Pricing a Machine That Predicts Breakdowns
No competitor, no reference price. Build the number from the customer's P&L.
The Prompt
Your client built an IoT retrofit kit for textile-mill machinery: vibration and temperature sensors plus software that predicts loom failures 48 hours ahead. Pilot results: unplanned downtime cut by 60%. There is no direct competitor in India. The client asks: how do we price this — per sensor, per machine, per mill? And at what number?
Opening exchange
With no reference price, this is economic-value pricing: quantify what a breakdown costs the mill, how much we prevent, and take a fair share of that. First — what does one hour of unplanned loom downtime cost a typical mill, and how many such hours do they suffer?
A mid-size mill runs 80 looms. Each loom contributes about ₹1,800/hour. Unplanned downtime averages 22 hours per loom per year, plus each breakdown event costs ~₹15,000 in emergency repairs versus ₹6,000 planned. Typical mill: ~70 breakdown events a year.
Everything needed for an EVC (economic value to customer) build is now on the table. The case is won by doing this math cleanly.
I'll build the value pool per mill per year, apply our 60% prevention rate, then split that surplus between customer and client — B2B convention says the customer keeps the larger share, or adoption stalls. Then choose the metric (per loom/month) to scale naturally with mill size.
Structure & Hypothesis
Analysis & Data
Why 30% capture and not 50%? And why per-loom-per-month instead of selling the hardware outright?
On the split: this is a new category sold to skeptical, cash-tight mills — the pilot proves value we believe; the buyer hasn't lived it yet. A 70/30 split in the customer's favour makes the ROI pitch trivial: "pay ₹6.8 lakh, save ₹23 lakh." As the category matures and trust builds, capture can rise toward 40–50% on renewals or premium tiers. On the metric: outright hardware sale caps revenue at one transaction and makes us a capex line competing with a new loom; per-loom-month is opex, scales with mill size, keeps us paid for the software's ongoing value, and builds a recurring base a future acquirer will pay for.
Two classic B2B pricing arguments: share-of-surplus calibrated to buyer risk, and metric chosen for adoption + recurring revenue.
A large mill group with 1,200 looms demands 40% off. Respond.
Never discount the rate card 40% — it reprices the whole market through word of mouth. Instead: volume tiers built into the metric (e.g., ₹700 first 200 looms, ₹560 beyond), an enterprise SLA tier that adds value instead of cutting price, and multi-year lock-in as the concession currency. Headline rate survives; their effective rate lands near ₹590.
Recommendation
Recommend
- Price at ₹699 per loom per month, subscription including hardware, installation, and the prediction software — no upfront capex for the mill.
- Sell with the customer's own P&L: "keep ₹16 lakh of the ₹23 lakh we save you" — the 70/30 split is the sales pitch, not a concession.
- Handle large groups with volume tiers and SLA-tier upsells; never cut the headline rate.
- Re-anchor capture toward 40% at renewal once realized savings are in the customer's own data.
Key Takeaway
What this case teaches
When there's no reference price, build the customer's value pool and split it — and remember the metric (per what?) is as strategic as the number. New-category B2B pricing buys adoption with a generous split, then earns capture back with proof.