Sunday, March 29, 2020

Re-inventing the Product : From Industry 4.0 to Digital Experiences




1. Introduction

This blog post will talk about product development, digital manufacturing and the necessary integration between software and hardware engineering. This topic is dear to my hear because of the many years that I have spent at Bouygues Telecom developing new internet gateways and set top boxes. I will talk today about Eric Schaeffer and David Sovie’s book, “Reinventing the Product”, which is one of the important sources for my own new book, to be published in the month to come. The focus on product makes this book a great companion to the larger vision of the Industrie 4.0 transformation, as described in the Acatech document « Industrie 4.0 Maturity Index - Managing the Digital Transformation of Companies » from Günther Schuh, Reiner Anderl, Jürgen Gausemeier, Michael ten Hompel and Wolfgang Wahlster.

Re-inventing the product is clearly about smart products, but it is also about new products built in smart, connected factories. It is driven by technology and the introduction of digital tools throughout the lifecycle, from design to usage and operations. It is also a transformation of mindset, culture and organizations. As a part of the larger digital transform, re-inventing the product starts with the customers, since the digital world allows to re-invent the customer relationship from listening to supporting. Physical products still require a strong emphasis on manufacturing and product lifecycle, but software engineering, software ecosystems and the power of platforms have become equally important ingredients of the product success. Because software moves fast, because “AI is eating software”, the role of software strategy plays an ever-increasing role with product differentiation and innovation.

I had the pleasure to invite Eric Schaeffer to give a talk at the NATF. He has been a keen observer of the manufacturing industry for many years, leveraging key positions at Accenture. His previous book “Industry X.0 : Realizing Digital Value in Industrial Sectors” was already dedicated to the digital transformation of the manufacturing world. “Re-inventing the product” tells the story about what happens when companies have deployed digital tools and connected product architectures, to transform these products into platforms and start selling outcomes and experiences instead of products and services (a simplistic but useful characterization being that B2B companies are now selling outcomes and B2C companies are selling experiences). Digitization transforms the value chain and moves value to the edges: the few who are the world best-in-class with technology and the companies that control the experiences. As always, digital transformation is about speed: Eric Schaeffer quotes Haier, who is about to move from the capture of a new customer need to the delivery of a new product in 30 days. This capacity is illustrated with a wonderful story about understanding from failure analysis that customers in Asia were using their washing machine to clean sweet potatoes, which lead to the design and delivery of a new specialized – and more robust – product for this very use.

2. Reinventing the Product



This blog post is structured about “Reinventing the Product – How to transform your business and create value in the digital”.  I have sorted the key take-aways in six subsections, that reflects both the relevance with my own product experience and how well suited these principles and examples are to the challenges that I can observe today. Thus, this is as usual a subjective and personal summary of this book. “Re-inventing the product” is especially rich with figures and diagram, which I will only describe briefly and will not reproduce. Therefore, I strongly advise you to get your own copy of this book after reading this summary. The authors have collected a very, very rich set of testimonies and quotes from key industry players. I have selected a few in this blog post (which is, therefore, not as short as I would have wished) but there is no way that I could do justice to the breadth of scope of the work assembled by Eric Schaeffer and David Sovie. This does not mean that the book covers all the possible aspects of smart product design (it is much too large a scope). The authors spend a lot of energy covering the interaction between software and hardware design, but only mention briefly the “systems of systems” aspect of modern products. In many cases, the “smart behavior” of a product is not the sum of the embedded intelligent software components, but the emergent consequence of “system of smart system” design.

2.1  Products in the digital age


The authors start their analysis of smart and connected product design with a useful two-dimension framework that illustrate the rising importance of experience design and contextual intelligence of products, thanks to the combination of connection to the environment and artificial intelligence. They propose to name IQ and EQ the two axes of this analysis : “Developing smart connected products is a journey … we have boiled down this journey into two key dimensions that are show on the Product Reinvention Grid … a product’s Intelligence Quotient (IQ) – the level of smartness, connectedness and cognitive intelligence – and it’s Experience Quotient (EQ), reflecting the quality of experience it can offer”.  Smart connected products are more than the combination of embedded intelligence and better design. The book proposes ten traits to characterize them : always on, sensorized for awareness, AI-smart, hardware value amplified by software, evergreen via update, digital age UI (user interface), Hyper-personalized, A platform for multiple parties, embedded in ecosystems, digital continuity throughout lifecycle (from design to recycling/grave). Evergreen is a critical concept here, it means that the connected product keep constantly up-to-date and in synch with its support software ecosystem, which is critical for cybersecurity.

Digital continuity is a key principle of Industry 4.0: it means that the digital data that represents, and with associated with, a product will follow the product throughout the product lifecycle, from design and manufacturing to delivery and usage. This makes the digital framework a collaborative platform for all company functions to work together, using the digital representation as a shared object. This digital continuity is described by the authors as “digital thread as an eternal umbilical cord”: “The brother, if you will, of the digital twin is the digital thread. It is the umbilical cord of a smart connected product that is never cut once a device, coming from complex multi-sourced delivery, has been released into the market”. Olivier Ribet, from Dassault Systems, talks about the use of Digital Twin as a collaborative platform: “The technology used to model and simulate an experience end-to-end makes you fast as creator and manager of a product. It orchestrates every step and weeds out processes that you do not need. It also can cut out a lot of linearity. So, designers, engineers, factory workers, distributors and marketers all work with one single version of the truth”.

When they become smart and connected, products become parts of large software ecosystems, a key idea that I will explain further later on. This makes the choice of ecosystem a strategic decision for product companies: “In fact, a large part of the business leader’s management skills will be about building or joining the right ecosystems. Businesses have two basic options here: they suggest and create a partner ecosystem defining the way they work, or they join an existing ecosystem that is already well established with widely accepted software or hardware standards”. The fact that software becomes a critical part of product development means that the product development process must evolve and adapt parts of software development methodologies, such as agile development. The authors quote Steve Myers, the CEO of Mindtribe : “Today there is still a big gap between engineering and product development. Product development is not taught in engineering schools and most engineers are developing products in ways that seem rational and logical, but that don’t benefit from the many lessons learned … my dream is to share what we have learned applying agile techniques to hardware to enable every team developing a product to spend more of their time developing things that matter to people and less time on things that don’t.”         



2.2 Start with the customer


Focusing on customer experience is a critical landmark of digital transformation. As the authors point out, thinking about experience is a quantum leap beyond designing for features and services. The principle of experience design forces to become user centric and to focus on the emotional aspect of the outcome, a combination of the perceived value (valence) and the expected satisfaction (arousal) : “The perception of an experience is formed by the interaction of the human with their environments and consists of two emotional qualities, arousal and valence”. The focus on experience design is critical to leverage the capabilities of digital to deliver personalized journeys. The analysis of customer journeys deliver the context that is necessary for successful experience design: “The way products are designed, engineered, manufactured and supported when in use will change. The work of research and development teams will be less and less predictable as manufacturers will be forced to embed new and not yet fully mature technologies to keep up with fast moving consumer markets”. Tesla is quoted as an example of experience focus: “one of the most important tasks for the car-maker’s designers is the development of the user interfaces integrating the vehicle capabilities to a unique experience from the user”.

This no longer a requirement for B2C companies only, B2B companies need to embrace customer journeys and experience design: “It has long been well known that customer experience is top priority for consumer-facing businesses. Now business-facing companies have started taking this seriously too … This is clearly because they know mere product features and functions are losing traction and the capacity to create traction in the marketplace”. As the authors point out, the product companies’ missions have moved from “understand consumer usage & expected product attributes to design and live customer experience journeys”. The book uses many concepts, principles and techniques that one would find in digital service design methods such as growth hacking. For instance, tracking “aha moments”: “Leading companies identify what we refer to as the key ‘moments that matter” across the customer journey”. Experience design now applied to B2B companies: “This has created a new and special discipline – experience engineering – focused on designing, creating and managing products while they are in use – with the single goal to create an experience for maximum user satisfaction”. Experience design for B2B companies it is actually harder: “In the B2B world the experience happens at enterprise levels. There are multiple touchpoint in parallel, which makes the experience much more complex to manage”.

The “Customer Feedback Learning Loop” (CFLL) is a key concept linked with the digital transformation.  CFLL is about learning from the customers, using three paths: the explicit (conversation), implicit (analytics) and social (communities) ways. There are multiple examples of CFLL described in this book, especially at Tesla. A Tesla manager says: “Continuous monitoring of real-world performance and usage data allows product makers to see and fix design problems that testing missed”. I mentioned the Haier example earlier in the introduction. Samsung tells a similar story about continuous design of products thanks to customer feedback: “Consumer listening is an important part of innovating home appliances. We do have data analytics teams. We have teams doing core research. We have teams who improve data analytics algorithms. We employ specialists looking into the data to come up with usage insights”.

2.3 Software is eating the product


The authors have a saying, “software is eating hardware, and digital is eating software”, which says that digital has brought a new way of producing software, together with new software capabilities, which have a huge impact on smart and connected products. When they talk about “digital software skills”, they mean for instance: design thinking, scrum masters, AI experts, product managers or cloud architects. Software becomes a critical component of product design: “The product architecture is becoming software-driven, which needs to be reflected in the overall product engineering approach”. This includes software engineering “inside the box” and “outside the box” (product as a platform that is part of an ecosystem). To succeed with this new challenge, it is critical to integrate hardware and software engineering. The book provides this quote from James Heppelmann, the CEO of PTC : “The engineering teams don’t know that much about the data concerns, the security concerns, the failover concerns and the cloud analytics capability of the business. … The IT people within the organization know a lot about that, but they never have been involved in a product delivery process”. A similar thinking is found with Dassault Systems: “That is the essential logic behind these technologies. From the first day you think about a smart connected product, you don’t want to separate development and engineering from product usage and ultimately the end-to-end experience it provides”. One of the key reasons for this tight integration is the need for speed: “The most important challenge is getting enough speed. Software can so quickly be changed and pushed into the market via downloads. The same pace is hard to achieve on the hardware side”.
Since software is eating the product, information systems classical strategic issues, such as modular architecture or data exchange infrastructure, becomes part of smart product design strategies as well. A key message from the authors is to look at software companies to import their best practices: “experts from software-making companies are role models for the more traditional product-making companies”. The book makes a few references to the common bimodal architecture pattern (system of records vs system of engagement, or CORE vs FAST IT). The authors insist on the importance of common, shared data models: “The data models used by the vast majority of product companies are decades old, and fundamentally ill-equipped for smart connected product world”. Shared data models are the necessary foundations to make data circulate between different roles and functions, which is the only way to extract all the value from the data that digitalization has produced. This is still a major challenge for many companies: “Today most companies are far, far away from such a unified business model. In a recent study, over 50 per cent of product companies reported having over 20 different product systems”. This deep (and business-focused) work on common business objects models is necessary to implement a company-wide version of the digital twin: “The digital twin is a complete digital representation of a physical product, including not only 3D modelling, but also the material properties, the software and data. The digital twin becomes the single version of the truth of all product-related master data”. 

Mastering the software engineering skills and the software development process is a way to leverage digital transformation towards a massive (10x) improvement in product development efficiency. The following is a simplified version of Figure 8.3 of the book.



2.4   AI is eating software


As explained earlier, software development in the digital age is evolving rapidly. This book, like many other great books that I have reviewed in this blog, clearly sees AI as a modality of software systems. Digitization has created the foundations that makes it possible to derive smart algorithms that are curated from the data that was collected : “More broadly, what has happened is this : over decades, in both hardware products, manufacturing method, electronic, micro-electronic and digital technologies have grown in importance, laying the foundations for increasing level of software. This in turn has created a bedrock for Artificial Intelligence”.  AI gives product designers the ability to develop smart and adaptative behaviors, at different scales (from product components to systems). Although smart connected products can leverage remote intelligence from back-end support systems, not everything happens in the cloud; embedded smart software gives smart products more flexibility and resilience : “AI-driven software is to be embedded to make the hardware intelligent while operating in the field”. The same idea is developed by Rajen Seth from Google: “It is clear that there will be multiple layers where intelligence can reside. It can sit at sensor level or at a product level. It can sit of an edge device such as a server in a retail store or on a mobile base station and if course it can reside in the cloud”.
This transformation to AI-driven software is not without challenges: “Major challenges for industrial manufacturers when embedding AI and digital technology in their product and services are : Data quality, Data/cyber security, the choice of making vs. partnering”. Agile organizations, and the close collaboration of engineering, product, business and data science skills is necessary to deliver value from AI algorithms: “Self-organizing, product-focused teams sit at the core of development activities around smart connected products. Only they can muster the agility for quick regrouping that is needed when products are so closely connected to their makers”. According to the authors, most operators in smart factories welcome this transformation towards digital manufacturing and smart products : “A 2018 global study, based on a survey of 14000 workers … found that 68% of highly skilled workers and nearly half of their lower-skilled peers speaking positively about the impact intelligent technologies will have on their work”. Although the will to leverage AI to improve products see to be ubiquitous, the book points out that many companies are lagging behind their implementation strategy : “So, belief is not the issue – yet many industrial players still seem to struggle to realize their AI dreams. Although most know they need to change, only 24 per cent recognize that digital reinvention drives their top- as well as bottom-line growth … Product companies need an increased sense of urgency to build AI capabilities and embed them in their product and experience roadmap”.

2.5 The power of platforms


A key idea of the book is that platforms play the role of a fulcrum to leverage the energy and contribution of other players: “the rise of the smart connected product is part and parcel with the development of seamless ecosystems that make life less stress from for users and more customized to the needs of the individual “. The book addresses briefly the strength of platforms and the network effects:  “All successful platforms create what are called network effects where the value of the platform increases as the number of users and the usage increases”, although I would suggest that other books give a much deeper view on the topic of platforms. What matters the most is that the platform is not a prerequisite but an asset that emerges as a co-development of the product: “The emergence of the product as a flexible and living platform goes hand in hand with the emergence of the ecosystem that builds organically around it”.  Tesla gives many great examples of this platform vision: “Smart connected products all have a platform character in my view. In many cases they are not one single platform but a portfolio of platforms”. One of Tesla manager offers this great quote: “Tesla absolutely views itself as a platform company and regularly published software updates that fundamentally upgrade the car … The same Tesla car a customer acquired in 2013 is today a much better car that it was when it was bought – due to permanent software updates”. The authors underline here to move from aging products to evergreen products.

The book gives multiple examples of types of platforms: Marketplace Platforms, Social & collaboration platforms, Sharing Platforms, IoT platforms and Developer Platforms. It then illustrate the platform approach with many examples, such as Ford who is building the Transport Mobility Cloud, a platform of mobility services for smart cities, a in a joint effort with Autonomic. The book proposed many other examples, such as: GE Predix, Schneider EcoStruxture Power, MyJohn Deere, Faurecia Cockpit Intelligence Platform or  Haier Cosmoplat.  Building a platform requires careful strategic positioning to make sure that a company known who it can count as allies and which other platforms they may be competing against: “Our view is that all product-based platform companies need to have a clear friend-or-foe strategy when it comes to defining their relationship to the Internet platform titans and other rivals”. The book includes some 2017 numbers about R&D spending from the better-known company (Amazon B$ 22.6, Google B$ 16.6, Samsung B$ 14.9, Volkswagen B$ 14.8, Microsoft B$ 13.9, Huawei B$ 13.3, Intel B$ 13.1, Apple B$12.1), to illustrate the importance to think carefully about the leading ecosystems and the need to leverage the strength of the major tech companies. Patrick Koller, CEO of Faurecia explains the importance of understanding your position in the global ecosystem: “As a system integrator we understand the full value chain of the cockpit of the future. So we can identify points where it does not make sense for us to invest on our own because the entry cost are too high and because you would anyway have world leaders, with stronger market positions on hand you could partner with, who are recognized experts in these domains”.

Not all software aspects in a smart product need to be seen as a platform, especially since cyber-security is of the most important challenge of smart product design. Steve Myers has this to say when he is asked about the most urgent topics to be tackled in a smart product world: “Product security comes to my mind; it is a really big deal in any IoT landscape. It is very easy to overlook all the different ways your product needs to be secure. We, as well as our clients, have to build more skills into product security; that is why we work with outside security auditors to review potential weaknesses”.  The need for security means that some aspects of the product software need to be kept proprietary. The authors recall the following testimony from a Caterpillar engineer: “We keep the system closed. In that regard our connected machinery is not a platform concept. We are not creating a Linux solution open to third parties when we create an expert system for a building project. The main reason is operational safety on the side, but another reason is data safety, as attempts have already been made to hack into construction equipment. In that light it is better to keep our systems closed and thoroughly encrypted”.

2.6 Enterprise transformation


Reinventing the product is a transformation journey for most companies. We have already seen it implicitly throughout the previous sections, to design, to build and to operate new smart connected products that deliver the expected outcomes and experiences, companies also need to re-invent themselves: “The move to “as a service” is an enterprise-wide challenge, based on five pillars : Business and product strategy, Enterprise operating model, Product & Experience Innovation process, Shared Product platform, Agile Workforce”.  The Haier company has completely reorganized itself, with a spectacular reduction of management to move from a consolidated and uniform group to a federation of one thousand smaller product companies. These new forms of organizations need to accommodate new ways of working and to be more attractive for top talents: “Top talent is likewise vital for delivering top-class experiences. And, as we have shown, to retain the best personnel, organizations must deliver an engaging work experience”. As we showed in Section 2.2, re-inventing the product means rethinking the role of the customer. As mentioned by a Tesla manager, a smart product like the Tesla car finds itself in a permanent optimization mode : “Due to its quality as a smart and connected product, you can follow in a meticulous way how the product is being used. You see which features are more relevant to the customer and which are not. You can adjust interfaces along their actual usage patterns. You can prioritize features that are most commonly used and deprioritize that that are less used”.

Smart product companies need to organize themselves to bring together multiple forms of skill and engineering, such has hardware, software engineering, design, operations and data management. The following quote from a Tesla manager is a great blueprint for such an organization: “There is the universal recognition that all three engineering components that deliver the final product are equally important and need maximum dedication. There is the physical product itself, the combination of hardware and software that provides the operation of the vehicle. Then there is the development of the user interfaces integrating the vehicle’s capabilities to a unique user experience for the user. And then there is thirdly the back-end infrastructure that monitors, manages and enhances the fleet of products in the field, enables improvement via software updates and operates the data collection. Tesla recognizes that all three need to be executed extremely well; most companies don’t yet have that done”. The book makes a few references to exponential organizations, as described by Salim Ismael : “To bring exponential organizations to life requires diverse, complementary teams and smart systemic steering. But once this has been achieved, they can be 10 times faster, better and cost-effective compared to their rivals”. Without any surprise, the need for experimentations is a cornerstone of smart product companies, as illustrated by this quote from Jeff Bezos : “If you double the number of experiments you do per year, you are going to double your inventiveness”, which is very similar to Thomas j. Watson famous quote: “If you want to increase your success rate, double your failure rate”.

There is no standard recipe to design the perfect organization for the product company of the future: “While there are standard components and skills required for a digital innovation factory, there is no standard organizational model. Every company must create its own, based on its existing position within the product evolution space and its ambition for the future”. However, the need for intense communication, colocation of various expertise and deep collaboration is quoted by most successful product companies. Tesla advocates for the use of one-roof organizations, which reminds me of the same expression that we borrowed at Bouygues Telecom ten years ago when we launched ourselves into the challenge of designing our own set top boxes: “The need for improved coordination across all the various business functions increases dramatically in the as-as-service world”. This last quote is a pivotal sentence to understand the rise of the networked model of organizations. A Tesla manager says that “Elon Must actively encourages engineer-to-engineer communication and strongly discourages hierarchical communication”. Rich Lerz, the CEO of Nytec, has strong words about the need for colocation for designing the next generation of connected smart products: “The breadth of skills needed for these new future-generation products is very broad, and we found that we needed to house all these skills in the same physical location. All members of the team work along every step of the product lifecycle together, from ideation to prototyping to the manufacturing ramp using the same agile integration development methodology”.

3. Conclusion


I will conclude this blog post with the key take-aways that I take from reading – twice – this book. The following list is even more personal and biased with my own experience than the rest of this post.

  1. Product development in the digital world is a continuous and iterative approach, based on short customer-centric learning loops.
  2. A product digital twin is a knowledge engineering collaborative platform that unites multiples teams through out the lifecycle of the product with a shared “single point of truth”.
  3. The massive use of real-time data from multiple sources, together with a new generation of AI algorithms based on machine learning, makes it possible to dynamically optimize operations even when forecasting is not possible.
  4. Experience design plays a critical role in the success of designing and developing successful connected smart products.
  5. The ability to collect data in a transverse manner from all functions in a company, throughout the product lifecycle, stems from the existence of a unique, shared data model. Data collection is a continuous cycle, since there is more value in future data than data from the past.
  6. Software engineering and information systems make the backbone of smart connected products. Their role is to allow the enterprise to leverage the continuous flow of digital technology innovation.
  7. Success in the world of digital products and experiences is a race to continuous learning. Designing the right organization, securing the access to powerful computing resources and extensive data flows are key factors to improve the speed of learning.





No comments:

 
Technorati Profile