15 Business Model Cases: Proposals and Economics
More
Depth: Practitioner
Progress saves as you read. Mark sections complete as you go; your progress appears in the sidebar.
- Podcast audio coming soon
- Character narration audio coming soon
15.1 Turn the Service Idea into an Approval Decision
A facilities team understands why lighting could become a service, but it still has to approve a contract. The proposal must show cash flows, risks, payback timing, and what changes if equipment life, energy prices or service promises differ from the sales case. A university campus manager replacing 50,000 aging fixtures across 15 buildings gives us a running decision to follow.
This is part 2 of 2. Business Model Cases: Evidence and Service Value supplies the evidence and service-value background. Here, the task is to turn that idea into an evidence-bound Lighting-as-a-Service proposal, compare lifetime cost and discount-rate effects, then evaluate hardware subsidies and governed data revenue before selecting a fit.
The problem is that all three approaches can promise recurring value while placing very different demands on cash, operations and trust. We will first compare the campus proposals, then examine payment timing and service responsibility. The smart-speaker and agricultural examples test whether the same discipline holds when the revenue comes from ecosystem activity or insights.
KEY CONCEPTS
- Service responsibility
The lighting provider carries the maintenance and replacement promises.
- Cash flow and lifetime cost
Payment timing changes the comparison between purchase and service.
- Subsidy recovery
Recurring contribution must repay the entry-device loss before churn.
- Governed insight revenue
A buyer, useful insight and consent framework come before collection.
15.2 Learning Objectives
By the end of this chapter, you will be able to:
- build an evidence-bound Lighting-as-a-Service proposal
- compare lifetime cost, payback timing, and discount-rate effects
- evaluate razor-and-blade and governed data-revenue models
15.3 Compare the Campus Proposals
This proposal comparison is an illustrative campus-scale model. It uses round numbers to teach lifetime cost and discounted cash-flow reasoning; it is not reported Schiphol contract pricing.
The campus facilities manager receives two proposals for the 50,000 aging fluorescent fixtures across 15 buildings.
Proposal A (Traditional): Purchase 50,000 LED fixtures outright at $200 each = $10M upfront. Estimated 15-year maintenance cost: $3.75M ($250K/year). Annual energy cost: $5.256M (calculated at $0.12/kWh, 200W average per fixture, 12 hours/day).
Proposal B (Lighting-as-a-Service, or LaaS): Zero upfront cost. Monthly service fee: $45K/month ($540K/year) for 15 years = $8.1M total. In this illustrative model, the service provider handles maintenance, replaces fixtures after 7 years, guarantees 99.5% uptime, and uses 50% less electricity than Proposal A’s LED installation.
The manager needs to compare Total Cost of Ownership (TCO), the costs included over the full 15-year period, before choosing between the proposals. The following arithmetic keeps hardware, maintenance, energy and service fees separate.
For Proposal A, add the upfront purchase to maintenance and energy over the contract period:
- Initial hardware: $10M
- Maintenance (15 years): $3.75M
- Energy: $5.256M/year x 15 = $78.84M
- Total: $92.59M
For Proposal B, maintenance and replacement are included in the service obligation, so add service fees and energy:
- Service fees: $540K/year x 15 = $8.1M
- Energy (50% less than Proposal A): $2.628M/year x 15 = $39.42M
- Total: $47.52M
The resulting comparison gives the facilities team three different decision inputs:
- Savings with LaaS: $92.59M - $47.52M = $45.07M (48.7% TCO reduction)
- Cash flow advantage: No $10M upfront fixture purchase; the service contract’s balance-sheet treatment depends on whether it contains a lease.
- Risk transfer: The illustrative service provider takes on the maintenance and replacement responsibilities stated in this model.
Key Insight: In this illustrative model, Proposal B is assumed to use 50% less electricity than Proposal A’s LEDs. The reported Schiphol reduction was relative to conventional lighting. The model’s energy and service assumptions drive the comparison; they are not reported Schiphol contract terms.
The calculator is a sensitivity tool: increasing energy rates or operating hours makes the assumed efficiency difference more valuable. Its original default service fee is $2 per fixture per month, which is $100K/month for 50,000 fixtures, rather than the proposal’s $45K/month. Set the fee consistently before comparing results; calculator defaults and a written proposal are separate assumption sets.
15.4 Account for Timing and Contract Risk
A lower lifetime bill does not remove the need to examine when money leaves the business. Subsidized equipment, installation labor, onboarding and service operations are paid early, while recurring revenue arrives slowly. The campus customer avoids a $10M initial purchase under Proposal B, but the provider still needs patient capital to fund hardware, support and maintenance before the contract repays that investment.
A 15-year contract also needs time-value-of-money discounting. $540K paid in Year 15 is worth far less than $540K paid in Year 1.
At a 5% discount rate, the $8.1M nominal LaaS service-fee stream has a present value of about $5.61M. The traditional purchase’s $10M upfront payment stays at $10M PV because it’s paid immediately.
For a multi-year IoT contract, calculate Net Present Value (NPV) by discounting each future payment to its value today. The technical detail below gives the payment formula and rate choice.
In the corrected 200 W university scenario, the 15-year nominal LaaS saving is $45.07M; at 5% the NPV saving is $34.27M.
The campus decision record should show both the numerical result and the obligations that depend on the contract:
| Factor | Traditional Purchase | Product-as-a-Service | Winner |
|---|---|---|---|
| Upfront Cost | $10M | $0 | PaaS |
| 15-Year TCO (nominal) | $92.59M | $47.52M | PaaS |
| 15-Year TCO (NPV at 5%) | $67.15M | $32.88M | PaaS |
| Balance Sheet Impact | Upfront fixture purchase | Contract-dependent; assess whether it contains a lease | Depends on contract |
| Technology Risk | Customer owns the fixtures | Maintenance and replacement terms are contract-dependent | Depends on contract |
| Flexibility | Owns assets, can sell | Locked into 15-year contract | Purchase |
NPV = Σ (Payment_year / (1 + discount_rate)^year). The annual payment formula discounts each year’s payment by (1 + rate)^year, then adds the discounted amounts. Immediate expenditure is not discounted. Use the organization’s Weighted Average Cost of Capital (WACC) as the discount rate and assess tax treatment separately because it depends on the contract and applicable accounting and tax rules.
In this scenario, discounting reduces the value of both future cost streams. The $34.27M NPV saving is smaller than the $45.07M nominal saving; deferred LaaS payments do not make NPV savings automatically greater than nominal savings. Comparing like-for-like timing prevents that mistaken interpretation.
15.5 Connect the Revenue Model to Operating Responsibility
The campus spreadsheet depends on service delivery. Product-as-a-Service sells an assured asset outcome, such as illumination, while the provider retains the lighting assets and service responsibility. Philips/Signify’s Lighting-as-a-Service at Schiphol is the public example; the campus prices and performance promises remain illustrative.
Figure 15.1 shows the stages of Philips’ move from fixture sales to Lighting-as-a-Service, each building on the one before.
LED fixtures become easier to compare, making service, financing, and maintenance more important differentiators.
Philips shifts from selling fixtures outright to charging for illumination outcomes, starting with Schiphol Airport.
Managed lighting contracts turn the buyer relationship into a long-running operating account instead of a one-time fixture sale.
LaaS reaches airports, hospitals, warehouses, and offices while Philips leans on service operations and financing at scale.
Lighting contracts increasingly combine efficient hardware, maintenance, upgrades, and circular-economy responsibilities.
The transformation worked because Philips paired lower customer upfront cost with long contracts, uptime guarantees, and retained ownership of the lighting infrastructure.
LED fixtures look like a commodity business, so Philips needs a defensible revenue model.
Customers pay monthly for illumination outcomes while Philips owns maintenance and replacement risk.
Recurring service contracts show that lighting can be sold as an operating outcome, not only as installed equipment.
Managed services remain strategically useful because they connect efficiency, maintenance, upgrades, and asset circularity.
Philips' shift from fixture sales toward managed Lighting-as-a-Service contracts, using Schiphol as the public launch example.
In Figure 15.1, the journey spans 2010–2023. In 2010, the commodity squeeze makes LED fixtures easier to compare, increasing the importance of service, financing and maintenance as differentiators. The 2015 Schiphol launch shifts attention from selling fixtures outright to charging for illumination outcomes. In 2018, managed lighting contracts illustrate a continuing operating account in place of a one-time fixture sale. The 2020 stage describes expansion across airports, hospitals, warehouses and offices; by 2023, the service model combines efficient hardware, maintenance, upgrades and circular-economy responsibilities.
Read those stages as an operating journey: financing, maintenance, customer trust and circular asset handling have to mature together. Lower customer upfront cost works alongside long contracts, uptime guarantees and retained ownership. The dates alone do not establish the provider’s margins or the campus proposal’s feasibility.
The four archetypes organize different revenue promises:
| Model | What generates revenue | Operating signal and example |
|---|---|---|
| Product-as-a-Service | Outcome-based subscription | Philips LaaS retains asset and service responsibility. |
| Subscription / razor-and-blade | Hardware subsidy plus recurring services | A smart-speaker ecosystem must repay its subsidy through attach-rate revenue. |
| Pay-per-use | Metered consumption or completed activity | Rolls-Royce Power-by-the-Hour links maintenance pricing to customer activity. |
| Data monetization | Aggregated analytics sold to third parties | Agricultural telemetry insights need governance and consent. |
Example: Philips LaaS
Revenue: Outcome-based subscription
Signal: Provider retains asset and service responsibility
Example: Smart speaker ecosystem model
Revenue: Hardware subsidy plus recurring services
Signal: Subsidy only works when attach-rate revenue repays it
Example: Rolls-Royce Power-by-the-Hour
Revenue: Metered usage tied to customer activity
Signal: Price follows actual consumption, not ownership
Example: Agricultural telemetry insights
Revenue: Aggregated analytics sold to third parties
Signal: Governance and consent matter as much as the data itself
Comparison of the four major IoT business model archetypes and the revenue logic behind each one.
A subscription exchanges continuing access or service for recurring payment. Pay-per-use follows actual consumption rather than transferring ownership. Rolls-Royce TotalCare charges engine maintenance by engine flying hour and transfers maintenance risk; that meter does not establish who owns the engine, and airlines may own or lease aircraft. These distinctions keep billing frequency separate from the obligation being sold.
15.6 Test Whether Recurring Contribution Repays the Subsidy
Lighting-as-a-Service keeps the provider close to the asset. The next pattern lowers the entry-device price and depends on recurring ecosystem value to repay the subsidy before churn. An attach rate is the share of device customers using a particular paid service or revenue-generating activity.
Scenario: A smart-speaker provider sells a hub below cost to grow its installed base. The ecosystem model below is illustrative: it assumes revenue from music subscriptions, smart-home purchases, and shopping margin. The point is to test whether the recurring attach-rate revenue repays the hardware subsidy.
The central question is whether a $75 subsidy can be recovered from $594 of ecosystem inflow over 3 years. The answer also needs service and acquisition costs. If that inflow falls to $150, the same cost discipline matters even more: the smaller total leaves less room to fund delivery and recover the entry-device loss.
A razor-and-blade strategy sells or subsidizes the entry device to drive recurring services. Low-cost hardware reduces adoption barriers, but the model fails if attach rates, retention, or service margin do not repay the subsidy.
Revenue Breakdown (3-Year Scenario Ecosystem Inflow):
| Revenue Source | Attach Rate | Monthly Revenue | 36-Month Total |
|---|---|---|---|
| Music streaming | 30% | $10 x 0.30 = $3 | $108 |
| Smart home platform fee | 40% | $50 x 0.20 x 0.40 = $4 | $144 |
| Voice shopping margin | 5% | $100 x 0.10 x 0.05 = $0.50 | $18 |
| Membership/commerce uplift | 60% | $14.99 x 0.60 = $9 | $324 |
| Total scenario ecosystem inflow | - | ~$16.50/month | $594 |
Business Model Comparison:
| Strategy | Hardware Pricing | Revenue Source | Illustrative Example |
|---|---|---|---|
| Razor-and-Blade | Below cost (subsidy) | Recurring services | Compare contribution after service costs with the subsidy |
| Platform Model | Market rate | Transaction fees | Different: no hardware subsidy |
| Freemium | Free software | Paid upgrades | Different: software, not hardware |
| Outcome-Based | Varies | Results achieved | Different: not ecosystem revenue |
Financial Calculation:
- Hardware loss: -$75 (average subsidy per device)
- 3-year scenario ecosystem inflow: +$594
- Ecosystem inflow less hardware subsidy, before service and acquisition costs: $519
- Profit, ROI, and breakeven cannot be calculated without service margins and other costs.
Why This Works in the Model:
- Low adoption barrier: $59 price point vs $200+ competitors
- Ecosystem lock-in: Voice shopping, music, smart home create switching costs
- High-margin services: Digital services can earn higher gross margins than hardware when their incremental delivery costs are lower
- Platform network effects: More devices leads to more developers leads to better ecosystem leads to more devices
Calculation note: The $594 ecosystem inflow is a scenario assumption, not a reported company metric or contribution LTV. In a real model, product teams should validate attach rate, retention, service margin, and attribution before assessing whether a hardware subsidy is repaid.
Similar Razor-and-Blade Models:
- HP Instant Ink pattern: Printers and ink subscriptions show how recurring consumables can subsidize competitive hardware pricing
- Peloton: Bikes with monthly class subscriptions ($1,495 bike, $528/year subscription)
- Kindle: Devices subsidized, e-book revenue ($120 device, $15/book x 20 books/year = $300)
For a device costing $110 to manufacture and selling for $59, the subsidy is $51 rather than the $75 baseline above. The same $594 inflow assumption still needs service and acquisition costs before it can support a contribution, ROI or payback estimate. Keep every assumption visible: subsidy, monthly margin, attach rate, churn, contract length, discount rate and support cost. Changing them can reverse the conclusion.
The Financial Metrics and Analysis chapter develops customer lifetime value (LTV), customer acquisition cost (CAC), churn and payback calculations. Pair LTV:CAC with payback and churn risk before scaling: an attractive lifetime total can still leave the business short of cash.
15.7 Count the Costs Before Calling Data a Product
Once recurring value is clear, data can look like the next revenue stream. Data monetization creates revenue from governed information products. Telemetry becomes commercially useful when a buyer can use the resulting insight to make a decision.
Many IoT companies assume that collecting massive amounts of sensor data automatically creates monetization opportunities. The belief is: “We’ll gather all the data we can, then figure out how to monetize it later.”
Four costs and constraints need to be examined before treating telemetry as a revenue stream:
-
Storage Costs Exceed Revenue: The original scenario budgets $23–50/month per TB for IoT time-series storage (AWS S3/Timestream); treat this as a modeling assumption rather than a verified tariff. A smart building with 500 sensors generating 1 MB/day each creates 15 GB in a 30-day month, or 0.015 TB, rather than 15 TB. At the assumed linear rate, that new volume represents about $0.345–0.75/month of storage. The original $345–750/month estimate would apply to 15 TB. Retention and analytics still add costs; without a clear buyer, collection remains an expense.
-
Data Without Insights Is Harder to Sell: Raw sensor readings (temperature: 22.3C, humidity: 45%) have limited value without context. Buyers pay for actionable insights such as which schedule change lowers heating, ventilation and air conditioning (HVAC) energy use. The transformation from data to insight requires analytics infrastructure (additional cost).
-
Privacy Rules Constrain Monetization: GDPR, CCPA, and sector-specific regulations (HIPAA in healthcare, FERPA in education) constrain data use and sharing where they apply. Compliance costs ($50K-500K for data governance systems) often exceed potential revenue.
-
Anonymization Reduces Value: Removing personally identifiable information (PII) is part of anonymization, but removing those fields alone does not establish that a data product is anonymous or that resale is permitted. Anonymization can reduce the granularity advertisers buy, but its effect on commercial value depends on the data product and buyer.
The original case examples illustrate trust and governance risks. Treat the company-specific strategy and causation claims below as claims requiring source evidence before using them in an approval proposal:
| Company | Data Collection Strategy | Outcome | Lesson |
|---|---|---|---|
| Fitbit (pre-Google) | Collected detailed health data, explored selling to insurers | User backlash, privacy concerns, strategy abandoned | Users don’t trust health data monetization |
| Facebook Portal | Smart display whose usage data, such as call frequency and duration, could inform ad targeting | Poor sales (privacy concerns), discontinued 2022 | In-home surveillance too invasive for consumers |
| Smart TV manufacturers (Vizio) | Sold viewing data to advertisers without clear consent | $2.2M FTC fine (2017), required explicit opt-in | Implied consent insufficient, explicit required |
| Ring (pre-acquisition) | Police partnerships accessing doorbell footage | Public outcry, policy changes, trust damage | Law enforcement data sharing harms brand |
These examples distinguish a possible buyer benefit from the costs and trust needed to deliver it. Raw readings such as temperature 22.3°C and humidity 45% do not by themselves establish a market. A recommendation about an HVAC schedule is a product only if it improves a decision the buyer values enough to pay for.
15.8 Govern the Insight and the Farmer’s Value Exchange
With the cost limits visible, work backward from the buyer’s decision rather than collecting broadly and searching for a buyer later.
| Wrong Strategy | Right Strategy | Revenue Impact |
|---|---|---|
| Collect everything, monetize later | Define monetization strategy first, collect only needed data | Reduces unnecessary storage |
| Sell raw data dumps | Sell curated insights/analytics dashboards | Can increase revenue when insights support a buyer decision |
| Assume consent (“implied by usage”) | Explicit opt-in with clear value exchange | Avoids regulatory fines ($50K-$5M+) |
| Generic data marketplace | Vertical-specific insights (agriculture, smart cities) | Can raise willingness to pay when the insight fits a specific workflow |
Data Monetization Success Formula:
- Start with Customer Problem: What decision does the buyer need to make? (Energy procurement, maintenance scheduling, inventory optimization)
- Work Backward to Required Data: Collect only sensors/metrics needed for that decision
- Build Analytics First: Develop insight generation before scaling data collection
- Establish Consent Framework: Explicit user opt-in with transparent value exchange
- Calculate Unit Economics: Ensure (insight revenue per user) > (collection cost + storage cost + compliance cost)
The agricultural telemetry example puts the sequence into practice:
- Problem Identified: Farmers need yield optimization recommendations
- Data Collected: Soil moisture, yield maps, weather; the illustrative collection plan excludes GPS tracking and personal data
- Insight Generated: “Plant corn variety X in northeast field for 12% yield increase”
- Revenue Model: Sell aggregated, consented insight products to seed companies, insurers, or research partners while giving operational analytics back to participating farmers
- Consent Model: Farmers explicitly opt-in, retain data ownership, can revoke access
- Result: Data revenue is tied to trust, transparency, and a clear value exchange rather than raw data extraction
The operating rule is to define the buyer, value proposition and consent framework before scaling collection. The farmer needs useful operational analytics, while the commercial buyer needs a defensible insight product. Those are related benefits, but the contract must still say who can use which data and for what purpose.
Fixed compliance costs create a minimum viable scale in the model; small deployments have fewer customers over which to spread them. Increasing sensor count only helps when the resulting insight revenue covers the added collection and operating burden. Track insight revenue per user against collection, storage and compliance costs, including the analytics needed to make the insight usable.
The consent-led decision sequence is therefore: name the paying buyer and decision; collect only the required telemetry; build the recommendation, score or benchmark before scaling collection; secure opt-in, ownership boundaries, anonymization and revocation; then validate per-customer economics. If there is no buyer yet, collecting everything creates storage cost and governance risk. If there is no insight layer, raw dumps have little pricing power. If consent is missing, backlash and regulatory exposure can erase the upside.
15.9 Select the Model by Its Proof Metric
Return to the campus manager’s approval problem: a proposal needs evidence for customer value, operating cost, adoption and the risk transferred between partners. The models can now be compared by what must be proved, rather than grouped together as recurring revenue.
| Model | Key metric | What to check |
|---|---|---|
| Product-as-a-Service | Payback and service margin | Contract life must cover equipment, maintenance, financing and support. |
| Razor-and-blade | LTV-to-subsidy ratio | Recurring attach-rate contribution must repay the hardware loss before churn. |
| Data monetization | Insight revenue per user versus collection cost | Revenue must remain positive after consent, governance, storage and compliance work. |
| Platform | Network effect strength | Each participant group should make the others more valuable. |
A smart-home platform connects device manufacturers, app developers and consumers. More devices attract developers, more apps attract consumers, and more consumers attract manufacturers. These network effects create value through participation across groups; transaction fees and a share of app revenue monetize that activity. Making switching difficult is a different mechanism. A maker selling its own sensor with a paid monitoring tier needs evidence for subscription revenue rather than assuming a multi-sided marketplace exists.
For the campus proposal, retained ownership connects Product-as-a-Service to subscription pricing. For subsidized hardware, CAC joins subsidy recovery in the funding decision. For agricultural telemetry, privacy regulation and the farmer’s value exchange shape the insight product. IoT Business Model Fundamentals develops CapEx versus OpEx, recurring revenue and LTV. Pricing Strategies connects willingness-to-pay, competitive benchmarks and value-based pricing to subscription prices, freemium tiers and outcome-based offers.
After choosing a model, Go-to-Market Strategy develops B2B launch strategies, sales cycles and pilot-to-scale frameworks. Illustrative math supports that decision process, but it does not become public company financial evidence unless a source actually reports the figures. Preserve opt-in, revocation, aggregation and anonymization boundaries as the product moves toward launch.
15.10 Summary
The campus comparison shows how an illustrative $10M purchase and $45K/month service offer produce different lifetime costs, cash flows and responsibilities. The $92.59M versus $47.52M nominal totals depend heavily on the energy assumption; the 5% discounted comparison is $67.15M versus $32.88M. Contract terms still determine maintenance, replacement, lease treatment and flexibility.
A hardware subsidy needs measured attach rates, retention and service margin to repay the entry-device loss. A data business needs a defined buyer, useful insight and explicit value exchange before scaling collection. Patient capital funds the gap between early hardware and operating costs and later recurring receipts. Consent, revenue sharing and granular opt-in support trust; the Vizio $2.2M penalty, Fitbit backlash and Ring trust concerns show why governance belongs in the economics.
Approve the revenue model only when its customer value, payment timing, operating obligations and evidence support one another. A promising lifetime total still needs a feasible payback period and a service the provider can deliver.
Continue to Financial Metrics and Analysis to calculate LTV, CAC, churn and payback for the proposal you can defend.

