IoT Value and Product Strategy · Study deck

Device Evolution: IoT Products and Enablers

A smart washer senses its load, checks energy prices, and changes the cycle.

Ada is your guide for this deck.

evolution
Ada, the module guide, in a scene from this chapter.
iotclass.org

After studying this chapter

Learning objectives

You will be able to:

  • Trace how sensing, edge compute, BLE, and cloud services enable an IoT product.
  • Test a product benefit against battery, privacy, and operating-cost limits.
  • Explain: Machine learning: Learns patterns and improves over time.
  • Explain: A light bulb controlled from a phone can meet the baseline IoT test, but it should not be sold as adaptive.
iotclass.org

Major section

IoT Products (2010s-Present)

Machine learning: Learns patterns and improves over time.

  • Example: A smart washing machine that automatically selects optimal washing cycles based on water usage, energy efficiency, and load type.
  • It detects fabric types via sensors, learns your usage patterns (you always wash jeans on Wednesdays), adjusts water temperature based on energy prices from the smart grid, and orders detergent automatically when running low.
  • Integrates with smart home ecosystem (starts cycle when solar panels generate excess power).
iotclass.org

Major section

Common Pitfalls and Misconceptions

A light bulb controlled from a phone can meet the baseline IoT test, but it should not be sold as adaptive.

  • A more mature lighting system may learn a routine, use ambient-light sensors, and coordinate with other devices.
  • Confusing Connectivity with Intelligence: Connectivity can establish an IoT data path, but it does not prove intelligence.
  • Ignoring the Embedded Foundation: IoT devices do not replace embedded systems -- they build on top of them.

Why it matters

The distinction matters because products lose trust when basic connectivity is sold as autonomous optimisation.

iotclass.org

Major section

Common Pitfalls and Misconceptions (continued)

A thermostat with an app is connected IoT; a thermostat that learns your schedule and adjusts from occupancy, weather, and energy prices is adaptive IoT.

  • Every IoT device still contains an embedded core that must balance the cost-power-performance triangle.
  • Assuming Linear Evolution: Not every product should become connected or adaptive.
  • A simple kitchen timer can remain embedded; it does not need ML or cloud analytics.
iotclass.org

Major section

Common Pitfalls and Misconceptions (continued)

The distinction matters because products lose trust when basic connectivity is sold as autonomous optimisation.

  • Forcing connectivity or intelligence into products where users do not need it adds cost, complexity, and attack surface without proportional value.
  • Overlooking the Ecosystem Opportunity: Ecosystem integration is not required for baseline IoT, but it can extend value.
  • A thermostat may gain useful capabilities by coordinating with a home platform, blinds, and occupancy sensors.
iotclass.org

Major section

Embedded to Connected to IoT

Early networks connect computers for direct communication; the Web connects people to information and services.

  • Mobile access removes the desktop boundary, and social plus cloud platforms join people to shared services and APIs.
  • IoT then connects things to things and people so systems can sense and act.
Evolution from embedded products to connected products to full IoT products
Evolution from embedded products to connected products to full IoT products
iotclass.org

Major section

Microwave Evolution

Scenario: A kitchen appliance manufacturer is planning their product roadmap for the next 5 years.

  • Computation: Digital timer, preset controls.
  • Value: Basic functionality, operates independently.
  • Computation: Digital controls + app interface.
  • Connected: ecobee3 lite ($140) - control remotely via app Google reports Nest users saved 10–12% on heating and 15% on cooling on average; price varies by model and date.

Why it matters

Value: Autonomous optimization, learns food types, prevents overcooking.

iotclass.org

Major section

Defining Principle

"An embedded system is a computerized system that is purpose-built for its application.": Elicia White, Making Embedded Systems (O'Reilly).

  • Unlike desktop software where you can always throw more RAM or CPU at a problem, embedded systems force engineers to make hard trade-offs.

Key terms

Embedded systems
Embedded systems are fundamental to modern electronics, providing tailored solutions to a wide range of technological challenges.

Why it matters

The middle block is the only part that differs between a microwave and an anti-lock brake, because the same processor, memory, and peripherals run a different dedicated program.

Diagram of an embedded system showing components like processors, converters, and amplifiers.
Diagram of an embedded system showing components like processors, converters, and amplifiers.
iotclass.org

Major section

Defining Principle (continued)

Purpose-Built Functionality: Embedded systems are designed to perform a single, specialized task, making them highly efficient in operation.

  • Low Power Consumption: Due to their focused functionality, embedded systems operate with minimal power requirements, allowing them to fit in small spaces and extend device longevity.
  • Cost-Effective: Embedded systems are typically inexpensive, making them an economical solution for controlling devices in various applications.
  • Optimization Requirements: Designers must optimize code and use minimal microcontroller resources to keep costs low and efficiency high.
  • Embedded systems are fundamental to modern electronics, providing tailored solutions to a wide range of technological challenges.
iotclass.org

Major section

Defining Principle (continued)

Figure: Quantitative embedded design budgets for compute resources puts numbers on why adding a radio to a product is never a free decision.

  • The Cortex-M family provides 32-bit capability, 8-96 kB RAM, 64-512 kB code flash, and sleep current below 1 uA; consumer electronics still work with typical MCU costs of $0.50-$5.
  • The BLE example adds a concrete product claim: a 30-byte packet once per second can target a year on a coin cell.
  • Connectivity therefore has to fit inside the same compute, cost, and power envelope rather than arriving with resources of its own.
Quantitative embedded design budgets for compute resources, MCU cost, and battery current
Quantitative embedded design budgets for compute resources, MCU cost, and battery current
iotclass.org

Major section

Phoebe's Field Notes: What The "Coin Cell For A Year" Claim Actually Budgets

The mathematical gist: A 225 mAh CR2032 at 3.0 V stores 0.675 Wh by nameplate.

  • One year of 1% self-discharge leaves 222.75 mAh, so charge, energy, pulse sag, and retention must be checked together.

Let Battery Bruno connect charge, watt-hours, burst current, cell resistance, and self-discharge.

iotclass.org

Deck summary

Key takeaways

Machine learning: Learns patterns and improves over time.

  • A light bulb controlled from a phone can meet the baseline IoT test, but it should not be sold as adaptive.
  • A thermostat with an app is connected IoT; a thermostat that learns your schedule and adjusts from occupancy, weather, and energy prices is adaptive IoT.
  • The distinction matters because products lose trust when basic connectivity is sold as autonomous optimisation.
  • Early networks connect computers for direct communication; the Web connects people to information and services.
iotclass.org

Retrieval practice

Recall check 1 of 6

Ada says: answer from memory, then check your reasoning.

Q1A washing machine adjusts its cycle using load and energy context. Which capability distinguishes the example from remote control?

AA phone button that repeats a manual command
BAn internet label on the appliance enclosure
CData-informed adaptation of the washing cycle
DA cloud connection without local control logic
Show answer

Answer: C The example combines sensing and logic to select and improve operating behavior.

iotclass.org

Retrieval practice

Recall check 2 of 6

Ada says: answer from memory, then check your reasoning.

Q2A simple timer gains cloud control but no useful new outcome. What should the team decide?

AKeep the cloud feature because every device should be adaptive
BPrice the timer as an intelligent service because timing automation itself creates ongoing service value.
CMove timing logic to cloud because remote scheduling provides the required local control.
DTest whether benefit justifies energy, privacy, security and service costs.
Show answer

Answer: D The added feature needs a measured benefit that outweighs its operating limits.

iotclass.org

Retrieval practice

Recall check 3 of 6

Ada says: answer from memory, then check your reasoning.

Q3A phone-controlled bulb is marketed as adaptive lighting. What should the review require for that stronger claim?

AProof that the phone can send a command
BA higher price for the connected label
CRemoval of the embedded control requirements
DEvidence that it learns, decides, and adapts
Show answer

Answer: D The maturity test distinguishes basic connectivity from adaptive behavior.

iotclass.org

Retrieval practice

Recall check 4 of 6

Ada says: answer from memory, then check your reasoning.

Q4A startup is designing a pet feeder. Version 1 dispenses food on a fixed schedule. Version 2 adds Wi-Fi so owners can trigger feeding from their phone. Version 3 uses weight sensors, pet activity tracking, and ML to automatically adjust portion sizes based on the pet's health data. Which version can most justifiably charge a monthly subscription fee, and why?

AVersion 1 - because it is the most reliable
BVersion 2 - because it requires cloud servers for the app
CVersion 3 - because it delivers ongoing value through personalized health optimization
DAll versions can equally justify subscriptions
Show answer

Answer: C A recurring fee is most defensible when the service keeps improving feeding decisions from new weight and activity data.

iotclass.org

Retrieval practice

Recall check 5 of 6

Ada says: answer from memory, then check your reasoning.

Q5Why was the ARM Cortex-M processor considered a "game-changer" for IoT development?

AIt was the first processor that could run Linux
BIt provided 32-bit computing capability with sleep currents below 1 microamp
CIt was the cheapest processor ever manufactured
DIt was the first processor with built-in Wi-Fi
Show answer

Answer: B The processor closed a design gap: older low-power 8-bit chips limited work, while earlier 32-bit designs drew too much battery power.

iotclass.org

Retrieval practice

Recall check 6 of 6

Ada says: answer from memory, then check your reasoning.

Q6Which 2013 event highlighted BLE for consumer proximity and beacon use cases?

AApple incorporated BLE into iBeacon in 2013
BGoogle released the Eddystone beacon protocol
CThe Bluetooth SIG released the BLE specification in 2006
DSamsung included BLE in the Galaxy S3
Show answer

Answer: A BLE-capable phones predated iBeacon.

iotclass.org

Print reference

Answers 1 of 2

Answer key.

  1. C · The example combines sensing and logic to select and improve operating behavior.
  2. D · The added feature needs a measured benefit that outweighs its operating limits.
  3. D · The maturity test distinguishes basic connectivity from adaptive behavior.
  4. C · A recurring fee is most defensible when the service keeps improving feeding decisions from new weight and activity data.
  5. B · The processor closed a design gap: older low-power 8-bit chips limited work, while earlier 32-bit designs drew too much battery power.
iotclass.org

Print reference

Answers 2 of 2

Answer key.

  1. A · BLE-capable phones predated iBeacon.
iotclass.org