# Ansoff Matrix — Worked Example (Industrial)

All companies, products, URLs, and figures are **fictional**. Industrial sibling of
[`sample.md`](sample.md): Northfield Automation maps its growth options after the retrofit
investigations. Two contrasts with the SaaS sibling worth noticing: the diversification quadrant
is *populated* this time (with the evidence bar visibly applied), and market-development moves
carry certification and channel gates that have no SaaS equivalent.

**Invocation:** `Ansoff growth options for Northfield Automation — core: software-defined
automation projects for enterprise manufacturers, North America. Outcome: +50% revenue in 36
months. Constraints: no acquisitions; hardware only through partners.`

---

# Ansoff Growth Options: Northfield Automation
**As-of date:** 2026-11-18
**Current core:** custom software-defined automation projects for enterprise manufacturers, North America
**Growth outcome sought:** +50% revenue in 36 months; constraints: no acquisitions, hardware via partners only

## 1. Market Penetration (existing product, existing market — lowest risk)
- **Expand share-of-wallet in existing enterprise accounts: monitoring add-on to completed projects** — signal: 60%+ of past projects have no ongoing software subscription — **Fact** (internal contract data); completed-project sites are pre-integrated — **Fact** — risk: **low** — known buyers, deployed footprint
- **Win-back sweep of stalled proposals citing "timing"** — signal: capital-spend survey shows modernization budgets recovering — **Fact** ([trade survey](https://example.com/nam-survey)) — risk: **low**

## 2. Market Development (existing product, new market)
- **Down-market: productized retrofit monitoring for 100-500-employee plants via integrators** — signal: the full investigation chain (scan → snapshot → forces) supports it — mixed-vendor gap, channel structure, profit-pool read — **Inference** (stacked across three artifacts) — risk: **medium** — the integrator-margin assumption remains unvalidated after three artifacts flagged it
- **Canada, same enterprise motion** — signal: two inbound RFQs from Canadian multi-plant operators this year — **Fact** (internal) — risk: **low-medium** — certification transfers under existing mutual-recognition arrangements — **Fact** ([regulatory guidance](https://example.com/regs/mutual-recognition))

## 3. Product Development (new product, existing market)
- **Machinery-safety documentation automation for enterprise customers** — signal: new documentation burden is regulatory fact — **Fact** ([register](https://example.com/regs/machinery)); zero competitors ship it — **Fact** (absence, three vendors checked); customers already ask our services team to compile these files manually — **Fact** (services backlog data) — risk: **medium** — real build, but demand is documented in our own paid work
- **AI anomaly-detection layer over customers' existing SCADA/historian data** — signal: the five-forces read named this the category's looming substitute; better to own it than meet it — **Inference**; pilot interest from two existing accounts — **Fact** (internal) — risk: **medium-high** — model performance on messy plant data is unproven

## 4. Diversification (new product, new market — highest risk)
- **Warehouse/logistics monitoring (Meridian's home turf, in reverse)** — signal: our asset-agnostic architecture transfers the same way theirs does — **Inference** — but no expressed demand from logistics buyers exists in our evidence base, and we'd enter *their* fortress exactly as they enter ours — risk: **high, rated honestly**. Verdict: logged, not recommended — the symmetry is seductive and the evidence is one architecture inference. The quadrant holds one entry to show the bar, not to greenlight it.

## 5. Recommended Sequence (the "so what")
- **First:** monitoring add-on into existing accounts (penetration) — strongest evidence, near-zero acquisition cost, and it builds the subscription muscle every later move needs
- **Then:** safety-documentation automation (product development) — demand documented in our own services backlog; ships to buyers we already have; becomes a differentiator *inside* the down-market move
- **Then:** the integrator-channel retrofit line (market development) — launched with the documentation feature as its wedge, per the battle card's evidence
- **Not yet:** the AI anomaly layer as a *product* — run it as two paid pilots first; the substitute threat clock (five-forces watch) sets the urgency, not enthusiasm
- **The assumption that breaks this sequence:** integrator margins. If the channel won't carry us profitably, market development recedes and the AI layer jumps the queue as the direct-sale differentiator.

### Assumptions to Validate
- Integrator margin economics (flagged in four artifacts now — validate before any 2027 commitment)
- Services backlog demand for documentation converts to product subscription, not one-off fees
- AI pilot performance on two real plants' historian data

---

## Why this example works

- **Diversification is populated *and* rejected** — the opposite lesson from the SaaS sibling's
  empty quadrant, and just as important: the entry shows what the extraordinary-evidence bar looks
  like when a seductive move (mirror-image entry into Meridian's market) fails it in writing.
- **The sequence compounds through capabilities, not just cash.** Penetration builds subscription
  muscle → documentation automation becomes the down-market wedge → the channel move launches
  armed. Industrial growth sequences are capability ladders as much as funding ladders.
- **A four-times-flagged assumption finally gets a deadline.** The integrator-margin question has
  survived the scan, snapshot, watch, and now the matrix; here it's named the sequence-breaker
  with an explicit validate-before date. Assumptions that survive multiple artifacts unvalidated
  are the ones that kill strategies.
- **The substitute became a product candidate.** The five-forces read ("own the dormant-data play
  before someone else productizes it") shows up here as a staged, pilot-gated move — frameworks
  feeding frameworks, with the risk gradient still enforced.