Hey Friends đď¸,
Previously weâve talked about the hardware product development process which takes an idea from concept to launch. But what about before that? How do companies develop innovation processes and new technologies which serve as a precursor to product development?
Before the iPhone we needed touch and camera technologies to be ready for productization. Before the Vision Pro, Apple spent almost a decade developing display technologies to achieve a high resolution, small form factor. Before Tesla became mainstream, EV batteries had to be capable of sustaining range and thermal conditions.
Today weâll be covering đ
Vision or reasons for innovation
Types of innovation
Technology development process
Ideation
Alignment to customer/technology themes
Prioritization
Testing
1. Vision
Innovation is a means to an end and doesnât happen in a vacuum. It grows in process driven environments and is a team endeavor. In other words itâs a multi variable outcome which happens with intent and structure, contrary to the lone inventor myth.
This typically starts by companies defining strategic goals, or why they need innovation. If the vision is not articulated or the r&d activities are not serving a specific goal then time and effort will be misdirected. Stephen Wunker in the Innovative Leader talks about some applications below.
Use Cases for Innovation
Industry trend anticipation
New technology response
Inefficiency improvements
Cost reduction or revenue generation
Company culture or branding enhancement
Traditional Ways to Measure Innovation
Technology development
Patents
New product releases
In this article weâll be focusing on the technology investigation side of things. However before diving in, itâs worthwhile to understand the technology s-curve. This is a model that illustrates how technologies mature in the market.
Birth Phase: Is characterized by slow development, delayed adoption, and high cost due to poor understanding of the technologyâs value.
Ex: Lead-acid batteries used in the first electric vehicles which were limited in range and performance.
Growth: Is when adoption speeds up and enhancements are made such as reduced cost and performance improvements.
Ex: Advancements in lithium-ion battery technology led to gains in range, affordability, and efficiency for the industry.
Maturity/Decline: Is when the technology becomes a market standard and reaches a point of saturation, slowing growth.
Ex: Current day batteries focusing on improving recycling processes and energy density.
2. Types of Innovation
Among the many types of innovation, there are three which are particularly relevant to physical product development.
Roadmap Innovation: Evolution of products from one generation to the next to stay competitive, like making them faster or lighter.
Ex: Improved ergonomics and haptic feedback on Playstationâs DualShock controllers.
Breakout innovation: Significant changes which alter an existing product category, usually having a major impact on the industry.
Ex: Dysonâs bagless vacuum cleaners which, through cyclonic technology, offered significantly better suction and efficiency. This changed the vacuum industry.
Disruptive Innovation: Technologies which create a new product category or market.
Ex: Google Nestâs thermostat which introduced a smart device that learns usersâ habits to optimize HVAC energy usage and enhance convenience. Other examples are Uber and Tesla.
3. Technology Development Process
Idea generation is typically not the bottleneck. Most organizations have no shortage of ideas. But what prevents them from innovating is lacking systematic frameworks to nurture these ideas to fruition.
A good technology introduction process needs to allow for fluidity, iteration, risk taking, ambiguity, and complexity. It also requires input from diverse stakeholders - product, design, operations, legal, marketing, and engineering.
Stage 1: Ideation
We start with inputs, which are basically new ideas. They usually come from three sources.
Academia
A timeless source of new ideas has been through universities. This is from professors, researchers, or scholars that provide new scientific or technical knowledge.
An example is Dr. Gierad Laput, who prior to creating machine learning technologies for products at Apple, published several academic papers comprising of sensing, gesture recognization, and human-computer interaction.
He later worked on productizing key new technologies into Apple devices such as activity recognition, crash detection, and hands free interaction on Apple Watches and AirPods.
However the reality is that most of the time academia lags versus the industry given that companies have incentive to productize, determine technology robustness, and move quickly. This is why firms usually hire professors as research or applied scientists.
Structured Activities
Companies like Meta run periodic hackathons while some teams at Amazon have internal incubators for new technology investigations. Competitions, innovation committees, accelerators, and internal start-up programs allow firms to foster creative environments for new ideas.
Some organizations even encourage their employees to take a portion of their time (15-20%) for brainstorming in areas unrelated to their current work assignment.
Friction
Friction is the most organic source of idea generation. Examples include:
Resolving inconsistencies or frustrations within a process or product
Failure in one domain but same concept being applied to another
Changes in perception or industry popular opinion
Stage 2: Align and Review
As mentioned above, ideas canât be pursued arbitrarily or in isolation. They need to be aligned with key areas of interest which intersect with a companyâs strategy. Two major themes are customer experience and technology roadmaps.
Customer Experience
These are themes which enhance user experience, provide customer delight, and are critical towards solving pain points that are aligned with company vision. Some industry examples include:
Wearables
Comfort and fit: Developing ergonomic, lightweight solutions that accommodate longterm use.
Automotive
Range anxiety reduction: Maximizing longevity of EV batteries.
Consumer Electronics
Intelligent updates & privacy: Reducing notification fatigue ensuring robust data protection.
Smart Homes
Frictionless home activity updates: Offering latency free access and interoperability with multiple sensor, camera, and device feeds.
Technology Roadmaps
There are various technology categories which companies see as advantageous to invest into. These include, but are not limited to: audio, sensing, biometrics, displays, batteries, materials, recycling, edge processing, and more. Some industry examples:
Wave guides
Some of the most sophisticated technology is in augmented reality waveguide development. Think of waveguides as display systems. Theyâre optical components that direct light from a display into a userâs eyes by refracting it through a serrated lens. Theyâre critical for designing slim, light AR glasses.
(Left) how wave guide displays work, (Right) refraction index vs FOV and immersiveness. Source: coherent.com Wave guides have existed for years but have only recently been on the path to mainstream productization.
Technology development studies in this area include materials design (plastics, glass, coatings), micro-structure engineering, light coupling enhancements, FOV improvements and more.
Multi Modal Sensor Fusion
This involves the integration of various sensing modalities (audio, cameras, motion, PIR) to lower HW costs and improve detection accuracy using ML.
For example, combining microphone, accelerometer, and video input from smart home devices to confirm which door was opened, and who opened it.
Another example is in smart wearable devices where sensors like accelerometers, temperature, and ECG are used to determine physical activities (walking, running, sleeping) and monitor health.
Multi modal sensing in wearable devices. Source: Sanfilippo and Petterson. Computer Vision
CV enhancement will be a key technology theme for many companies that use cameras. Think smart homes, automotive, security, and inspection.
Smart home companies will need to improve facial recognition algorithms to recognize home owners vs visitors in various lighting conditions while integrating anonymization techniques to protect privacy.
Another application will be deep learning models to detect and classify activities like package deliveries, suspicious behavior, or animals playing.
Stage 3: Prioritize
The next step, as Wunker and co describe in the Innovative Leader, is that companies need scorecards to holistically and objectively prioritize ideas given the finite nature of time and resources. A simple example template is shown below with criteria and stop light colors for each.
Stage 4: Test
The last, and sometimes longest, step prior to exiting from a technology development process into a product development process is to vet out the selected tech using a series of experiments, tests, builds, and prototypes. The goal is to validate feasibility.
A great example of this is mentioned in the Vanity Fair, where we get an insight into Tim Cookâs impressions of the Vision Pro long before it was productized.
Note the level of new technology investigations that had to be completed prior to packaging the device into its final form factor. One instance is engineering micro-OLED displays to fit 23 million pixels in front of a userâs eyes. Others are captured by the 5000+ patents and 7+ years of development.
Also consider how the early demos look drastically different than the final commercialized product.
âMike Rockwell, vice president of Appleâs Vision Products Group, is there when Cook enters and sees it. Itâs like a âmonster,â Cook tells me. âAn apparatus.â Cookâs told to take a seat, and this massive, monstrous machine is placed around his face. Itâs crude, like a giant box, and itâs got screens in it, half a dozen of them layered on top of each other, and cameras sticking out like whiskers. âYou werenât really wearing it at that time,â he tells me. âIt wasnât wearable by any means of the imagination.â And itâs whirring, with big fansâa steady, deep humming soundâon both sides of his face. And this apparatus has these wires coming out of it that sinuate all over the floor and stretch into another room, where theyâre connected to a supercomputer, and then buttons are pressed and lights go on and the CPU and GPU start pulsating at billions of cycles per second andâŚTim Cook is on the moon!
âŚ..What Cook didnât know is how his engineers were going to take this thing that needs a supercomputer in another room, and fans and multiple screens, and shrink it down to the size of a pair of goggles that weighs a little more than a box of spaghetti. âIâve known for years we would get here,â Cook told me. âI didnât know when, but I knew that we would arrive here.â
After all of the above comes step 5, a formal exit from the technology development process into the NPI (new product introduction) process. In reality, while only a few ideas make it all the way through, itâs important to not feel discouraged.
An inspiring team is Google X or Googleâs moonshot factory which is an innovation lab that focuses on solving complex problems with audacious technologies. While theyâve had some massively successful productization examples like Waymo (self driving cars) theyâve also had a considerable amount of failures.
Sometimes the industry is just not ready for the technology (like Google glass at the time) and thatâs just a part of the journey.
Thanks for Reading
Make sure you check out some other articles:
If you enjoyed this, like and spread the word. If you have any questions on hardware PMing, product development, or careers letâs talk.
Read because of the visuals restacked in a note, stayed because of the content, subscribed because of everything you provide. Great post!
Reading this article was inspiring! Thank you for sharing. I love how you broke down the âinnovationâ. Also Tim cooks example was the cherry on top.