Shortage-resolution autopilot for hardware supply chains that reallocates inventory and execution plans across ERP and logistics stacks.

1. Large enterprises are already buying overlay autonomy that sits on top of existing systems, which lowers the adoption barrier for a new exception layer.
2. The trust boundary has moved from analytics to action because the category now writes updated plans back into ERP and logistics tools instead of just flagging issues.
3. Inventory reallocation is explicitly moving into autonomous software, making shortage triage newly productizable rather than forever human middleware.
4. A $50 million Series B behind Auger shows there is real capital and urgency behind the category, not just pilots.

Catalyst. Auger's funding and named customers show enterprises now budget for overlay autonomy that acts inside their existing supply-chain stack rather than another visibility dashboard.

Build a system that ingests order, inventory, shipment, and planning data across the customer's existing stack and assembles a real-time exception view for one product family or launch program. For each shortage event, the product scores tradeoffs across service level, revenue, margin, and transport cost, then generates a recommended reallocation and execution plan. The first version keeps humans in the loop with one-click approvals and guaranteed write-back into existing systems instead of replacing them. Over time, every accepted or rejected action becomes training data for policy tuning, which lets the software take on more of the shortage desk autonomously.

What's different. This is not a visibility dashboard and not a full planning-suite replacement. The product owns the high-stakes last mile of exception execution: who gets scarce inventory, what gets decommitted, and which logistics plan is pushed back into the system right now. Its defensible asset is the decision-policy graph linking customer priority, SLA, margin, and network state to realized outcomes, plus the audit trail and write-back integrations that make deeper autonomy trusted over time.

Jobs to be done

flowchart LR
  Buyer[Supply chain control-tower lead] --> Pain[Launch shortages force manual reallocation]
  Pain --> Product[Shortage-resolution autopilot]
  Product --> Outcome[Higher fill rate with less expedite spend]

Executive takeaways

• The wedge is real: both new entrants and incumbents now frame the unsolved problem as turning cross-stack signals into coordinated action, which means the market has moved beyond pure dashboards but is not yet saturated with shortage-first point solutions. [2][21][24][27]
• Buyer urgency is episodic but acute: launch shortages, OTIF penalties, tariff shifts, and transportation shocks can destroy service and margin quickly, so a one-product-family pilot can attach to an already-budgeted war room. [7][12][25]
• The best first sale is not “autonomous supply chain” writ large; it is human-approved shortage resolution for one business unit, with write-back into existing systems and clear recovered-revenue, expedite, and cycle-time proof. [5][19][20][31]
• Competitive intensity is real but fragmented across planning suites, visibility networks, and AI-native overlays; a startup can still win if it lands faster than suites and goes deeper on policy-driven allocation than logistics-heavy towers. [18][21][24][27][33][34][35]

Market definition

Software that sits above ERP, planning, warehouse, and transportation systems to resolve supply-chain exceptions—especially shortages—by ranking trade-offs, generating cross-system actions, and progressively moving from recommendation to governed execution. [3][5][21][24][29]

Customer and buyer

The day-to-day user is a control-tower lead, allocation planner, supply-demand planner, or network planner who is already reconciling conflicting signals across orders, inventory, logistics, and customer priority. The economic buyer is usually a VP Supply Chain, COO, or business-unit GM, with an IT/data counterpart when write-back and governance matter. [2][4][32]

Buying triggers

• A supplier miss, demand spike, or launch slip creates immediate manual allocation calls across regions, channels, and customers. [2][6][8]
• Rising tariff, export-control, or forced-labor-compliance pressure changes sourcing or flow decisions fast enough that spreadsheet response becomes too slow. [7][12][13][14][15]
• OTIF exposure, expedites, and duplicated or shorted orders make the cost of late exception handling visible to finance and customer teams. [25][26][28][32]

Willingness to pay

This is enterprise-operations spend, not productivity SaaS. Buyers already pay for buffer inventory, expedites, OTIF misses, and large planning programs; case studies and industry commentary suggest six-figure ACVs are plausible when the tool measurably improves in-stock performance, reduces inventory, or shortens response times. [7][10][19][25][28][32]

Category dynamics

Growth signal Adjacency signal: 4.4% 2025 U.S. consumer technology revenue growth

Validation signals

• Auger already sells an overlay autonomy layer to named enterprise customers and says one Fanatics workflow is mostly autonomous.
• Atomic says pilot customers cut inventory while preserving service, showing buyers will test narrow AI-led planning workflows.
• Doss and Didero both pitch AI layers that sit above existing ERPs rather than full system replacements.
• FourKites and Blue Yonder now market inventory-risk detection, recommended mitigation, and execution orchestration instead of visibility alone.
• Loop’s large round shows investors also see value in supply-chain intelligence layers built from messy operational data.

Regulatory & technical constraints

• Any autonomous action that changes origin, route, or supplier can interact with tariffs, export controls, or forced-labor documentation obligations.
• AI-assisted operational decisions increasingly need logs, documentation, and human oversight to be governable at scale.
• Safe write-back depends on real APIs and transaction models across ERP and inventory systems, not just dashboards.
• Planning and control-tower users already expect inventory, allocation, and scenario simulation inside existing suites, raising the integration bar for entrants.

Competition splits into three camps: planning suites with deep simulation and enterprise trust (Kinaxis, o9, Blue Yonder), visibility/control-tower networks with richer external execution data (FourKites, Blue Yonder), and AI-native overlays that sit on top of ERP rather than replace it (Auger, Doss, Didero, Loop in adjacent workflows). Many buyers still bridge the gaps with spreadsheets, email, and manual ERP updates. [3][18][21][24][27][33][34][35]

Why incumbents do not win by default

• ERP and data-cloud platforms. Systems of record and data foundations are necessary, but they do not by themselves decide how to reallocate constrained inventory, escalate trade-offs, or coordinate cross-functional response.
• Planning suites. Kinaxis, o9, and Blue Yonder already model supply, inventory, and response at enterprise scale, but they typically enter as broader transformation platforms rather than a fast shortage-desk wedge.
• Visibility and network-control towers. Visibility networks contribute crucial ETA and partner-state data, yet they are strongest when the problem is logistics awareness; a shortage-resolution wedge can still differentiate on customer-priority policy, decommit logic, and governed write-back.
• In-house war rooms and consultants. Human coordination remains the default because companies trust spreadsheets and meetings more than autonomous actions, but that also leaves the response loop slow, person-dependent, and hard to scale.

Launch Shortage Autopilot should start as a shortage-resolution operating layer for North American consumer-electronics, AR/VR, and connected-device business units that still run weekly launch war rooms. The product is not a planning-suite replacement; it ingests order priority, inventory, shipment, and transport state across ERP, WMS, TMS, and planning tools and produces a human-approved reallocation, decommit, or expedite plan for one launch-critical product family. The first customer is a $1B+ hardware brand with SAP or Oracle ERP, a separate planning stack, three or more regional fulfillment nodes, and an upcoming launch or supplier slip that makes spreadsheet triage too slow. Research supports the category because Auger, FourKites, Blue Yonder, Kinaxis, and o9 all now market visibility-to-action workflows, and Auger shows enterprises will buy overlay autonomy without replacing core systems. The first sale should be a paid pilot on one business unit that proves faster alert-to-action cycle time, lower expedite spend, and preserved fill rate before broader automation rights are requested. The market is large enough for a venture wedge only if expansion works: current estimates are $1.2B TAM, $125.0M beachhead SAM, and $9.0M modeled year-3 SOM. The main execution risks are dirty cross-system data, buyer reluctance to permit governed write-back, and competitive bundling from suites or better-funded overlays. The key gaps are the true count of qualifying North American business units and the phase-one approval boundary, so the first 90 days must focus on ICP mapping, data-readiness audits, and governance design.

Problem

• Control-tower teams still resolve launch shortages with spreadsheets, email, and manual ERP, WMS, and TMS transactions because no system adjudicates trade-offs across order priority, inventory location, in-transit stock, and customer commitments in one loop.
• When a component slips or demand spikes, planners must decide which orders to short, which nodes to reallocate, and whether to expedite freight after the selling window has already narrowed.
• Visibility and planning tools flag risk but rarely push an executable, audit-ready cross-system plan back into the stack, so fill rate, launch revenue, and margin leak during every major exception.

Solution

• Ingest order, inventory, shipment, ETA, and business-priority data for one product family and present a live shortage queue with recommended reallocate, decommit, substitute, or expedite actions.
• Keep planners in the loop at first with role-based approvals, confidence scores, and one-click write-back into incumbent systems rather than asking the customer to replace ERP or planning software.
• Learn from accepted and rejected actions to tune shortage policy by channel, customer tier, SLA, margin, and node state, then expand into governed auto-execution for low-risk cases.

Why we win

• The wedge is narrower and faster to deploy than Kinaxis, o9, or Blue Yonder because it sells a shortage desk, not a full planning transformation.
• Vendor-neutral write-back across mixed ERP, WMS, TMS, and planning stacks is harder for logistics-heavy visibility networks or in-house war rooms to reproduce.
• The decision-policy history linking priority rules, approved actions, and realized service or margin outcomes becomes a proprietary asset that broad suites do not capture from spreadsheet-driven workflows.
• Audit trails and human-approval logs directly address tariff, export-control, forced-labor, and AI-governance concerns that slow autonomous action adoption.

Milestones

flowchart LR
  Wedge[One product-family shortage desk] --> MVP[Human-approved reallocation and write-back]
  MVP --> Proof[Faster cycle time and lower expedite spend]
  Proof --> Expansion[Multi-corridor hardware accounts and adjacent verticals]

Founding team

Experiment roadmap

Risk assessment

What must be true

• At least 15 North American consumer-device business units match the beachhead profile and run recurring shortage war rooms with budget authority.
• At least 5 of the first 8 design-partner prospects can expose order, inventory, shipment, and priority data and permit governed write-back into at least one core system within 45 days.
• Human-approved pilots cut median shortage alert-to-action time by 30%+ and reduce expedite spend or decommit volume by 10%+ on a live product family.
• At least 50% of paid pilots convert to $300k+ annual production contracts within 120 days of pilot completion.
• Bundled suite, overlay, or internal alternatives win fewer than 75% of the first 12 competitive evaluations.

Open diligence questions

• How many North American consumer-device business units actually fit the beachhead and own shortage P&L?
• Which ERP, planning, WMS, and TMS combinations dominate the first 10 target accounts, and what write-back objects are available in practice?
• What baseline fill rate, expedite spend, decommit volume, and alert-to-action metrics can design partners share before a pilot starts?
• What approval boundary will supply-chain leaders, IT, and internal audit accept in phase one, and what evidence expands that boundary?
• Which budget line pays first inside the account—business-unit operations, control-tower modernization, or broader supply chain transformation?
• When buyers compare Auger, Kinaxis, o9, Blue Yonder, FourKites, or internal teams, what exact gap makes them switch?

Model sanity

• Revenue engine. Base revenue is driven by growing from 2 active paying business units at M12 to 15 by Q4Y3 while corridor value climbs from pilot pricing to roughly $600K mature ARR.
• Must go right. Standard connectors and human-approved write-back must compress pilot-to-production cycles toward one quarter, or the jump from 6 paying units at Q4Y2 to 15 at Q4Y3 will slip.
• Model breaks if. If production value stalls near $500K ARR or gross margin stays below the 70% target, Y3 EBITDA stays negative and the $3.0M pre-seed no longer carries a six-month buffer comfortably.
• Next-round proof. The seed story is 6 production business units, 2 referenceable logos, 25% partner-sourced pipeline, and 1 governed low-risk auto-execution deployment with cash still above roughly $1.0M.

Scenarios

Sensitivity

flowchart LR
  TargetAccounts[Target hardware business units] --> PaidPilots[Paid pilots]
  PaidPilots --> ProductionBUs[Production business units]
  ProductionBUs --> ExpansionModules[Node and channel expansion]
  ExpansionModules --> Revenue[Revenue]
  Revenue --> GrossProfit[Gross profit]
  GrossProfit --> Cash[Cash]

Flags: The step from 6 active paying business units at Q4Y2 to 15 at Q4Y3 is ambitious and depends on the partner channel becoming real, not merely promised. · customersEop includes paid pilots and production business units, so recurring-only production count lags the headline paying count through most of Y1 and early Y2. · Gross margin only reaches the BP target of 70% if connector reuse materially reduces bespoke implementation; dirty customer data could hold margins in the mid-60s. · Cash is modeled as EBITDA, so milestone billing timing, implementation prepayments, and enterprise procurement delays could shift the actual cash curve. · The $125M beachhead SAM is concentrated, so one or two delayed target accounts can move Y2 bookings and the timing of the next round materially.

• Autonomy trust gap. Supply-chain leaders may refuse autonomous write-back on high-stakes allocations until governance is proven. Mitigation: Start with human approval, role-based guardrails, explainable tradeoff scoring, and audit-ready logs before expanding to auto-execution.
• Dirty cross-system data. Late or inconsistent order, inventory, and transport data can make a recommended reallocation wrong at the moment it is executed. Mitigation: Launch on customers with stable interfaces, score every action for data confidence, and fall back to recommendation-only mode when data freshness drops.
• Incumbent suite response. ERP, planning, or control-tower vendors could add adjacent exception workflows once the ROI becomes visible. Mitigation: Win by integrating across mixed stacks, owning the cross-system decision-outcome dataset, and delivering value on one exception corridor faster than suites can ship.