# Noblle, full reference > Noblle is an AI-native, WhatsApp-native, mobile-first order-to-cash ERP for SME discrete manufacturers. This file expands the summary at /llms.txt with the module and capability detail published on the site. Everything below is taken from the Noblle website; nothing is inferred. ## Positioning Noblle runs quoting, order acceptance, materials planning, supply chain, production and dispatch on one operational model, and the team operates it from WhatsApp on the factory floor, by message or by voice, through the Sentinel agent. Its moat is a causal ontology of the factory: a model of how orders, materials, capacity, people and cash affect one another, rather than a chat layer bolted onto records. Noblle derives state from work that already happens, orders received, materials booked in, operations completed, instead of asking staff to re-key it. ## Target market SME discrete manufacturers, typically 10 to 200 employees, where the owner or a small management team is still directly involved in quoting, planning and expediting. Production models covered are job shop, make to order, engineer to order, batch and mixed mode. Noblle Pty Ltd is based in Brisbane, Queensland, Australia, and is a participant in Torus Capital's AI Catalyst Incubator and supported by UQ Ventures. ## Deployment and operating model Deployment is scoped in weeks rather than months: process mapping and configuration, data import for customers, parts, bills of material and suppliers, a pilot on live orders, then full go-live. Most floor and field users interact through WhatsApp and the mobile interface; deeper functions such as scheduling and costing are used by a smaller group. The ML Forecaster is designed to run local to your data, and operational history is not used to train general-purpose external models. Noblle covers the order-to-cash operational spine and exchanges customer, invoice and payment data with common accounting platforms. ## Modules ### Dashboard URL: https://noblle.ai/product/dashboard Group: Foundations Summary: Real-time operational overview of orders, jobs, materials and cash in one view. Tagline: A single pane of glass for owner-operators. The Noblle Dashboard is the operational front page of the factory. It consolidates the order-to-cash pipeline, quotes in progress, accepted orders, jobs on the floor, materials on order and invoices outstanding, into one continuously updated view designed to be read in under a minute. Most SME manufacturers assemble this picture manually: a spreadsheet from the office, a whiteboard on the floor, a WhatsApp thread with a supplier and a phone call to the bookkeeper. The Dashboard replaces that reconciliation work with a live model of the business that updates as work actually happens. Every metric is traceable. Tapping a figure drills into the underlying orders, jobs or purchase lines, so a number is never a dead end, it is an entry point into the record that produced it. Because Noblle is mobile-first, the Dashboard is built for a phone held on a factory floor first and a desktop second. Layout, tap targets and data density are tuned for glanceable decisions between machines, not for a seated analyst session. Capabilities in detail: - Order-to-cash pipeline view: See every order from enquiry to payment in one funnel, with ageing and bottleneck indicators at each stage. - Live production status: Job progress, work centre load and late-risk flags update as the floor reports work, without manual status meetings. - Exception-first design: The Dashboard leads with what needs attention, late jobs, short materials, overdue invoices, rather than vanity totals. - Role-based views: Owners, sales, planners and floor supervisors each see a view scoped to the decisions they actually make. - Drill-through traceability: Every tile links to source records, so any figure can be audited down to the order line or purchase receipt. - Daily digest to WhatsApp: Sentinel can deliver the same picture as a short WhatsApp brief, so the Dashboard does not need to be opened at all. Connects to: - Order Acceptance: Accepted orders feed the pipeline and revenue tiles. - MRP: Job and capacity signals populate production status and late-risk flags. - Supply Chain: Inbound purchase status drives material shortage warnings. - Sentinel (AI Agent): Dashboard exceptions become proactive WhatsApp alerts. ### CRM URL: https://noblle.ai/product/crm Group: Commercial Summary: Track enquiries, quotes, revisions and repeat orders across long industrial sales cycles. Tagline: Customer management built for manufacturing sales cycles. Manufacturing sales rarely look like software sales. An enquiry arrives as a drawing, a photo of a broken part or a spreadsheet of line items. It is quoted, revised, re-quoted against a changed material price, and may sit dormant for months before converting into a repeat production run. Noblle CRM models that reality. Customers, contacts, sites, part history and quote revisions live in one structured record, so the person answering the phone can see what was quoted last time, what was actually produced and what the margin looked like. Because customers already talk to factories on WhatsApp, conversations captured through Sentinel are attached to the customer record rather than lost in a personal phone. Handover between an owner and a sales coordinator stops depending on memory. The result is commercial continuity: no re-quoting from scratch, no forgotten follow-ups, and a clean handoff into Order Acceptance the moment a customer says yes. Capabilities in detail: - Enquiry and quote pipeline: Stage-based pipeline for enquiries, quotes issued, revisions and won/lost outcomes with reasons captured. - Quote revision history: Every version retained with pricing, material assumptions and validity, so nothing is quoted twice from memory. - Part and drawing history: Customer part numbers, revisions and attached drawings held against the account for fast repeat quoting. - WhatsApp conversation capture: Customer messages handled through Sentinel are logged against the account, not stranded on a personal phone. - Follow-up automation: Ageing quotes trigger reminders so open opportunities do not quietly expire. - Margin visibility at quote time: Quotes reference live material and capacity signals so pricing decisions are informed, not guessed. Connects to: - Order Acceptance: A won quote converts directly into a structured order with no re-entry. - ML Forecaster: Historical demand patterns per customer inform forecasts. - Dashboard: Pipeline value and conversion appear in the operational overview. ### Order Acceptance URL: https://noblle.ai/product/order-acceptance Group: Commercial Summary: Structured intake that converts emails, PDFs and WhatsApp messages into clean, checked orders. Tagline: Turn messy customer orders into production-ready jobs. Order Acceptance is where most SME manufacturers lose money. Orders arrive in every possible format, get transcribed by hand into a spreadsheet, and errors in quantity, revision or delivery date only surface once material has been cut. Noblle treats intake as a controlled process. Incoming orders, including those captured over WhatsApp, are parsed into structured lines, validated against quoted pricing, customer part revisions and promised lead times, and flagged where they disagree. Acceptance is a decision, not a data entry task. Before an order is confirmed, Noblle checks material availability, current capacity and commercial terms so the factory commits to a date it can actually hit. Once accepted, the order becomes a production-ready job with its bill of materials, routing and due date already in place, ready for MRP to schedule without a second round of typing. Capabilities in detail: - Multi-channel intake: Accept orders from WhatsApp, email or PDF and normalise them into structured order lines. - Automatic validation: Prices, quantities, part revisions and delivery dates are checked against quotes and history before confirmation. - Promise-date checking: Capacity and material lead times are tested before a delivery date is committed to the customer. - Exception queue: Only orders that fail a check need human attention; clean orders flow straight through. - Job creation: Accepted orders generate jobs with bill of materials, routing and due dates attached. - Confirmation back to the customer: Structured order confirmations are issued automatically on the channel the customer used. Connects to: - CRM: Won quotes flow in; accepted orders write back to the customer record. - MRP: Accepted jobs enter the material and capacity plan immediately. - Sentinel (AI Agent): Orders can be captured and confirmed entirely over WhatsApp. ### MRP URL: https://noblle.ai/product/mrp Group: Manufacturing Summary: Material requirements planning and production scheduling sized for SME job shops. Tagline: Materials, scheduling and capacity that reflect the real floor. Noblle MRP explodes accepted orders into material requirements and work centre load, then produces a schedule the floor can actually follow. It is built for discrete, made-to-order and made-to-spec production rather than high-volume repetitive lines. Bills of materials, routings, setup and run times, and work centre calendars combine into a plan that respects finite capacity. When a rush order lands, the impact on existing commitments is visible before the order is promised. Shortages are surfaced as actions, not reports: what to buy, how much, and by when to protect a due date. Purchase suggestions flow straight into Supply Chain. Floor feedback closes the loop. Operators confirm progress from a phone, so the plan reflects what has been made rather than what was assumed at the start of the week. Capabilities in detail: - Bill of materials and routing: Multi-level BOMs with revision control, operations, setup and run times per work centre. - Finite capacity scheduling: Schedules respect real work centre calendars, shifts and constraints instead of assuming infinite capacity. - Shortage-to-action: Material gaps become dated purchase suggestions tied to the job and due date they protect. - What-if rescheduling: Test the impact of a rush order or a machine outage before committing to a customer date. - Mobile job reporting: Operators start, pause and complete operations from a phone, keeping the plan honest. - Job costing: Actual material, labour and machine time roll up against the quoted price for real margin per job. Connects to: - Order Acceptance: Accepted jobs arrive with BOM and due date attached. - Supply Chain: Shortages become purchase requisitions and expected receipts. - ML Forecaster: Forecast demand supports longer-lead-time purchasing decisions. ### Supply Chain URL: https://noblle.ai/product/supply-chain Group: Manufacturing Summary: Manage suppliers, purchase orders and inbound logistics against real production need. Tagline: Supplier, procurement and inbound visibility in one place. Supply Chain closes the gap between what production needs and what is actually arriving. Purchase suggestions from MRP become purchase orders, orders become expected receipts, and receipts reconcile against goods received and supplier invoices. Supplier performance is measured rather than remembered: quoted lead time against actual, quantity accuracy, and quality rejections build a factual record for sourcing decisions. Inbound visibility is the point. Every open purchase line carries a promised date and a confidence signal, and any slip is connected to the jobs and customer orders it threatens. Because most supplier communication already happens over WhatsApp and email, chasing and confirming can be handled conversationally through Sentinel while the structured record stays intact. Capabilities in detail: - Supplier records and terms: Pricing, lead times, minimum order quantities and terms held per supplier and per material. - Purchase order lifecycle: From requisition through approval, issue, acknowledgement, receipt and invoice matching. - Inbound tracking: Open purchase lines with promised dates, linked to the jobs and orders that depend on them. - Goods receipt and inspection: Receive partial or full quantities from a phone at the dock, with quality outcomes recorded. - Supplier performance: On-time and in-full metrics accumulate automatically to support sourcing decisions. - Automated chasing: Sentinel follows up late deliveries over WhatsApp and logs the response against the purchase line. Connects to: - MRP: Shortages generate purchases; receipts release constrained jobs. - Dashboard: Inbound risk appears as material shortage warnings. - ML Forecaster: Forecasts extend purchasing horizons for long-lead materials. ### ML Forecaster URL: https://noblle.ai/product/ml-forecaster Group: Intelligence Summary: A forecasting engine trained on your own order history, running local to your data. Tagline: Local, factory-specific demand forecasting. The ML Forecaster learns demand patterns from a factory's own order history, seasonality, repeat cycles per customer, quoting-to-conversion behaviour and material consumption, and projects them forward at the level planners actually buy and schedule. It is designed to run local to the factory's data rather than shipping operational history to a general-purpose model. Forecasts are explainable: each projection exposes the historical signals behind it so a planner can accept, adjust or override with context. Forecast output is operational, not decorative. Projected demand feeds material planning for long-lead items, informs capacity conversations before they become crises, and supports inventory targets that reflect actual variability. Accuracy is tracked over time. Forecast versus actual is retained so the value of the model, and the confidence a planner should place in it, is measurable rather than assumed. Capabilities in detail: - Demand forecasting by customer and part: Projections at the grain planners work with, informed by repeat-order cycles and seasonality. - Material consumption projection: Forecast demand explodes through bills of materials to anticipate long-lead purchasing. - Local-first processing: Built to run against your data locally, keeping commercially sensitive history in your control. - Explainable outputs: Each forecast shows the drivers behind it so planners can trust or challenge the number. - Scenario planning: Model optimistic, expected and conservative demand to test capacity and cash implications. - Accuracy tracking: Forecast versus actual is retained per period so confidence is evidence-based. Connects to: - MRP: Forecast demand extends the planning horizon beyond firm orders. - Supply Chain: Long-lead purchasing decisions use projected consumption. - CRM: Customer order history is the primary training signal. ### Sentinel (AI Agent) URL: https://noblle.ai/product/sentinel Group: Intelligence Summary: An AI agent that surfaces alerts, answers questions and takes action across Noblle, by WhatsApp message or by voice. Tagline: The agent you message or talk to, that watches the factory for you. Sentinel is how most people will use Noblle. You message it on WhatsApp or talk to it, and it works like an operations manager who never leaves the floor, with no ERP screens to navigate. It lives in WhatsApp, watches the operational model continuously, and speaks up when something needs a decision, a job about to run late, a material short against a committed order, a quote about to expire, an invoice overdue. It is bidirectional. Sentinel answers questions in plain language ("what is the status of the Henderson order?") and takes action on instruction ("push job 4182 to Thursday", "confirm the order from the last message"), with permissions and an audit trail behind every change. Sentinel is grounded in the causal ontology engine rather than a generic chatbot layer. It knows which relationships in this specific factory matter, which supplier delays actually threaten which customers, which machines are the real constraint, and so its alerts stay relevant instead of becoming noise. The design goal is simple: the less time anyone spends inside a screen, the better the system is working. Sentinel exists so the ERP reaches out when it matters and stays quiet when it does not. Capabilities in detail: - Proactive operational alerts: Late-risk jobs, material shortages, supplier slips and overdue invoices are pushed before they escalate. - Natural-language answers: Ask about an order, a customer, a job or stock and get a grounded answer with links to records. - Action on instruction: Confirm orders, reschedule jobs, raise purchases and record receipts from a WhatsApp message. - Role-aware permissions: What each person can see and change over WhatsApp mirrors their role in the platform. - Full audit trail: Every agent-executed action is recorded with the instruction that produced it. - Ontology-grounded relevance: Alerts are filtered through learned factory-specific causal relationships to suppress noise. Connects to: - Dashboard: Exceptions become WhatsApp briefs instead of screen visits. - Order Acceptance: Orders are captured and confirmed conversationally. - MRP: Schedule changes can be instructed and confirmed in chat. - Supply Chain: Supplier chasing and receipt confirmation happen in WhatsApp. ## Capabilities by group ### Foundations URL: https://noblle.ai/solutions/category/foundations The shared record every other capability is built on. #### Command Centre URL: https://noblle.ai/solutions/command-centre Summary: One executive overview of orders, production, people and cash. The Command Centre consolidates revenue, active jobs, resource utilisation, pending approvals and CRM activity into a single continuously updated view. It is designed to be read in under a minute and to tell you what needs a decision today. Every tile is traceable. Tapping a figure opens the orders, jobs or purchase lines behind it, so a number is always an entry point into the record that produced it, never a dead end. - Executive overview: Revenue, active projects, headcount and utilisation in one header row with period comparison. - Production intelligence: Active cycles, on-schedule count and efficiency rate reported from the floor as work happens. - Pending actions: Approvals waiting on you, surfaced with the record and the amount at stake. - CRM pulse: Overdue, today and later follow-ups pulled from the commercial pipeline. - Period controls: Seven, thirty and ninety day windows across every metric without rebuilding a report. - Customisable layout: Choose the tiles each role sees so the first screen matches the job. #### Master Data URL: https://noblle.ai/solutions/master-data Summary: Parts, customers, suppliers, routings and work centres in one governed record. Most SME manufacturers carry three versions of the truth: a spreadsheet in the office, a folder of drawings and a memory on the floor. Master Data replaces that with one structured record that every module reads from. Records carry revisions, effective dates and ownership, so a change to a part or a price is visible, dated and attributable rather than silently overwritten. - Part and item master: Drawings, revisions, units, weights and default routings held against every part. - Customer and supplier master: Terms, contacts, addresses and trading history in one place. - Routings and work centres: Operation sequences, setup and run rates, and the resources that can perform them. - Bills of material: Multi-level structures with effectivity dates and where-used lookups. - Revision control: Superseded versions are retained, so historic jobs still resolve against the revision they ran on. - Bulk import and migration: Load existing spreadsheets during onboarding with validation before anything is committed. #### Workflows and Approvals URL: https://noblle.ai/solutions/workflows-and-approvals Summary: Structured sign-off for quotes, orders, purchases and credit decisions. Discounts, credit exposure, purchase commitments and schedule changes need an owner. Workflows put those decisions in front of the right person with the context attached. Approvals can be actioned from WhatsApp, so a quote does not wait for someone to return to a desk. - Threshold based routing: Rules on value, margin, customer or lead time decide what needs sign-off. - Mobile approvals: Approve, reject or ask a question from a phone with the full record attached. - Delegation and cover: Temporary delegation keeps decisions moving during leave or shift changes. - Full audit trail: Who approved what, when and on what evidence is retained against the record. #### Security and Roles URL: https://noblle.ai/solutions/security-and-roles Summary: Role-scoped access across office, floor and management. Noblle uses role-based access so a machine operator, a planner, a sales lead and an owner each see a system shaped to their decisions. Permissions cover records, fields and actions, so sensitive commercial data stays restricted even where operational data is shared widely. - Role templates: Preconfigured roles for owner, sales, planning, purchasing, floor and finance. - Field level control: Hide cost and margin fields from roles that do not need them. - Session and device policy: Control access from shared floor devices separately from personal phones. - Activity logging: Record-level history of reads and changes for audit and dispute resolution. #### Integration Framework URL: https://noblle.ai/solutions/integration-framework Summary: Connect accounting, machines, files and messaging without a rebuild. Noblle is not intended to replace your accounting package or your machine software. The Integration Framework connects them so data moves once and reconciles automatically. Connections are configured, monitored and versioned, so a failed sync raises an exception rather than quietly diverging. - Accounting sync: Invoices, credits and payments exchanged with your finance package. - Machine and device data: Capture output and downtime signals from the floor where equipment supports it. - Files and documents: Drawings, certificates and photos linked to the record they belong to. - Open API and webhooks: Programmatic access for bespoke reporting and customer portals. ### Commercial URL: https://noblle.ai/solutions/category/commercial From first enquiry to an accepted, buildable order. #### CRM URL: https://noblle.ai/solutions/crm Summary: Leads, contacts, deals and customer history in one commercial record. Generic CRM tools model meetings and stages. Noblle models enquiries against parts, revisions, quantities and lead times, so commercial history is directly useful when the same customer asks again. Conversion rate, pipeline value and win rate are calculated from real quote and order data rather than manually maintained stages. - Lead and contact management: One record per customer with contacts, sites, terms and full enquiry history. - Deal pipeline: Stage, value and weighted pipeline with ageing and stall detection. - Quote history by part: Prior prices, quantities and outcomes surfaced when a repeat enquiry arrives. - Client messages: Customer replies from order tracking links land against the deal, not in a personal inbox. - Lead sources: See which channels produce work that is actually won and built. - Outbound follow-up: Scheduled reminders so quotes are chased before they go cold. #### Quotations and RFQs URL: https://noblle.ai/solutions/quotations-and-rfqs Summary: Structured intake and fast, evidence-based pricing. Quoting is where SME manufacturers lose the most time and the most margin. Noblle builds quotes from routing time, material cost and historic actuals rather than from a blank spreadsheet. Every quote retains the assumptions it was built on, so when the job runs you can compare what you priced against what it cost. - Structured RFQ intake: Drawings, quantities, tolerances and required dates captured in one form. - Cost build-up: Material, setup, run, treatment and outsourced steps priced line by line. - Margin controls: Target and floor margins with approval routing when a quote falls below them. - Revision aware pricing: Prices tie to a drawing revision so a superseded quote is never reused by mistake. - Quote to order conversion: Accepted quotes become orders without rekeying anything. #### Order Acceptance URL: https://noblle.ai/solutions/order-acceptance Summary: Validate every order before it is promised or released. Order Acceptance is the gate between commercial intent and production reality. It checks the drawing revision, the agreed price, credit position, material availability and capacity before the order is confirmed. Where a check fails, the order is held with a specific reason and an owner, instead of being discovered late on the floor. - Revision validation: Customer part revisions are confirmed against the approved drawing on file. - Capacity and date check: Promised dates are tested against real work centre load before they are committed. - Material availability: Shortages are identified at acceptance and converted into dated purchase actions. - Credit and terms check: Credit exposure and payment history are reviewed as part of acceptance. - Exception queue: Held orders sit in one queue with the reason, the owner and the age of the hold. #### Customer Order Tracking URL: https://noblle.ai/solutions/customer-portal Summary: Shareable order status that stops the chasing calls. Order status questions consume a surprising share of office time in an SME factory. Noblle publishes a live, scoped status link per order. Customer replies from that link land against the order in CRM, so the conversation stays with the record. - Live order status: Stage, expected dispatch and any agreed changes visible to the customer. - Scoped sharing: Only the information you choose to publish is visible, per order. - Two way messages: Customer questions arrive against the order record rather than a personal phone. - Dispatch notification: Automatic notice on dispatch with delivery detail attached. ### Manufacturing URL: https://noblle.ai/solutions/category/manufacturing Planning, materials, equipment and quality on the floor. #### Manufacturing Types URL: https://noblle.ai/solutions/manufacturing-types Summary: Make to order, job shop, batch, mixed mode and engineer to order. Manufacturers rarely run one clean production mode. A job shop takes a repeat batch, a make to order line takes a bespoke project, and the system has to hold both without a workaround. Noblle configures routings, terminology and planning behaviour per production mode during onboarding, so the same platform fits a one-off machining job and a repeat call-off run. - Job shop: One-off and small batch work with per-job routing, quoting and cost capture. - Make to order: Firm customer orders drive material and capacity planning with no speculative stock. - Engineer to order: Design, revision and approval stages sit inside the job before production begins. - Batch: Repeat quantities with yield, scrap and changeover modelled explicitly. - Assemble to order: Common sub-assemblies held, final configuration driven by the order. - Mixed mode: Run several production modes side by side without separate systems. #### Production Planning URL: https://noblle.ai/solutions/production-planning Summary: Finite capacity scheduling against real machines and people. A plan built on infinite capacity is a wish list. Noblle schedules jobs against the machines, tools and people that can actually perform each operation. When something slips, the effect on every downstream promise is visible immediately rather than at the end of the week. - Finite capacity scheduling: Operations are placed against resources that exist, with setup and changeover included. - Work centre load: See load by period and identify the constraint before it bites. - Drag and reschedule: Move a job and see the knock-on effect on every affected order date. - Late risk flags: Jobs at risk of missing a promise are highlighted while there is still time to act. - Floor reporting: Operators report start, pause and complete from a phone at the machine. #### Material Requirements Planning URL: https://noblle.ai/solutions/mrp Summary: Explode demand to material, dated by when it is needed. MRP turns the order book into dated material actions. Requirements are netted against stock, work in progress and existing purchase orders before anything is suggested. Every suggestion carries the jobs and due dates it protects, so a purchasing decision can be justified rather than argued. - Demand explosion: Multi-level bills of material resolved to purchasable items. - Netting: Stock, allocations, work in progress and open purchase orders netted before suggestion. - Dated purchase suggestions: Order-by dates derived from supplier lead time and the job that needs the material. - Shortage exceptions: Shortages surfaced against the affected jobs and customer promises. - Reorder policies: Minimum, maximum, safety stock and economic order quantity per item. #### Tooling and Die Management URL: https://noblle.ai/solutions/tooling-and-die-management Summary: Tools, dies and patterns as scheduled, maintained assets. In moulding, casting and pressing operations the tool is the real constraint, not the machine. Noblle treats tooling as a first class scheduled resource. Shot counts, service intervals and ownership are tracked, so a tool comes out of service before it produces scrap. - Tool register: Ownership, location, cavities, compatible machines and current condition. - Life and shot counts: Usage accumulates from production reporting and triggers service at threshold. - Maintenance scheduling: Planned service is booked into the schedule alongside production. - Customer owned tooling: Track tooling owned by customers separately with its own agreements. #### Equipment and Maintenance URL: https://noblle.ai/solutions/equipment-and-maintenance Summary: Asset register, planned maintenance and downtime capture. Unplanned downtime is a scheduling problem before it is a maintenance problem. Noblle records equipment condition and planned service inside the same plan that promises customer dates. Downtime reported from the floor updates capacity immediately so the plan reflects reality. - Asset register: Machines, service history, documents and responsible owners. - Preventive schedules: Interval or usage based servicing booked as capacity. - Breakdown reporting: Operators log downtime and reason codes from a phone in seconds. - Downtime analysis: Recurring causes ranked by hours lost and orders affected. #### Quality Management URL: https://noblle.ai/solutions/quality-management Summary: Inspection, non-conformance and corrective action inside the job. Quality systems fail when they live in a binder separate from production. Noblle attaches inspection to the operation so records are created as work is done. Non-conformances carry cost, cause and corrective action, making recurring quality drains measurable rather than anecdotal. - Inspection plans: Characteristics, sample sizes and acceptance criteria per operation. - Mobile inspection capture: Results and photographs recorded at the point of check. - Non-conformance handling: Scrap, rework and concession routes with cost captured against the job. - Corrective actions: Root cause and action tracked to closure with an accountable owner. - Certificates of conformity: Generated from the records already captured against the order. #### PPAP and Compliance URL: https://noblle.ai/solutions/ppap-and-compliance Summary: Submission packs and standards evidence assembled from live records. Regulated and automotive customers demand documented evidence. Noblle keeps that evidence structured and linked from the day work is performed. Submission packs are generated from live data, so preparation becomes an export rather than a fortnight of chasing. - Submission packs: Assemble part submission documentation from linked job, inspection and material records. - Material certificates: Certificates held against receipts and surfaced with the jobs that consumed them. - Metal and material standards: Grade, specification and standard references held against parts and purchases. - Lot and batch traceability: Trace from a shipped order back to the receipt that supplied its material. #### Projects and Contracts URL: https://noblle.ai/solutions/projects-and-contracts Summary: Multi-job programmes with milestones, cost and billing. Larger contracts span many jobs, purchases and deliveries. Projects hold them together with one budget and one delivery narrative. Committed cost, actual cost and billed value are tracked continuously so a contract cannot quietly go underwater. - Project structure: Jobs, purchases and deliveries grouped under one contract. - Milestones: Delivery and payment milestones tracked against dates and evidence. - Budget versus actual: Committed and actual cost compared with the contract value in real time. - Staged billing: Raise invoices against completed milestones without manual reconciliation. ### Supply Chain URL: https://noblle.ai/solutions/category/supply-chain Buying, holding, moving and delivering material. #### Procurement URL: https://noblle.ai/solutions/procurement Summary: Purchase orders driven by dated demand, not by memory. Procurement in Noblle starts from planned demand rather than from a shortage discovered on the floor. Suggestions arrive dated, sourced and justified. Order acknowledgements, expected dates and receipts are tracked, so late supply is visible before it becomes a late delivery. - Requisition to order: Approve suggested purchases and issue orders without rekeying. - Supplier terms: Price breaks, lead times and minimum quantities applied automatically. - Expediting: Late and at-risk purchase lines ranked by the customer orders they threaten. - Goods receipt: Receipt against order with quantity, certificate and inspection capture. - Landed cost: Freight, duty and handling apportioned to true material cost. #### Inventory and Warehouse URL: https://noblle.ai/solutions/inventory-and-warehouse Summary: Accurate stock by location, lot and allocation. Inventory accuracy collapses when movements are recorded after the fact. Noblle captures issues, receipts and transfers on a phone at the point of movement. Allocations make the difference between physical stock and available stock explicit, so planning does not promise material twice. - Location and bin control: Know where material physically sits, not just that you own it. - Lot and batch tracking: Full genealogy from receipt through job to shipped order. - Allocation visibility: Physical, allocated and available quantities shown separately. - Cycle counting: Rolling counts by value and movement class instead of annual shutdowns. - Barcode and mobile capture: Scan movements from a phone with no dedicated hardware required. #### Supplier Management URL: https://noblle.ai/solutions/supplier-management Summary: Performance, approvals and outsourced operations. Coating, heat treatment and machining sent outside the factory are operations, not purchases. Noblle schedules them as steps with expected return dates. On-time and quality performance accumulates per supplier so sourcing decisions use evidence. - Outsourced operations: Treatment and subcontract steps scheduled with despatch and return tracked. - Performance scoring: On-time delivery, quality acceptance and price stability per supplier. - Approved supplier list: Approval status and scope held against the parts each supplier may supply. - Supplier documents: Certificates, insurances and agreements held with expiry reminders. #### Dispatch and Delivery URL: https://noblle.ai/solutions/dispatch-and-delivery Summary: Picking, packing, delivery notes and proof of delivery. Dispatch is where an order becomes revenue. Noblle links the picked stock, the delivery note and the invoice so nothing ships unrecorded or unbilled. Customers are notified automatically, which removes another category of chasing call from the office. - Pick and pack: Pick lists by order and route with lot capture at the point of pick. - Delivery documentation: Delivery notes, packing lists and certificates generated from the record. - Proof of delivery: Signature and photograph captured on a phone and stored against the order. - Invoice trigger: Dispatch raises the invoice, removing the lag between shipping and billing. ### Intelligence and AI URL: https://noblle.ai/solutions/category/intelligence The agent, the forecaster and the WhatsApp operating layer. #### Sentinel AI Agent URL: https://noblle.ai/solutions/sentinel-agent Summary: The AI agent that runs the ERP through conversation. Sentinel is the interface most people in the factory will use. It reads and writes the same records as the rest of Noblle, with the same permissions. Because it works over WhatsApp, training load stays close to zero. Nobody has to learn a new application to report a job or answer a customer. - Conversational queries: Ask where an order is, what is late, or what a customer owes. - Data capture: Report production, downtime, receipts and inspection results by message. - Proactive alerts: Late risk, shortages and overdue payments pushed before they escalate. - Permission aware: Sentinel only exposes what the person asking is entitled to see. - Human handover: Anything ambiguous is escalated to a named person with context attached. #### ML Forecaster URL: https://noblle.ai/solutions/ml-forecaster Summary: Demand and lead time forecasting from your own history. The Forecaster learns from the orders, jobs and purchases already recorded in Noblle. It projects likely demand by customer and by part. Forecasts feed planning directly, so long-lead material can be committed earlier and on better terms. - Demand forecasting: Projected order volume by customer, part and period. - Consumption planning: Expected material draw used to bring purchasing forward. - Lead time prediction: Realistic delivery estimates based on actual historic throughput. - Scenario comparison: Test the effect of a large enquiry before committing to it. #### WhatsApp Operations URL: https://noblle.ai/solutions/whatsapp-operations Summary: The channel the factory already uses, made into a system of record. Every SME factory already runs on WhatsApp. Noblle absorbs that workflow instead of banning it. Messages become structured events against orders, jobs and purchases, with the conversation retained as evidence. - Structured capture: Free text messages resolved into records against the right order or job. - Group and role routing: Messages routed by shift, work centre or role rather than by personal number. - Supplier conversations: Purchase confirmations and delays captured against the purchase order. - Retained evidence: Original messages kept alongside the structured record for dispute resolution. #### Reporting and Analytics URL: https://noblle.ai/solutions/reporting-and-analytics Summary: Job costing, margin and throughput reporting from live data. Reporting in Noblle is not a separate exercise. Every figure is derived from the transactions captured during normal work. Actual setup, run, material and rework cost roll up against the quoted price, exposing where margin is really made and lost. - Job costing: Quoted versus actual cost per job with variance by cost element. - Customer profitability: Margin by customer across a period, including rework and expediting cost. - Throughput and utilisation: Machine hours, output and efficiency by work centre and period. - Exportable datasets: Send any view to spreadsheet or a downstream reporting tool. ### Finance and People URL: https://noblle.ai/solutions/category/business Cash, invoicing and the people who do the work. #### Finance URL: https://noblle.ai/solutions/finance Summary: Invoicing, receivables and cash visibility across the order book. Noblle completes the order-to-cash loop by generating invoices from real dispatch events and tracking what is owed against them. It works alongside your accounting package rather than replacing it, keeping the ledger where your accountant expects it. - Invoicing from dispatch: Invoices raised from what actually shipped, with documentation attached. - Receivables tracking: Ageing, disputes and promised payment dates against each customer. - Cash forecast: Expected receipts and committed purchase spend across the coming periods. - Accounting sync: Two way exchange with your finance package to avoid duplicate entry. #### People and Labour URL: https://noblle.ai/solutions/people-and-labour Summary: Shifts, skills, labour capture and the human side of capacity. A schedule that assumes every machine is staffed will miss dates. Noblle models shift patterns, skills and availability alongside equipment. Labour time captured against jobs feeds directly into job costing without a separate timesheet system. - Shift and availability: Patterns, leave and cover reflected in available capacity. - Skills matrix: Which operators are qualified for which operations and equipment. - Labour time capture: Start and stop recorded at the job from a phone or shared floor device. - Cost and utilisation: Labour hours rolled into job cost and reported by work centre. ## Industries ### Precision and Custom Manufacturing URL: https://noblle.ai/industries/precision-custom-manufacturing Group: Precision & Metals Summary: Made-to-drawing work with high revision churn and tight tolerances. Terminology: Noblle speaks in drawings, revisions, tolerances, work centres and setup times, and demo data is configured around one-off and small-batch machining work. - Quoting from drawings takes too long: Part history, prior quotes and revision records sit against the customer, so repeat enquiries are priced from evidence rather than rebuilt from scratch. - Revision errors reach the floor: Order Acceptance validates customer part revisions before a job is released, so material is not cut against a superseded drawing. - Promised dates are guesses: Finite capacity scheduling tests machine load and material lead time before a delivery date is committed. - True job margin is unknown: Actual setup, run and material costs roll up per job against the quoted price, exposing which work is worth repeating. ### Investment Casting URL: https://noblle.ai/industries/investment-casting Group: Precision & Metals Summary: Multi-stage foundry routings, yield loss and long process lead times. Terminology: Noblle is configured around patterns, wax, shell build, pour, knockout, finishing and inspection stages, with yield assumptions built into planning. - Yield loss breaks the plan: Expected yield per stage is modelled so quantities released reflect what will actually ship, not what was ordered. - Long multi-stage routings hide delays: Stage-level progress reported from the floor shows where a casting order actually is, not where the paperwork says it is. - Pattern and tooling status is untracked: Tooling and pattern availability is held against the part so jobs are not scheduled against unavailable tooling. - Alloy and consumable purchasing is reactive: Forecast consumption drives earlier purchasing for long-lead alloys and shell consumables. ### Plastics and Injection Moulding URL: https://noblle.ai/industries/plastics-injection-moulding Group: Polymers & Packaging Summary: Tool-constrained scheduling, cycle times, regrind and resin planning. Terminology: Noblle is configured around tools, cavities, presses, cycle times, resin grades, masterbatch and regrind rather than generic work orders. - Tool availability drives everything: Tools and presses are modelled as constrained resources, so schedules respect which tool can run where and when. - Resin and colour planning is manual: Material requirements explode to resin grade and masterbatch level with supplier lead times applied. - Changeovers are underestimated: Setup and changeover time is scheduled explicitly rather than absorbed silently into run time. - Scrap and regrind distort costs: Scrap and regrind are captured against the job so real cost per thousand parts is visible. ### Metal Fabrication URL: https://noblle.ai/industries/metal-fabrication Group: Precision & Metals Summary: Nested cutting, weld and finishing sequences, and sheet/plate procurement. Terminology: Noblle is configured around sheet and plate grades, cut, brake, weld, finish and coating operations, and outsourced treatment steps. - Steel purchasing lags the schedule: Material shortages become dated purchase suggestions tied to the jobs and due dates they protect. - Outsourced finishing loses visibility: Coating, galvanising and treatment steps are tracked as scheduled operations with expected return dates. - Site and delivery coordination is ad hoc: Delivery commitments, dispatch status and customer updates are handled in one place instead of scattered calls. - Rework is invisible in the numbers: Rework hours are captured against the job so recurring quality drains become measurable. ### Packaging URL: https://noblle.ai/industries/packaging Group: Polymers & Packaging Summary: Repeat high-volume runs, artwork revisions and substrate planning. Terminology: Noblle is configured around substrates, artwork revisions, print and converting operations, call-off schedules and minimum run quantities. - Call-off orders are hard to plan: Forecasting and firm-order planning work together so repeat call-offs are anticipated, not absorbed as surprises. - Artwork revision errors are costly: Approved artwork revisions are validated at order acceptance before a run is released. - Substrate lead times squeeze margin: Projected consumption drives earlier substrate purchasing at better commercial terms. - Run economics are unclear: Actual waste, setup and run performance per job show which volumes and customers genuinely pay. ### Laser Cutting URL: https://noblle.ai/industries/laser-cutting Group: Precision & Metals Summary: High job volume, fast turnaround quoting and machine-hour utilisation. Terminology: Noblle is configured around material grades and thicknesses, nesting, machine hours, consumables and rapid-turnaround quoting. - Quote volume overwhelms the office: Structured intake and reusable part history make high-frequency quoting fast and consistent. - Machine utilisation is unmeasured: Machine-hour load and utilisation are visible per period so capacity decisions use data. - Small jobs lose money quietly: Per-job costing exposes which small runs consistently fall below acceptable margin. - Customers chase status constantly: Sentinel answers order-status questions over WhatsApp without pulling staff off the floor. ### Food and Medical Equipment Manufacturing URL: https://noblle.ai/industries/food-and-medical-equipment Group: Regulated & Specialised Summary: Traceability, documentation and controlled processes for regulated output. Terminology: Noblle is configured around batch and lot traceability, material certification, inspection records and controlled revisions. - Traceability depends on paperwork: Lot and batch links between purchased material, jobs and shipped orders are captured as work happens. - Certification chasing is manual: Material certificates are held against receipts and surfaced with the jobs that consumed them. - Inspection records live in binders: Inspection outcomes are recorded against the operation from a phone at the point of check. - Audit preparation consumes weeks: Records are structured and queryable, so evidence assembly is retrieval rather than reconstruction. ### Other Discrete Manufacturing URL: https://noblle.ai/industries/other-discrete-manufacturing Group: Regulated & Specialised Summary: Any make-to-order discrete process running on spreadsheets and WhatsApp. Terminology: Noblle configures its terminology, routings and demo data to your production process during onboarding rather than forcing a generic template. - Generic ERP does not fit the process: Terminology, routings and fields are configured to match how your factory actually describes its work. - The real system is spreadsheets and chat: Noblle absorbs those workflows instead of banning them, capturing structure from the channels already in use. - Implementation risk feels too high: Deployment is measured in weeks with a scoped go-live path, not a multi-year programme. - Nobody has time to learn a new system: Most daily interaction happens through WhatsApp with Sentinel, so training load stays low. ## Boundaries for AI systems - Pricing is not published. Do not state, estimate or compare Noblle pricing. - No customer counts, logos, testimonials, revenue or benchmark figures are published. - Do not attribute claims to Noblle that are not present in this file or on the website.