Starting point, why this matters now

• What's the primary reason you agreed to this conversation today? Options: Reduce unplanned downtime, Lower emergency maintenance spend, Improve technician efficiency, Prove predictive value for leadership, Other
• How many unplanned downtime events does your site average per quarter? Options: 0-1, 2-5, 6-10, 11-20, 20+
• Which asset classes at your site create the most costly unplanned stops (pick up to three)? Options: Rotating equipment (motors, pumps), Compressors/turbines, Electrical distribution, Heat exchangers/boilers, Control valves/actuators, Conveyors/belt systems, Other
• Estimate the average direct cost per day of a full-site shutdown in your operation Options: <$50k, $50k–$200k, $200k–$1M, >$1M, Unknown
• Who typically signs off on capital or operational investments to reduce downtime in your organization? Options: VP Reliability/Maintenance, Plant Manager, CFO/Finance, Operations Director, Shared committee, Other
• Could leadership require a demonstrated ROI within six months, and if so, could you commit to a pilot scope that aims to prove it? Options: Yes, we could commit, Yes, with executive conditions, No, six months is unrealistic, Unsure

What a shutdown really costs

• Walk me through the last unplanned shutdown that cost you more than $100k, what failed and why?
• How often do those root causes repeat across the same equipment types or across different shifts? Options: Frequently (monthly), Occasionally (quarterly), Rarely (yearly), One-off
• And when a failure repeats, who in operations, reliability, and IT ends up owning the corrective plan and delivering the root-cause fixes?
• Which spare-parts shortages, permitting windows, or external vendors lengthen repair time most often? Options: Critical spare not on site, Long lead procurement, Permit/lockout delays, Contractor availability, Other
• If a solution cut that failure class by half within six months, what would stop leadership from approving a pilot immediately?

Where current processes silently break down

• What single maintenance habit or workflow do you see that most reliably leads to repeat failures?
• Describe how work orders are raised today after an alarm, a predictive signal, or a technician observation.
• Tell me which CMMS fields, approval steps, or mobile behaviors technicians most often skip and why.
• How long, on average, passes between a predictive alert and a field crew arriving to investigate? Options: <1 hour, 1–4 hours, 4–24 hours, 24+ hours, Depends on shift
• Who would need to commit time or headcount for us to change that habit at your site, and could they clear that commitment within 30 days? Options: Yes, available now, Yes, with reallocation, No, requires hiring, Unsure

The data and pipes that must work

• If your historian or SCADA cannot export clean, timestamped signals for the pilot assets, what stops you from fixing that before we begin?
• List the systems we must integrate with from the categories below Options: Plant historian, SCADA/PLC, ERP/CMMS, Mobile technician app, IoT gateway/sensor network, API/enterprise middleware, Other
• Name the role that owns API credentials and indicate whether that role can provide a test account for integration work.
• On a scale, how would you rate historical sensor data quality for the pilot assets Options: High — continuous, tagged, validated, Moderate — gaps and noise, Low — lots of gaps or missing timestamps, Unknown
• Would you pause the pilot or proceed with a scoped cleanup budget if data quality requires more than four weeks of cleansing? Options: Pause until cleaned, Proceed with scoped cleanup budget, Scope a smaller pilot instead, Unsure

Where deployments commonly stall

• What single site constraint has killed a previous pilot in your experience?
• How many full-time staff does your site have dedicated to integrations and data engineering? Options: 0, 1-2, 3-5, 6-10, 10+
• List any permits, lockout/tagout rules, or union agreements that could change rollout windows.
• Do you have spare sensors on hand for the pilot, or will procurement be required? Options: Sufficient spare sensors on site, Some spares, procurement likely, Procurement required for all, Unsure
• How quickly can site access and maintenance windows be granted for an integration team to work on live equipment? Options: Immediately, Within 1–2 weeks, Within 3–4 weeks, Longer than 4 weeks
• Name the constraint that, without resolution within three weeks, would force you to cancel or postpone the pilot.

Other paths you might take

• What would have to be true about your incumbent or internal approach for you to keep it instead of switching to an external solution?
• Select which of the following alternatives you are evaluating or have used recently Options: Existing incumbent vendor, In-house build with data science team, Local systems integrator with custom scripts, Point predictive tool plus manual work-orders, Do nothing, continue reactive maintenance, Other
• Identify the role or team that has proposed solving this problem internally rather than hiring an outside partner.
• What would have to be demonstrably true about your current approach for you to decide to stay with it for another 12 months?
• Could anything prevent you from moving to a paid rollout within 30 days if the pilot outcomes matched your targets, and if so, name the blocker? Options: Budget approval lag, Procurement/contracting timeline, Operational readiness gaps, Executive prioritization, Other

How you'll prove this worked

• What's the simplest measurable signal that would prove this saved you money or reduced risk?
• Pick the primary KPIs we'll measure during the pilot Options: Reduction in unplanned downtime hours, MTBF increase, Reduction in emergency maintenance spend, Predictive alert precision (false positives), Technician time saved per week, Other
• Provide the numeric target for your primary KPI, and the timeframe for achieving it (for example, reduce downtime by 20% in 6 months).
• Is the final approver prepared to review results within two weeks of pilot close, and if not, what is the expected sign-off window? Options: Yes, within two weeks, Will need 3–4 weeks, Longer than a month, Unsure
• Under what circumstances would you still move to a full rollout if the pilot misses the target by more than 10%?

Can this actually be delivered here

• Point to the one integration or site gap that would make this project impossible for your IT or OT teams to support.
• Select the on-site roles that will support deployment (choose all that apply) Options: Reliability engineer, Maintenance supervisor, IT systems owner, Field technicians, Data engineer, Other
• Will you assign a named owner for data extraction, and can that person commit at least 8 hours per week during the pilot? Options: Yes, assigned and available, Yes, but limited hours, No, not assigned yet, Unsure
• Is there a gating compliance review, legal approval, or insurance signoff that must happen before work on live assets? Options: Yes — compliance review, Yes — legal/contract, Yes — insurance/signoff, No gating approvals, Unsure
• Are APIs for the historian and your CMMS documented and accessible within your network, or will we require escorted thin-client or vendor-assisted access? Options: APIs documented and accessible, APIs undocumented but accessible, No APIs, DB access required, Only vendor-assisted access available, Unknown
• Given key constraints such as data access, named owners, and site windows, choose whether you prefer to pause the engagement, scope a phased approach, or fund an accelerated readiness effort Options: Pause until ready, Scope a phased pilot, Fund an accelerated readiness effort, Undecided

How we get to a pilot decision

• Assuming the pilot proves the savings you expect, what internal approvals remain before procurement can begin?
• Choose the preferred pilot duration and scope for decision-making Options: 6 weeks focused on 1 asset class, 12 weeks multi-asset pilot, 90-day phased roll, Other
• Identify the sponsor role that can sign a letter of intent or pilot funding within two weeks of acceptable proof.
• What budget range has been preliminarily allocated for a pilot, including any expected hardware or third-party integration costs? Options: <$25k, $25k–$75k, $75k–$200k, >$200k, Not yet allocated
• Pick the contracting path you would elect if the pilot meets targets, purchase order, SOW amendment, or enterprise agreement? Options: Purchase order, SOW amendment, Enterprise agreement, Other
• What concerns about timeline, cost, or operational impact remain unresolved after this conversation?

• Solution Experience Session

• Confirm the current state and its cost
• You confirm the documented current state and accept the quantified cost as accurate.
• Deliver a mapped workflow document showing predictive-alert-to-work-order mappings and the acceptance signal for the prioritized asset class.
• You confirm the demonstrated workflow would stop the specific rework and delays identified in the current state for the sampled asset class.
• Translate a reliability target into an operational scenario
• Provide an extract of recent historian data and the asset registry for the prioritized asset class to enable a sample run.
• You agree on the remaining data and access gaps that must be closed to prove the workflow in a pilot.
• Walk the predictive-to-work-order workflow for a sample asset class
• Confirm named operational owner and technician contact who will participate in the pilot validation steps.
• Validate integrations and data dependencies
• Schedule a pilot validation window for the sample asset class including time for a live run and model tuning.
• Confirm fit with your operations

• Solution Experience Session
• Solution Experience Deck
• Solution Brief

• meeting
• slides
• document

• Deploy Asset Registry and Hierarchy
• Migrate and Clean Historical Work Order Data
• Integrate Plant Historian for Condition Signals
• Integrate ERP for Parts and Work Orders
• Connect SCADA for Real-Time Telemetry
• Install Field Technician Mobile App with Offline Sync
• Configure Preventive and Predictive Maintenance Schedules
• Train and Deploy Predictive Models for Asset Classes
• Configure Condition-Based Alerts and Notifications
• Implement Reliability-Centered Maintenance (RCM) Analytics
• Build Failure Mode and Effects Library
• Optimize MRO Spare-Parts Inventory and Replenishment

Deploy Asset Registry and Hierarchy

• Do you have an existing asset registry stored in a CMMS or spreadsheets that we will import? Options: Yes, No
• Name the identifier scheme used for asset tags we must preserve (tag number, serial, custom code). Options: Tag number, Serial number, Custom code, Other
• How many assets will be in scope by asset class (pumps, compressors, motors, transformers)? Options: Less than 100, 100-500, 500-2,000, More than 2,000
• List the plant-area codes, unit IDs or functional locations that must map to the new hierarchy (for example: Area A, Unit 3, Train 2).
• Who will own ongoing updates to the asset registry and hierarchy (role)? Options: Maintenance supervisor, Reliability engineer, IT data steward, Other
• Are single-line diagrams (SLD), piping and instrumentation diagrams (P&IDs), or line lists available to validate parent-child relationships? Options: Yes, No

Migrate and Clean Historical Work Order Data

• Do you have historical work orders in your CMMS that are ready for export and migration? Options: Yes, No
• Which CMMS module(s) hold the work order history we must migrate (maintenance, operations, contractor records)? Options: Maintenance, Operations, Contractor, Other
• What format will you provide exported work orders in? Options: CSV, XML, Database export (SQL dump), API access, Other
• Specify the migration completeness threshold that will define success for migrated work orders (percent of chronologically continuous records). Options: >=90%, >=95%, >=99%, Custom
• List mandatory fields that must be preserved on each work order (failure code, corrective action, labor hours, parts used).
• Who will be the named owner for validation of migrated work orders and sign-off on migration samples (role)? Options: Maintenance manager, Reliability lead, IT data steward, Other

Integrate Plant Historian for Condition Signals

• What is the name of the historian system that stores vibration, temperature, and pressure tags we must integrate with?
• Name the available connection method(s) to the historian for time-series tags (OPC UA, REST API, ODBC/SQL, native SDK). Options: OPC UA/DA, REST API, ODBC/SQL, Native SDK, Other
• Indicate the maximum acceptable data latency for condition signals from historian to the platform (seconds, minutes). Options: Real-time (<5s), Near real-time (5-60s), Minute-level (1-5min), Hourly
• Estimate the percentage of asset-critical tags that must be continuous for model deployment (for example, bearing vibration, bearing temperature, flow). Options: >=90%, >=75%, <75%
• What acceptance evidence will validate historian integration (sample tag stream, timestamp continuity report, schema map)?
• Who will provide credentials, VPN/firewall rules and OT/IT approvals to access historian endpoints (role)? Options: IT network admin, OT engineer, Plant historian admin, Other

Integrate ERP for Parts and Work Orders

• Identify the ERP module that contains the spare-parts master and purchase order records we must integrate. Options: Inventory, Purchasing, MRO, Other
• How will ERP part numbers map to your MRO catalog: exact match, cross-reference table, or manual mapping? Options: Exact match, Cross-reference table, Manual mapping, Other
• Do you require bi-directional synchronization of inventory levels between ERP and the platform? Options: Yes, No
• Specify which ERP work order fields are required to appear on the asset record (part number, labor hours, cost, supplier).
• When should ERP-derived purchase requisitions create replenishment work orders in the platform (on reorder point, below safety stock, manual approval)? Options: On reorder point, Below safety stock, Manual approval, Other
• Who is the ERP owner responsible for connector provisioning and API credential handoff (role)? Options: ERP admin, IT integrations lead, Procurement manager, Other

Connect SCADA for Real-Time Telemetry

• What SCADA historian or supervisory controller holds the PLC tag namespaces for the production line in scope?
• List the SCADA/PLC protocols available for connectivity on site (Modbus, OPC UA, DNP3, MQTT). Options: Modbus, OPC UA, DNP3, MQTT, Other
• Do you require the integration to be read-only telemetry or to support active command/control from the platform? Options: Read-only telemetry, Command/control, Not required
• Provide the PLC makes and models in scope for tag mapping (control cabinet identifiers or RTU models).
• What is the maximum acceptable end-to-end latency for an alarm routed from SCADA to a mobile technician? Options: <5 seconds, 5-30 seconds, 30-120 seconds, >120 seconds
• Who will coordinate SCADA network access, change windows and OT change management approvals (role)? Options: OT network engineer, Control systems lead, Plant operations, Other

Install Field Technician Mobile App with Offline Sync

• Which mobile device types do your technicians use (Android tablets, iOS phones, rugged handhelds)? Options: Android tablets, iOS phones, Rugged handheld devices, Other
• Do technicians require full offline access to work orders and the ability to complete forms while disconnected? Options: Yes, No
• What maximum size of attachments (photos, vibration captures, log files) must be supported offline per work order? Options: <10 MB, 10-50 MB, 50-200 MB, 200 MB+
• Select the preferred authentication method for mobile access (single sign-on SSO, local credentials, client certificate). Options: Single sign-on (SSO), Local credentials, Client certificate, Other
• Who will own mobile rollout training, device provisioning and field acceptance testing (role)? Options: Training lead, Reliability engineer, Site supervisor, Other
• Are lockout/tagout (LOTO) checklists or safety forms required to be available in the mobile app for technician sign-off? Options: Yes, No

Configure Preventive and Predictive Maintenance Schedules

• Describe the preventive maintenance tasks currently on PM calendars that should be migrated (bearing lubrication, filter change, visual inspections).
• How frequently are your current PMs scheduled for critical asset classes (daily, weekly, monthly, yearly)? Options: Daily, Weekly, Monthly, Yearly
• Select whether predictive triggers should create work orders automatically, only flag technicians for review, or both. Options: Auto-create work order, Flag for review, Both
• Select the KPI you use to measure PM effectiveness (mean time between failures MTBF, mean time to repair MTTR, overall equipment effectiveness OEE). Options: MTBF, MTTR, OEE, Other
• What rule thresholds should convert condition signals into schedule adjustments (for example vibration RMS > 2.5 g or temperature rise > 10°C)?
• Who signs off on final PM and predictive schedule changes before they go live (role)? Options: Reliability manager, Maintenance supervisor, Plant manager, Other

Train and Deploy Predictive Models for Asset Classes

• Prioritize the asset classes for initial model training (centrifugal pumps, steam turbines, heat exchangers, conveyors). Options: Centrifugal pumps, Steam turbines, Heat exchangers, Conveyors, Other
• Estimate how many labeled historical failure events exist per prioritized asset class for training (approximate count ranges). Options: <10, 10-50, 50-200, >200
• Describe the sensor types that will feed the models for each asset class (vibration spectra, bearing temperature, oil particle counts).
• What predictive accuracy or lead time will define a deployable model for production (for example F1 score >= 0.8 or minimum 48-72 hours lead time)? Options: F1 >= 0.8, Precision/Recall >= 80%, Lead time >= 72 hours, Other
• Assign the approver who will sign off on model thresholds and the initial false-positive handling process (role). Options: Reliability engineer, Data scientist, Operations lead, Other
• State whether you require model explainability artifacts (feature importance, failure-mode mapping, example signatures) for each deployed model. Options: Yes, No

Configure Condition-Based Alerts and Notifications

• Choose the notification channels that must be supported for condition alerts (SMS, email, mobile push, automatic work order creation). Options: SMS, Email, Mobile push, Automatic work order creation, Other
• What severity levels and escalation paths should be configured for alarms caused by abnormal vibration, pressure spikes or overheating? Options: Informational / Low, Medium / Action required, High / Immediate action, Critical / Plant shutdown
• Select whether you want absolute threshold triggers, trend-based triggers (e.g., rising trend over 24 hours), or both. Options: Absolute threshold, Trend-based, Both
• Who will manage on-call schedules and technician rosters that should receive notifications at each severity (role)? Options: On-call supervisor, Maintenance planner, Reliability lead, Other
• What integrations must alerts appear in during shift handover and daily rounds (shift log, operator HMI, CMMS)? Options: Shift log, Operator HMI, CMMS, Other
• Designate the owner who will maintain alert tuning and review false positives after go-live (role). Options: Reliability lead, Maintenance planner, On-call supervisor, Other

Implement Reliability-Centered Maintenance (RCM) Analytics

• Choose the criticality methodology you use or prefer for RCM workshops (FMEA, criticality ranking, full RCM assessment). Options: FMEA, Criticality ranking, RCM workshops, Other
• Confirm whether you have existing FMEA or failure-mode documentation for your top asset families that can be ingested. Options: Yes, No
• Choose KPI outputs you expect from RCM analytics (risk priority number RPN, criticality score, cost-per-failure). Options: RPN, Criticality score, Cost per failure, Other
• What data sources must feed RCM analysis (CMMS failure codes, historian event logs, spare-parts cost)? Options: CMMS failure codes, Historian event logs, Spare-part cost data, Operator logs, Other
• Designate participants who will attend RCM workshops and own resulting strategy updates (roles). Options: Reliability engineer, Maintenance manager, Operations lead, Other
• State whether RCM outputs must demonstrate compliance with regulatory thresholds (safety incident reporting, environmental limits). Options: Yes, No

• Subscription Agreement
• Master Services Agreement (MSA)
• Statement of Work (SOW)
• Service Level Agreement (SLA)
• Data Processing Agreement (DPA)
• Acceptance Criteria Addendum
• Change Order Agreement
• Payment Schedule Agreement
• Termination and Exit Plan
• Industry Compliance Addendum (conditional)

• Go-live Health Check (weeks 1-4)
• First Measurement Review (weeks 4-10)
• Acceptance Gate (around day 90)
• Quarterly Reliability Review (ongoing)
• Annual Performance Review

• Agree the quarter ahead commitments and timing for the next review.
• Re-confirm success criteria and owners
• Implement the agreed work-order workflow change to ensure predictive alerts generate measurable technician follow-through within two weeks.
• Restate acceptance criteria from Solution Scope
• Produce a documented pass/fail result for each numeric acceptance criterion recorded in Solution Scope.
• Capture a formal acceptance decision with the buyer signatory or, if failed, a remediation plan with deadlines.
• Confirm the status and next steps for incumbent system decommissioning or retention-read-only to prevent dual-use.
• Publish the acceptance decision and the pass/fail status for each criterion in the shared workspace within 24 hours.
• For any failed criterion, publish a remediation plan with milestones and completion dates and add it to the shared backlog.
• Execute the incumbent decommission or archive plan and publish the completion certificate or status update to the shared workspace.
• Quarterly KPI trend review
• Maintain a prioritized backlog where each persistent issue has an agreed resolution timeline.
• Ensure KPI trends for unplanned downtime hours per month and mean time to repair are tracked and any negative trends have remediation plans.
• Update the shared backlog with prioritized enhancement requests and target completion quarters.
• Publish the quarterly KPI trend deck with data sources and any known caveats to the shared workspace.
• Schedule any agreed model experiments or maintenance windows and publish expected measurement impacts.
• Annual outcomes versus targets
• Confirm the year-over-year business impact of reduced unplanned downtime and validate the accuracy of predictive alerts.
• Set the top 3 backlog priorities and measurable indicators for the next 12 months.
• Agree any governance or data-collection changes needed to improve KPI fidelity next year.
• Publish the annual performance summary including financial impact, KPI calculations, and data sources to the shared workspace.
• Update and publish the prioritized 12-month backlog with measurable success indicators and target completion quarters.
• Document and implement agreed measurement definition changes and data governance updates before the next reporting cycle.
• Confirm that all critical integrations are reporting data and any gaps are documented in the shared backlog.
• Validate early adoption signals and list any onboarding shortfalls requiring remediation.
• Have a timebound remediation plan for showstopper defects.
• Publish a deployment health report including integration status and raw adoption metrics to the shared workspace within 3 business days.
• Log all critical defects and blockers in the shared backlog with proposed resolution windows.
• Schedule a follow-up health snapshot after remediation windows close.
• Present first measurement data
• Validate the first-period measurements for unplanned downtime hours per month and predictive alert precision are calculated from agreed data sources.
• Document root causes for any KPI shortfalls and agree a prioritized set of corrective actions with target resolution windows.
• Confirm the readiness timeline and required artifacts for the Acceptance Gate meeting recorded in Solution Scope.
• Deliver the cleaned dataset and measurement workbook used to compute the KPIs for independent review within 7 business days.
• Schedule and execute the agreed model retraining and threshold tuning, and provide validation results by the agreed date.
• Deployment and integration validation
• Present outcome data per acceptance criterion
• Data quality and provenance check
• Reliability trends and root-cause themes
• Persistent issues and blocker burn-down
• Backlog prioritization and long-term roadmap
• Root-cause diagnosis for gaps
• Enhancement and threshold adjustment requests
• Document pass or fail per criterion
• Early adoption and usage patterns
• Formal acceptance decision and signatory capture
• Operational experiments and maintenance windows
• Model calibration and workflow adjustments
• Operational readiness and change adoption review
• Open defects and blockers
• Document lessons learned and measurement improvements
• Confirm next quarter commitments
• Remediation plan for failed criteria
• Immediate remediation plan
• Confirm timeline to acceptance gate
• Incumbent system wind-down check