The Innovation Edge.

Driven by the pandemic and the need to transform in a post COVID-19 era, companies continue to digitize every aspect of their business to remain relevant to customers — innovation is now a deeply entangled part of every business. A recent study by McKinsey & Company that analyzed US consumer sentiment and behaviors during the coronavirus crisis notes that "e-commerce sales continue to experience outsized growth, with online penetration remaining approximately 35 percent above pre-COVID-19 levels, and showing more than 40 percent growth over the past 12 months." Equally companies are finding that to remain competitive they will need to give customers compelling reasons to return to their stores — not surprisingly retailers are innovating experiences by blending social media as part of the in-store journey, accelerating click-and-collect capabilities (buy online, pick up in store) and exploring augmented reality to make the shopping experience fun and memorable. But finding the right balance between hyper-personalization and sustainable growth is not as simple as it sounds.

In addition to transforming jobs and skills, it is also overhauling industries such CPG, retail, banking and capital markets, health services and others — every sector is accelerating investments in innovation to create meaningful interactions. In the process, many are discovering that delivering memorable engagements require a closed-loop process in which every area of the business is hyper-concerned about delivering an awesome customer experience while continuously measuring the impact of innovation to the bottom-line.

While disruptive innovation can help create new business platforms and drive sustainable revenue streams, it's not a guarantee. In short, the pursuit of innovation for the sake of it being cool or promising, invariably leads to increased operating risk. In fact, as attractive as innovation may seem, it may not always yield the expected benefits in the context of a company's operating culture, structure or shareholder profitability expectations. This article explores some of those risks by evaluating what happens at the innovation edge — where innovation reaches a point of diminishing returns or places the business at risk.

IMPACT ON THE BOTTOM-LINE .

Without a doubt innovation is a critical ingredient to success in a world that continuous to digitize itself. But it's one element of success. Innovation can be risky, and getting the most out of your investments is often about having a clear line of sight on those capabilities that are noise versus those that represent strong signals for adoption. Too, as businesses accelerate the pace of innovation, profitability may stagnate, even erode in the short-term causing the potential pullback of investments resulting in lower than anticipated shareholder value.

Take the case of a fashion retailer who to compete and deliver a memorable experience, enabled customers to customize every aspect of their standard handbag. Handle, tabs, feet, body, straps, piping, were all customizable. The company launched a brilliant website that allowed customers to visualize their product and dynamically customize their handbag. Advanced analytics helped predict what styles would likely be more attractive for the buyer while using complex basket analysis algorithms to suggest additional products at checkout. Not surprisingly, the company over time experienced diminished profitability — the reason: too much complexity which then spread through all parts of the company's operations. Imagine the number of parts and safety stock levels that needed to be maintained (from various suppliers across the globe) to meet unpredictable demand variations for specific SKUs.

"Hyper-personalization that results in significant product proliferation can lead to unsustainable product and service complexities that negatively impact profitability."

To address this specific innovation challenge, companies start with a zero-complexity baseline model that highlights the production costs of selling an absolute minimum number of standard parts. Variety is then added into the product mix and is subsequently modeled out to understand the resulting impact on all aspects of the business (i.e., sales, marketing, supply-chain, finance) — the goal: "find the edge of innovation that intersects with hyper-personalization and commercial viability."

But the challenges of innovation go much deeper and affect many functions and the underlying culture of the business including its moral integrity. In this article we explore a few of these.

WHAT HAPPENS AT THE INNOVATION EDGE?

While there are many areas where companies pursue innovation, in this article we focus on five: customer engagement, machine learning analytics, the Cloud, Hybrid ERP and DevSecOps.

Each area represents an opportunity to deliver seismic changes in the way a company connects with its customers, improves operating efficiency and delivers fast-paced value to shareholders. However, making the right choices at the right time can be tricky. We call this managing at the innovation edge.

"While the benefits of innovation are significant, it has the potential for yielding diminishing returns and in some cases approach the point where risks far outweigh benefits if not properly managed."

The more one approximates the edge, the greater the risk of diminishing returns due to implementation risks, costs, complexities, required skill sets and other factors.

CUSTOMER ENGAGEMENT.

Leveraging complex data unification and machine learning analytics, companies across sectors are becoming better at delivering a hyper-personalized customer experience. A study by McKinsey & Company notes that a large percentage of customers are likely to be more loyal to brands that deliver personalized services based on individual preferences. By using complex analytics that help precisely identify customer buying trends, needs and wants, companies are able to fine tune services and product recommendations.

“We see our customers as invited guests to a party, and we are the hosts. It’s our job every day to make every important aspect of the customer experience a little bit better” - Jeff Bezos, CEO, Amazon

To enable an awesome experience, businesses continuously strive to reduce friction during the buying process — the goal is to create an effective engagement journey. Businesses are therefore streamlining checkout, simplifying customer onboarding, untangling complex order-to-cash (OTC) processes, hyper-personalizing interaction and product and service choices to create memorable customer experiences. However, the more one approximates the innovation edge, the greater the risk of adversely impacting customer sentiment and potentially the company's moral integrity.

Many businesses are revamping their e-commerce platforms, driving mobile engagement, pursuing consistent interaction across channels (omni-channel), experimenting with conversational AI and augmented reality and other innovations. Many of these initiatives deliver significant benefits to the bottom line by creating new business platforms, and revenue streams. However, if not managed correctly, innovation can also result in significant risks.

First there is the point at which customers feel that the company's communication strategy has become annoying and intrusive — there is too much knowledge about their personal behavior, wants and needs. Using data and machine learning algorityms to drive dynamic messaging and targeting can also result in incorrect social media signals and destructful engagement which can impact the company's moral integrity.

Secondly, customers may feel that the company's efforts to drive operating efficiency by automating various human functions, results in an impersonal interaction that lacks human touch. Consider the annoyance of dealing with ineffective chatbots in the customer service, order management and tracking functions.

Third, as previously noted, hyper-personalization that drives customized communication and enables high product variability can also significantly impact profitability. Too few choices, not good — too many choices also not good.

Finally, as the company, and its peers, accelerate the adoption of innovation, they continuously move the features baseline — customers expect certain functions which in turn requires the company to adopt a continuous development and delivery cycle, dynamic scaling and managing site reliability to deliver a memorable experience. It's about not disappointing customers.

In all cases while analytics, machine learning and other technologies can drive significant transformation of the customer engagement, even if the goal is to drive improved engagement, too much of a good thing can have significant negative effects.

MACHINE LEARNING ANALYTICS.

For clarification, we are focusing on the topic of machine learning analytics and not traditional business intelligence (BI) analytics which is also valuable in identifying trends, and outliers. Machine learning analytics (MLA) on the other hand automates the entire data ingestion and analysis workflow to provide comprehensive and actionable insights while testing out hypotheses.

"Machine learning analytics is often used to determine commonalities between different data."

Machine learning analytics is typically used to determine commonalities between different data to understand for example what makes customers alike, what is the relationship between customers, products, seasonality, weather patterns, market trends, etc. It is also used to conduct deep outcome-based analysis to determine product and service strategies.

A recent study shows that "a company needs three things to offer such predictive analytics: 1. A Cloud-based, customer-level data lake, which is the foundation for better customer understanding, 2. algorithms based on data showing what influences customer behaviors to generate predictive scores for each customer based on journey features, and 3. an “action and insight engine” that provides information, insights and suggestions via an application-programming-interface (API) layer."

Machine learning analytics and resulting recommendations are used in several platforms including but not limited to Amazon, Netflix, and Facebook for sentiment analysis and product recommendations. In the retail sector MLA is used to make dynamic product suggestions by analyzing vast amounts of behavioral data in real-time. In the financial services sector MLA is used for deep fraud detection and to make recommendations on new products.

Machine learning analytics has its drawbacks too as it can be susceptible to data errors, and deep complexity in programmed models— complex algorithms and data sets can carry inherent bias which can skew results in the algorithm’s output — or worse drive incorrect actions in downstream systems (i.e., MRP).

The technology also has a deep impact in sectors that rely heavily on complex revenue management practices to drive profitability. Initially designed for the airline and hospitality industry, revenue management uses analytics and performance data to manage churn, improve forecasting, predict customer behavior, improve loyalty, and enable dynamic pricing.

Over time MLA has been adopted across various industry sectors that involve customers who are willing to pay different prices for the same product, at different times, and when there is a set inventory to be sold, that must be sold within a specific time (i.e., airline seats, hotel rooms). Aggregating, unifying market data, buying trends, customer desires, seasonality and other factors is crucial to the hospitality industry to maximize revenue, and profitability. However, aggregating and modeling the wrong data, can be devastating resulting in inaccurate trend assumptions, incorrect pricing, higher churn, inaccurate booking forecasting, and lower margins.

To support effective machine learning analytics, data scientists unify, prepare, transform, encode, and bring data to a state that is understandable by machine learning algorithms. The process is complex and requires a team with diverse skills including data scientists, data engineers, statisticians, modelers, developers and others. While many Cloud providers provide platforms to accelerate adoption (i.e., Amazon SageMaker, Azure ML), having the right skills to ensure successful execution is crucial.

Nonetheless even with a time-consuming data ingestion process, complex error diagnosis and correction, and investments in skilled resources to fine tune models, the resulting insights are powerful and often cannot be derived by throwing more person-hours to a problem.

THE CLOUD.

Gartner predicts that Worldwide end-user spending on public Cloud services is forecast to grow 18.4% in 2021 to total $304.9 billion, up from $257.5 billion in 2020. “The ability to use on-demand, scalable Cloud models to achieve cost efficiency and business continuity is providing the impetus for organizations to rapidly accelerate their digital business transformation plans. The increased use of public Cloud services has reinforced Cloud adoption to be the the new normal. Furthermore, Gartner notes that "The COVID-19 pandemic forced organizations to quickly focus on three priorities: preserve cash and optimize IT costs, support and secure a remote workforce, and ensure resiliency."

But as Cloud adoption accelerates, so are the disillusionments. The benefits are significant and include areas such as lower infrastructure costs (platforms and networking), lower investments in raw computing infrastructure, lower total cost of ownership (TCO), ability to deploy solutions faster, improved reliability and endless scale. But the Cloud migration journey remains a complex endeavor as companies grapple with interconnecting old and new systems, mobile platforms, addressing security and compliance risks and lack of skilled resources to containerize and make legacy applications Cloud-ready. While 96% of respondents in a recent Accenture analysis report achieving some degree of their expected Cloud outcomes, less than half are?“very satisfied” with the results.

While Cloud innovation and services continue to be critical many agree that there is also a huge, missed opportunity, especially when the strategy for migrating to the Cloud is primarily based on a lift and shift strategy for older/legacy applications.

"While Cloud innovation and services continue to be critical to accelerating a company's digital journey, many agree that there is a huge, missed opportunity."

But if you approach Cloud computing with a clear business-case and a long-term horizon view of the complexities, costs and benefits, drivers for choosing a particular Cloud model (public, private, hybrid) and migration strategy, the opportunities for change become clearer, and limit the potential of disillusionment.

The Cloud provides a huge opportunity to simplify the eco-system and modernize traditional monolithic software systems by aggressively driving micro services enablement, leveraging containerization, adopting an "everything-as-a-service" philosophy, and using modern integration platform as a service solution (IPaaS) to unify on-premise, Cloud and SaaS applications. Taking these factors into account drives improved adoption and expectation management for all stakeholders (business and IT teams).

HYBRID ERP.

For decades companies ran their operations on monolithic ERP systems that manage workflows such as orders and payments, materials requirements planning (MRP), inventory and warehouse management, govern supply-chains and control the company's financials to name a few. While ERP platforms are the foundation for the world's largest businesses, they are often critiqued for being difficult to adopt, having a closed data model, too rigid to customize or, expensive to run and operate — just licensing costs can be limiting.

Over the past years, most ERP vendors moved their application suites to the Cloud — SAP S/4 HANA, Oracle ERP Cloud, Salesforce, Microsoft, INFOR, IFS and others are delivered through software-as-a-service (SaaS) platforms. And in fact, most have established clear Cloud migration deadlines. Too, the pandemic accelerated Cloud adoption as businesses migrated their on-premises businesses to the Cloud which in turn drove a pent-up the demand for SaaS solutions. A recent Gartner report notes that SaaS still dominates one of the biggest market segments amongst Cloud services and is expected to grow at least $138.2 billion in 2022. In terms of functionalities, versatility, and accessibility, SaaS remains a viable option for enterprises in a vulnerable business environment.

However, in contrast to prior years' strategies to implement a single integrated ERP to manage all aspects of a business, many companies are pursuing a best-of-breed approach. Rising costs, specific industry functionality requirements and the need to deliver features in an agile manner, are forcing companies to consider a new approach that incorporates an ERP system to manage core transactions, while interconnecting this with third party solutions to perform specialized tasks such as: integrated business planning (IBP), warehouse management, transportation management, supply chain planning, e-commerce, machine learning platforms, advanced analytics and others.

"Companies increasingly consider a best-of-breed ERP approach that incorporates a main ERP system to manage core transactions, while interconnecting this with other third-party solutions."

However, several factors need to be considered in pursuing a best-of-breed strategy. Bringing multiple systems together is?complex — it requires well thought out integration strategies, a master data management (MDM) approach to synchronize data cross the enterprise eco-system, a well laid out digital architecture that integrates all systems in a frictionless manner and management of multiple vendors.

However, a hybrid ERP approach can augment on-premises legacy systems with Cloud-based SaaS solutions that would otherwise be complex and expensive to implement on top of existing legacy systems.

DevSecOps.

According to a Verified Market Research report , the Global DevSecOps Market was valued at $2.18 Billion in 2019 and is projected to reach?$17.16 Billion by 2027, growing at a?CAGR of 30.76% from 2020 to 2027. The report highlights that "the growing need for secure applications owing to the increasing number of cyber threats is the primary factor driving the growth of the market. Also, the rising demand for application delivery and increasing compliance on security is another factor that contributes to market growth."

"In fact, it is anticipated that cybercrime will cost the world more than $6 trillion this year alone."

In its simplest form development, security, and operations (DevSecOps) automates the integration of security at every phase of the software development?process and helps ensure the delivery of secure software. Rather than security being an afterthought, it's now an integral part of short-sprint software development lifecycles that need to rapidly deliver much needed features and functionality for companies to remain competitive. Too, while security in the past was a task relegated to a separate security team, DevSecOps makes application, services, the Cloud and infrastructure security a shared responsibility — development teams, cybersecurity architects and engineers, and IT?operations teams all share responsibility. Most modern development teams have adopted a shift left approach, a popular term in the DevSecOps community meaning cybersecurity is built into the application process as early as possible in the development lifecycle System testing is integrated into an automated test suite where all teams detail out requisite test methods, traceability tests, audibility controls and visibility features that are then used in a continuous integration and delivery pipeline to deliver software in an agile manner.

Automation and machine learning are the foundation for DevSecOps platforms such as GitLab, Aqua Security, Acunetix, SonarQube, Codacy and others.

Adopting DevSecOps frameworks and tools will become essential to businesses that need to rapidly deploy solutions, and feature enhancements — the goal is to deliver software, safer, sooner, the DevSecOps motto, by streamlining the provisioning, patching, hardening, and configuration of secure software.

However, pushing DevSecOps innovation also brings challenges. A significant culture change in the way technology and business teams interact is a pre-requisite for adopting a smart delivery pipeline. Security as noted is no longer owned by a single team but is a shared responsibility. This shared responsibility can result in a lack of accountability across teams. But the greatest risk lies in a lack of collaboration between teams early in the process as it goes counter to traditional thinking — security has to remain a separate and disconnected unit to ensure objectivity and proper vulnerability risk management. Too, modern DevSecOps tools may not necessarily integrate with existing security and vulnerability testing tools, thus requiring companies to make additional investments.

Nonetheless, in a world where customers continuously demand new features and capabilities, businesses need to deliver new functionality at a fast pace but in a secure manner. Therefore, while adoption challenges can be significant, in complex environments a company may not have much of choice but to adopt DevSecOps as part of its development methods and frameworks.

MAIN TAKEAWAYS.

The digital revolution is well under way and will continue to transform industries while creating an insatiable demand for new jobs and skills. And while innovation brings added complexity, one thing is certain: there’s no turning back.

But business disruption and uncertainty can fuel anxiety about the future that can carry deep organizational consequences. History tells us, of course, that businesses and global commerce will eventually adapt to the digital revolution. But leveraging the power of innovation should be well thought out to harness these capabilities and improve the probability of success without diminishing organizational energy and enthusiasm.

Tightly inter-related innovation trends such as those touched upon in this article invariably lead to improved interactions with customers, and other stakeholders while driving large productivity gains. But achieving the promise of innovation requires a culture of innovation that focuses on employee engagement at all levels. Innovation is not just about implementing new technologies and capabilities — rather it is the relentless pursuit of finding new ways to do things better, transforming business processes, and adapting to changes in a way that results in greater brand loyalty, better service, new revenue models and improved profitability. But perhaps more importantly, it's also about having moral integrity to protect data privacy and not to use this data to create harmful products and services for the sake of driving up profitability.

Pursuing novel innovation as an end all solution to solving problems creates anxiety and often negatively disrupts regular business. On the other hand, fostering a culture of innovation that pushes for continuous improvement as a natural part of everyday work motivates the entire company. And while there are many tactics for fostering an innovation culture none is more powerful than communicating a well thought out and results focused digital strategy that cascades to all levels of a business — clearly defined goals and desired outcomes to be achieved through innovation lowers anxiety and creates much needed organizational buy-in. Doing so also provides powerful indicators on the extent to which innovation can be driven, at what pace and overall level of tolerance for change.

Ultimately, it's about creating an innovation culture that promotes creativity and continuous improvement, ensures moral integrity, accepts failure as a necessary element of progress and perhaps a little bit of good luck.

Anthony DeLima