GettyGetty
Digital transformation (DT) projects often fail – we know that – but there are steps companies and their partners can take to maximize the likelihood of success. In a recent conversation with Rohit Kapoor, the Vice Chairman and CEO of EXL Service, it became clear to me that digital transformation – in spite of all the myths and horror stories – can transform business processes and whole business models if certain things are true. The discussion was more of a meeting of the minds than anything else, and was therefore fun. It was stimulated by a Harvard Business Review survey (conducted by EXL Service) published as part of HBR’s reporting on the state of the practice of “Digital Transformation: Bridging the Gap Between Expectations and Outcomes.” His and my experience (including data we’ve collected at Villanova University on digital transformation) – far from pie in the sky – describe a realistic approach to digital transformation. We now know that when certain things are not true, digital transformation projects will likely fail, and some will fail spectacularly. But we also know that when certain things are true, well, digital transformation can succeed.
So here we go again. I’ve written about digital transformation several times here and elsewhere. Even though I sometimes promise to stop reporting the obvious, I seem to always find something new to discuss (or complain about). Actually, in this moment, I’m encouraged. We’re making real progress, even toward the development of at least a framework for a digital transformation maturity model comprised of an identifiable set of best practices. My conversation with Rohit Kapoor contributed to some optimism about how to improve the chances of successful digital transformation projects.
So what are these best practices, the list of things that should be true for digital transformation to succeed? What have we learned? Can DT actually work? Absolutely – if we exploit the lessons learned.
What (We Now Know) Must Be True
The conversation crystalized a lot of about digital transformation. Let me summarize in five steps.
1. Right-Sized Transformation Opportunities & Risks
Kapoor and (others I’ve spoken to and worked with) emphasize the need to assess the environment in which transformation will occur. Some verticals, for example, like financial services and healthcare, are more likely to transform their processes than others, like insurance. Vested interests – like saving money by reducing fraud – are powerful drivers. Transformation leaders must be open to change and have the willingness to “sell” transformation for its business value: professional introverts need not apply. Nor should leaders be driven by panic. Support should also descend into business units (where resistance can be significant) by describing the business value of transformation at all levels, which should include the right incentives for the key players.
Transformers must also trust their vendor partnerships especially since it’s unlikely that client companies will have all the talent, technologies and even domain knowledge to cost-effectively transform. Roles may also be threatened, so transformation should be unambiguously assigned to the right leaders, including, for example, Chief Digital Officers (which frequently layer conventional Chief Information Officers who end up running the infrastructure that supports the application of emerging technology for digital transformation).
Kapoor recommends creating an Office of Digital Transformation comprised of a Digital Solutions Center, a Digital Product/Partner Program and a Digital Innovation Lab. Yes (yes, yes!) we need formal teams devoted to digital transformation. Put another way, the recommendation is not to pursue digital transformation out of one’s back pocket, but to officially legitimize its status. Makes sense.
2. Agile Digital Transformation
As Kapoor stresses, it’s easy to be seduced by big bang approaches to digital transformation, but all that does is inflate unrealistic expectations. The preferred approach? “Small steps – not mega transformation – a three-year road map,” makes the most sense. “Define the issues narrowly, learning along the way, generating feedback” throughout the process. In other words, adopt an Agile approach to digital transformation. Manage expectations down, not up. Focus on achievable results over realistic periods of time, not spectacular promises that cannot be kept.
3. Get the Right Data
The data better be there. Not just piles of structured and unstructured corporate data about customers, production, manufacturing, distribution, competitors and processes, but high quality data that’s clean, consistent and analytically accessible. Transformation projects that involve big data analytics and artificial intelligence are especially dependent upon quality data. Kapoor reports that sometimes clients must be told that their current data repositories will not enable digital transformation. Understandably, these are tough conversations, but sometimes necessary before spending a ton of money on projects destined to fail: honesty and transparency are still keys to project success. Agree again.
4. Identify the Core Skillsets
What do you need to transform? In addition to the right data and (Agile) methodology, digital transformation projects require additional capabilities that companies need to exhibit internally and externally. The assumption is that companies – no matter how enthused they might be about digital transformation – cannot master all of the skills necessary to accomplish their goals. This is a significant acknowledgment and suggests that definitions of DT “core competency” cannot be satisfied internally by most – if not all – companies. Vendors are necessary partners since they’re far more likely to be tracking emerging technologies than their clients deeply emerged in the domains of their business. For the same reasons why most companies are now in the cloud (and why most of us no longer fix our cars), companies need partners and should not attempt to build out their own wide, deep and expensive digital transformation teams. Instead, they should share capabilities with their partners. Perhaps most important is what Kapoor describes as “orchestration,” or the ability to coordinate methods, tools, technologies and outcomes before, during and after digital transformation projects. This is a top-down and bottom-up management process done collaboratively with clients. The cloud computing metaphor may work again here, where, as Kapoor suggests, clients should adopt a hybrid (skills) approach to digital transformation.
5. Define the Right Success Metrics – & Pay When You Save
Kapoor described his approach to digital transformation as win/win, but in an unconventional way. If clients fail to see the impact of their digital transformation projects, vendor compensation adjusts downward. Yes, this is unconventional, especially in a world where clients pay vendors for the coffee they serve. For example, clients focused on robotics must see empirical gains in automation. The same is true for fraud detection and other projects where compensation should be linked to measurable impact. This obviously reduces the risks around digital transformation.
Maturity? Maybe
So what have we learned over the past decade? The obvious: digital transformation is not a panacea. Best practices? DT requires good data, the right core skills and top-down and bottom-up support. Orchestration is necessary. Outcome metrics must be empirical. Start small and build iteratively over time: be Agile.
It takes time for reality to take hold, to manage inflated expectations about most everything. The Gartner Group actually has a name for the phenomenon: “the peck of inflated expectations” (followed by “the trough of disillusionment”). But the trough is followed by the “slope of enlightenment”on its way to the “plateau of productivity.” Digital transformation suffered the same history as so many other technologies and processes. At the end of the day, DT best practices must be anchored in business value. It must also be recognized as a multi-disciplinary process that – like any complex project – requires lots of skills, competencies, collaboration and coordination. Now that we know, we can do it right. We’re well on our way to a DT maturity model. (Note that I have no financial interest or relationship with EXL Service.)