While companies recognise the need to innovate on a regular basis, designing and managing an organization that does so remains a challenge. Innovation often appears to conflict with the predictable, process-driven engine that most companies believe to be ideal for efficient operations. A little over six years ago, in 8 Steps to Innovation: Going from Jugaad to Excellence, Vinay Dabholkar and I identified the best practices of companies that had cracked this problem.
Since then, the winds of digital transformation have hit businesses, often with gale force. Business leaders can speak of nothing but Industry 4.0, artificial intelligence (AI) and machine learning (ML), internet of things (IoT), automation, and the like. Corporate executives often ask me how managing innovation is different in this digital era. This column is devoted to answering this question.
Before we look at the challenges in managing innovation, let’s explore how the nature and content of innovation have changed.
Nature and Content of Innovation
Clearly, the advent of digital technologies has broadened the scope of innovation across all four types of innovation—product, process, customer experience and business model.
Digital technologies first transformed innovation in those businesses where the product itself could be digitised—books, music, movies, etc. But the use of sensors and IoT has transformed a wide range of products from industrial machinery to home appliances into quasi-digital products. Service businesses have been transformed as well—for example, earlier, motor insurance products were generic, based on broad demographics like age or marital status. Today the on-board sensor on a car can observe driver behaviour and help customise a product to a specific customer.
Add AI and ML to this mix, and you now have products with embedded intelligence and self-learning and correcting abilities. This shift obviously opens up a whole new arena of innovation. The main impact of this digital paradigm is products that are more aligned to user needs and are better customised. The embedded intelligence means that product performance can improve over time with more use.
Perhaps the greatest contribution of digital technologies is to process innovation. Robotic Process Automation coupled with AI and ML is transforming manufacturing and service delivery processes. Earlier trade-offs between scale and cost are getting upended. Speed, efficiency and integration are the outcomes.
AI is the common thread transforming both product and process innovation. As Iain Cockburn, Rebecca Henderson and Scott Stern point out in a recent article, AI has emerged as both a general-purpose technology (like the transistor or microprocessor in earlier eras) and a method of invention (like digital computing). This has sparked a whole new wave of innovation, leading to potentially lower R&D costs and new classes of problems that can be solved. An Indian start-up, Niramai, is representative of this trend. The use of AI and ML has enabled this company to use non-invasive and safe thermal imaging for breast cancer screening and detection, replacing potentially harmful or invasive approaches.
It’s often said that we live today in an experience economy. Product and customer experience innovation converge in convenient options like Ola cabs or OYO-certified hotel rooms. Digital technologies have clearly opened up vistas to provide new and diverse customer experiences whether it be through bots or virtual reality. Customer experience innovation is enhanced thanks to the possibilities of real-time optimisation, remote monitoring, greater control over conventional products, greater customisation, and co-creation facilitated by digital technologies.
Business model innovation opportunities are significant in the digital era due to the opportunities to create new platforms, cross-subsidise products and customer groups, develop alternate revenue and pricing models, transform traditional businesses, and facilitate “pay-by-use” models. However, Uber’s $5 billion loss in the latest quarter and the fact that Amazon’s profits arise from Amazon Web Services and not from its much-vaunted ecommerce business point to the challenges in developing sustainable business models in the digital era.
Besides this broadening of scope for product, process, customer experience and business model innovation, digitalisation is driving a need for organisational innovation. Much of this is driven by automation. The current buzz is around moving from automation to augmentation of automated processes by human ingenuity and creativity, and of harnessing collaborative intelligence. Companies are seeking new ways of managing the human-machine interface.
What does all of this mean for the innovation processes followed within organisations?
In 8 Steps to Innovation, we identified three main challenges that companies need to address to make innovation a more routine part of the organization:
- building and sustaining a pipeline of ideas (the pipeline problem)
- providing velocity to these ideas (the velocity problem)
- improving the conversion ratio of ideas into successful business outcomes (the batting average problem).
The pipeline problem could be solved by top management laying the foundation for innovation; building a challenge book; and enhancing participation in innovation.
The velocity problem could be addressed by experimenting at low cost and high speed; using champions to move from proof of concept to incubation; and iteration on business models.
And, the batting average problem could be answered through sandboxes, platform approaches and open innovation, and by building in margins of safety.
The video below talks about these in detail.
How are these different in the digital era?
These broad priorities remain the same, but there are some major differences in how these are achieved. The big changes are related to the challenge book, experimentation and how you enhance the batting average.
Understanding of user needs is critical to building and elaborating the challenge book. Notwithstanding Steve Jobs’ dismissal of market research as inadequate to understand market needs, market research through surveys and focus groups followed by test marketing to validate concepts has, for long, been the mainstay of the approach of consumer product companies. This approach is being replaced in situations where customer data is available or can be generated.
Earlier, whether it was Gillette creating a new shaving system for lower income markets, or Renault developing a Sports Utility Vehicle for the Indian market, immersion was the key to see how people use products and services and understand their needs. Today, much of this is possible through data. If you don’t have data or access to it, you need to find new ways of generating it. Bajaj Finserve combines data from millions of customers who have signed up for its EMI-based credit cards with data sourced from third parties like credit bureaus to understand consumer spending and repayment behaviour and thereby build one of the most successful consumer finance businesses in India. In an era of data, immersion in the traditional sense may not be that important.
Of course, market research itself is more sophisticated in the digital era thanks to better gadgets and online tools. But you get more accurate data from customers in actual use situations than through customer surveys.
The velocity problem referred to the inertia faced in trying out ideas. While the broad solution is creating an environment and infrastructure to facilitate experimentation, experimentation has become much easier in the digital domain. The time needed to do an experiment has come down and many more experiments are possible with the same resources. For example, special pricing offers can be made on a few servers with the mass of customers being served through other servers as the control group. In The Digital Transformation Playbook, David Rogers quotes Kaaren Hanson, CIO of Intuit: “We can run 50 different ideas through our rapid experimentation process in the time and resources it takes to run three ideas through our old process.”
As Rogers explains, two types of experimentation are distinct but equally important—open-ended or divergent experimentation to develop concepts and ideas, and convergent experimentation using strict statistical protocols to make A vs B decisions. Low cost digital tools are available to perform both types of experiments. With the increasing ease of experimentation in the digital paradigm, ideation and experimentation are less distinct phases. A classic example of this is Greg Linden at Amazon who felt that just as in the case of the candies strategically placed at the physical retail checkout counter, there is scope to offer additional products to the ecommerce customer just before she pays for the items in the shopping cart. This flew in the face of conventional ecommerce wisdom, but Linden’s “unapproved” experiment showed that his hunch was true.
Enhancing returns from innovation (i.e. improve the batting average) continues to be a challenge in the digital era. Ever since companies like Procter & Gamble realised the potential of going beyond organisational boundaries to scout for innovation ideas about 15 years ago, open innovation has become a mantra to enhance the quality of innovation and the alignment of innovation with the needs of the market. The increasing robustness of digital tools has made practising open innovation easier.
Yet, as Sunil Gupta underlines in Driving Digital Strategy, open innovation has become even more important. The average quality of open innovation ideas through contests may not always be great but a few ideas could be game changers. Gupta points out that companies need to get some key processes related to open innovation right—clear definition of the strategic intent and goals; careful problem definition; attractive contest design; and incentive alignment to attract the best external collaborators. Specialised skills related to open innovation include scouting and the management of IP across organisations.
But the contours of open innovation have widened well beyond what Gupta describes. With startups born as digital natives clearly having some distinctive skillsets compared to established firms, collaborating with startups has become an important dimension of corporate open innovation. The paradigm practised earlier by serial acquirers such as Cisco and General Electric of scanning, minority equity investments, and finally acquisition is now replaced by in-house accelerators working in concert with strategic business units (SBUs) and empowered venture arms. Companies like Cisco, NetApp and Wells Fargo run such active incubator programmes in India. Even public sector banks like the State Bank of India have active engagement with fintech startups to collaborate on new forms of banking. SBI’s YONO mobile platform is the outcome of such a collaboration.
In The Digital Matrix, Venkat Venkatraman points out that creating new platforms with partners in other industries and sometimes even competitors in the same industry has become key to driving network economies and platform effects. These platforms then become important loci for innovating products and services for customers as well as building new capabilities. The creation of alliances for platform development has become critical enough to be a part of the strategic decision-making process of top management. Venkatraman gives the example of Ford that has undertaken a number of strategic experiments on car-sharing, car swap, dynamic social shuttle buses and parking space location through such collaborative platforms to explore the future of transportation and the nexus of traditional and digital technologies. Such collaborative platforms are becoming important sandboxes of digital innovation.
But all the innovation in the world is of little use unless you can capture the value created out of innovation. So, let’s look next at how the digital era changes the challenges of value capture.
Capturing Value from Innovation
Capturing value from innovation is becoming more challenging in the digital era.
As I discussed in my earlier article on strategy in the digital era, digital has lowered barriers to entry in many industries. It has also impacted the salience and value of complementary assets. For years, Gillette (owned by P&G) had a strong grip on the shaving products industry. Its brand was synonymous with the category. It has deep product expertise and excellent distribution. It could build a wall of patents around new products. Potential competitors such as Unilever kept away. But digital + data + potential for customisation changed all this. The Dollar Shave Club (DSC) created a new franchise around complete customised shaving solutions that arrive at your doorstep. Unilever acquired DSC for $1 billion, and presto, Gillette has a formidable competitor! Clearly, traditional complementary assets no longer constitute the barrier to entry that they were in the pre-digital era.
Data has become critical to understand user needs and to drive AI engines. Netflix’s success in creating popular content is directly linked to its data-based understanding of viewer preferences. Using data to produce films better aligned to the market has even reached Bollywood. Mumbai-based digital startup Pocket Aces is using AI and ML to test genres and plots, and uses the feedback to produce serials and shows for Netflix.
Increasingly, issues surrounding data ownership and use will have serious implications for getting value out of innovation in the digital era. Who owns the data could have implications for who owns the IP, and give a new colour to IP disputes. New regulatory requirements such as those contained in the EU’s GDPR require companies to disclose what’s in the AI “black box”, thus potentially diluting a company’s ability to get proprietary value out of AI. India’s draft ecommerce policy also points to putting buyer trends in the public domain. Further, data localisation requirements could make the whole data piece far more complex.
In platform-driven businesses, the platform captures a significant part of the value created. But, clearly, not everyone can be a platform player. In any case, a winner-takes-all logic seems to be working in many platform businesses. Rogers and others have argued that unique value creators in the value chain do have a chance to capture value. But with the pace of change and the aggressive strategy of platforms to enter new businesses, capturing value looks more problematic. This problem is aggravated by the slow pace of grant of intellectual property protection in most jurisdictions.
Some People Dimensions of Managing Innovation
In most corporations, the focus of innovation today is on data, analytics and AI. Companies have re-aligned their budgets to focus on these and other dimensions of digital innovation. This has resulted in organisational changes—R&D in traditional major disciplines has receded to the background, data-crunching and machine learning have come to the foreground. Really well-qualified and capable innovators in the emerging digital disciplines are in short supply and the best are snapped up by the large platform players with their deep pockets. Existing R&D personnel who are willing to embrace the new methods are retained by companies. These organisations are side-lining or removing those who can’t or won’t make the transition. Thirty years ago, CK Prahalad cautioned American companies against focusing too narrowly and losing valuable competencies; it remains to be seen whether the current approach of companies will have similar results.
The willingness to experiment is becoming an essential requirement of digital jobs. I recently met a young manager in an online travel services company who told me that doing 50 experiments a quarter is part of his KRAs. Surprisingly (and counter to the spirit and philosophy of experimentation), the company expects a 50% success rate!
In a data-driven world, the leader’s role is no longer to back hunches or practise intuition. Increasingly the leader is being seen as the chief experimenter—piloting strategic experiments of the type undertaken by Ford and demanding more experimentation by members of his team, of both the convergent and divergent type as appropriate.
How is managing innovation different in the digital era?
As we would expect from a paradigm shift, the digital revolution has increased the scope for product, process, customer experience and business model innovation. While the broad challenges faced by organisations in innovating in the digital era are not different from before, some specific changes are worth noting.
- Data, particularly if digitally generated, plays a critical role and is a potential substitute for customer immersion. Digital not only makes experimentation easier, but practically makes it imperative for the organisation.
- Strategic experimentation, particularly through collaborative platforms, needs to be on the top management’s agenda. Such collaborative platforms have the potential to become important innovation sandboxes for the company.
- Open innovation is perhaps even more important than before, but needs to be strategically designed and include a startup focus.
- Capturing value from innovation is becoming more challenging. This implies a need to move faster than before. Shifting innovation resources to emerging areas like AI and ML seems inevitable but also carries risks. The role of the leader is increasingly becoming that of Chief Experimenter.