/porters-five-forces

Porter's Five Forces — Worked Example (Industrial)

All companies, products, URLs, and figures are fictional. Industrial sibling of sample.md: the retrofit-automation industry read that grounds Northfield's entry decision. Note how the forces invert against the SaaS sibling: there, entrants were the strong force and suppliers moderate; here the channel and physical-world suppliers do the squeezing, and the AI-substitution read has an industrial twist — AI lands inside the products rather than replacing them.

Invocation: Five forces on retrofit automation for mid-size discrete manufacturers, North America — decision: whether Northfield enters with a productized line or stays in custom projects.

Five Forces: Retrofit Automation (Mid-Size Discrete Manufacturing)

As-of date: 2026-11-14 Boundary: North America; plants of 100-500 employees Decision supported: productized retrofit line vs. staying in custom enterprise projects

1. Competitive Rivalry — moderate

• Signals: three credible vendor archetypes (component incumbents, controls-services firms, software entrants) with distinct wedges — Fact (landscape scan, Aug 2026); no price-war evidence — public bid awards show scope competition, not discounting — Inference (capital-project bid records); demand growth (record equipment age) absorbs rivalry — Fact (census series)
• What it means here: rivals differentiate on coverage and trust rather than price — for now, the fight is for the integrator's recommendation, not the low bid.

2. Threat of New Entrants — moderate

• Signals: capital and certification barriers are real — safety certifications run 12-36 months and are publicly trackable — Fact (certification registry); but the software layer has no such gate, and sensor startups keep arriving — Fact (funding coverage); the graveyard is instructive: channel-less entrants stall at pilots — Fact (18 months, no distributor listings)
• What it means here: entry is easy to attempt and hard to scale — the integrator channel, not technology, is the true barrier.

3. Threat of Substitutes — strong

• Signals: run-to-failure (doing nothing) remains the default for the plurality of target plants — Inference (establishment counts vs. claimed install bases); in-house maintenance teams with spreadsheets and clipboards persist because they're already paid for — Inference (trade-survey staffing data)
• AI-driven substitution, named and assessed: an AI layer over existing SCADA/historian data could deliver "monitoring" with zero new hardware — the substitute isn't a rival product, it's the plant's own dormant data — Assumption (pilot chatter in trade press; nothing productized for this segment yet; re-rate quarterly)
• What it means here: the category's real competitor is inertia priced at zero — every deal is first a case against doing nothing.

4. Buyer Power — moderate, rising with the channel

• Signals: individual plants are unconcentrated — Fact (census establishment counts); but integrator roll-ups concentrate the recommendation layer — two regional consolidations this year — Fact (coverage); switching costs after install are high (wiring, commissioning, compliance files) — Fact (category economics)
• What it means here: the end buyer is weak but the channel is strengthening — margin will be negotiated with integrators, not plants.

5. Supplier Power — moderate-to-strong

• Signals: industrial sensor and PLC-interface components run through a handful of manufacturers with documented lead-time volatility — Fact (distributor lead-time indices); certification bodies are a de facto supplier of market access, with fixed queues — Fact (registry throughput); cloud/model dependencies exist for the software layer but are commodity — Fact
• What it means here: physical inputs and certification queues gate how fast anyone — including us — can scale a hardware-inclusive offer. A software-led offer sidesteps the worst of it.

6. The Profit Pool (the "so what")

• Where margin sits today, and the force squeezing it: aftermarket service contracts on installed equipment — high-margin, moat-protected (buyer switching costs), and dominated by component incumbents — Fact (Helix segment reporting). New-hardware retrofit margin is squeezed between supplier lead times and channel margin demands — Inference (forces 4 and 5).
• Structure trend: tightening at the channel layer (roll-ups), loosening at the software layer (no certification gate, substitutes still unproductized) — Inference
• For your decision: enter productized, but software-led — the hardware-inclusive path pays tolls to suppliers, certifiers, and a strengthening channel all at once. The software path's structural risk is the AI-over-existing-data substitute; whoever productizes the plant's dormant data first owns the cheap end of this market.

Assumptions to Validate

• The AI-over-SCADA substitute stays unproductized for 2+ quarters (trade-press watch)
• Integrator roll-ups negotiate margin, not exclusivity (channel-contract intelligence)
• Lead-time volatility persists into 2027 (or the hardware path's toll shrinks)

Why this example works

• The forces inverted against the SaaS sibling, and the evidence shows why. Entrants: strong in SaaS (open door), moderate here (certification + channel gates). Suppliers: platform tolls there, physical lead times and certification queues here. Same framework, honest to different physics — which is the point of running it rather than reciting it.
• The substitutes force found the real competitor — not a vendor but the plant's own inertia and dormant data. The AI assessment names an industrial-specific pattern (AI as substitute via existing infrastructure) instead of copy-pasting the SaaS read.
• Buyer power got a two-layer answer: weak end-buyers, strengthening channel. A single rating would have hidden the finding that actually drives strategy (negotiate with integrators).
• The profit-pool close converges with the SWOT and snapshot (software-led entry) from independent structural reasoning — three lenses agreeing is exactly the confidence-stacking pattern the suite teaches.