAM?
2. What Linear Referencing System standard does the platform support (LRS-based, chainage, mileposts)?
3. What GIS platforms does the system integrate with, and is it two-way sync or one-way export?
4. Can a single work order carry both a linear location (from/to stationing) and an equipment hierarchy position?
5. What regulatory compliance reports are pre-built, and which require custom configuration?
6. What does "AI" mean in your platform — descriptive, predictive, prescriptive, or agentic?
7. Can the AI take autonomous action (create, modify, close records), and what is the audit trail for those actions?
8. What is the data model for mixed linear/discrete portfolios — one unified register or two federated schemas?
9. Where does data reside, and what certifications (ISO 27001, SOC 2, GDPR) does that hosting carry?
10. What is the typical implementation timeline for an organisation of our size and portfolio mix?
11. What does the pricing model look like at our scale — per user, per asset, per module?
12. Who are three reference customers with a portfolio mix similar to ours?

Red flags in vendor responses

"Linear is on our roadmap." It is not in the product. The timeline is uncertain. Plan for discrete-only capability in the near term.

"We have a GIS integration." Ask what "integration" means. If it is a scheduled data export or a read-only map overlay, it is not a peer integration. Ask for a live two-way sync demonstration.

"Our AI predicts failures." Predictive AI is table stakes. Ask what happens after the prediction — does the system recommend an action, and can it execute that action? If the answer is "we show the risk score and the maintenance team decides," that is analytics, not an agent.

"We handle all asset types." Ask to see a work order that carries a linear location and a discrete equipment hierarchy position simultaneously. If the demo switches screens, modules, or systems, the claim is overstated.

"Our reference customers are in [a different sector]." Regulatory requirements, asset data models, and operational workflows vary significantly by industry. A reference customer in manufacturing is not evidence of capability for a water utility managing both a distribution network and treatment plants.

Phase 3: Evaluate total cost of ownership

The costs that don't appear in the licence price

Licence cost is the most visible number in a vendor comparison. It is often the least predictive of total cost. The numbers that determine total cost of ownership over a 5-year contract:

Implementation consulting. Enterprise EAM implementations for complex portfolios routinely run 1–3x the first-year licence cost in consulting fees. Ask for a fixed-scope implementation quote, not just a day-rate estimate.

Data migration. Asset data in a legacy system is rarely in a form that a new platform can ingest directly. Allow for data cleansing, schema mapping, and validation — this is often underestimated by a factor of two.

Training and change management. A system that field technicians do not use effectively is not a system. Budget for role-specific training, not generic product training.

Integration maintenance. If the EAM integrates with an ERP, a GIS, and a SCADA system, those integrations require maintenance as each system is upgraded. Ask who maintains them and at what cost.

Module licensing. Platforms that charge separately for linear capability, AI features, mobile access, or compliance modules will cost more than the base licence suggests. Get a quote that includes everything you need, not just the base product.

Vendor stability and product roadmap

A platform you will run for 10 years needs a vendor likely to still be investing in it in year 5. Questions worth asking:

• How many customers does the platform have, and what is the annual churn rate?
• What percentage of revenue is reinvested in product development?
• What does the 18-month product roadmap look like, and which items are committed versus aspirational?
• What is the acquisition risk — is this a platform that a larger player is likely to absorb, and what would that mean for the roadmap?

Phase 4: EU-specific considerations

GDPR and data residency

For operators in the EU, data residency is a procurement requirement, not a preference. Asset management systems hold operational data that may include personally identifiable information (technician records, contractor data, customer service records). The relevant questions:

• Where are the production servers located?
• Where are backups stored?
• Is there sub-processor data processing outside the EU, and if so, what is the legal basis?
• What Data Processing Agreement does the vendor provide?

US-headquartered vendors can comply with GDPR via standard contractual clauses and EU-hosted infrastructure, but this requires explicit confirmation — not assumption.

NIS2 alignment

The NIS2 Directive applies to operators of critical infrastructure including water, energy, and transport. For EAM platforms managing these assets, relevant considerations include:

• Incident detection and reporting capabilities built into the platform
• Access control and authentication standards
• Supply chain security (what third-party components does the EAM vendor use?)
• Asset inventory capabilities that map to NIS2's asset classification requirements

NIS2 alignment is not a certification — it is a set of architectural and process requirements. Ask vendors to walk through how their platform supports the specific article requirements relevant to your sector.

The evaluation scorecard

Use this framework to score each shortlisted vendor across the dimensions that matter for your portfolio:

Dimension	Weight	Vendor A	Vendor B	Vendor C
Native linear capability	20%
Native discrete capability	15%
Unified data model	15%
Agentic AI capability	15%
GIS integration quality	10%
Implementation speed	10%
Total cost of ownership	10%
EU data residency	5%

Adjust weights for your portfolio. A pure-linear operator should weight native linear capability at 35%. A pure-discrete operator can drop it to 5%.

Further reading

• Linear asset management: what your EAM actually needs — covers LRS, GIS integration, segment-level criticality, and cascade modelling in depth.
• Agentic AI in EAM — covers the difference between analytics, copilots, and agents, and what to look for in a platform that is genuinely agent-native.
• How UniEAM compares to Maximo, SAP, IFS, and Hexagon — applies this framework to the five platforms most commonly shortlisted by mixed-portfolio operators.

FAQ

How long should an EAM evaluation take? A structured evaluation for a complex mixed-portfolio operator typically takes 3–6 months: 4–6 weeks for requirements definition and shortlisting, 6–8 weeks for structured vendor evaluations and demos, 4–6 weeks for reference checks, contract negotiation, and final decision. Rushing the evaluation usually surfaces problems in implementation.

Should we involve IT or operations in the decision? Both. IT evaluates integration architecture, data security, and hosting. Operations evaluates workflow fit, field usability, and reporting. Platforms that score well on IT criteria but poorly on operations usability tend to be rejected by the field and generate low adoption. Involve front-line maintenance planners and field supervisors in the demo evaluation.

What happens if we pick the wrong platform? EAM platform migrations are expensive — typically 2–4 years of disruption and significant consulting cost. The cost of a poor selection decision compounds for the life of the contract. Invest in the evaluation; it is significantly cheaper than a migration.

Is open-source EAM a viable option? For simple discrete asset management, open-source platforms (OpenMAINT, Asset Panda) can work at small scale. For mixed-portfolio operators with compliance requirements, regulatory reporting needs, and agentic AI requirements, open-source options currently lack the maturity to compete with purpose-built platforms. Evaluate them honestly against your requirements rather than on cost alone.