DOWNLOADSArticle (8 pages)
In this episode of The McKinsey Podcast, Kate Smaje and Rodney Zemmel—global leaders of McKinsey Digital—talk with McKinsey Global Publishing’s Lucia Rahilly about digital transformation: what it really means, how to deliver on it, and why it should remain front and center on the CEO agenda. Plus, stay tuned for Kate’s and Rodney’s quick takes on trends to watch.
After, in an Author Talks excerpt, Amy Webb, futurist and author of The Genesis Machine: Our Quest to Rewrite Life in the Age of Synthetic Biology (Hachette Book Group, February 2022), describes how synthetic biology could open the door to “bespoke children,” significantly longer life spans, and subterranean societies.
The McKinsey Podcast is cohosted by Roberta Fusaro and Lucia Rahilly.
Digital transformations are a long game
Most Popular Insights
Lucia Rahilly: The phrase “digital transformation” has been part of our business lexicon for many years now. And at this juncture, most companies have, presumably, invested a relatively substantial volume of resources in digital and tech. Have leaders made any meaningful progress in reinventing themselves digitally? Or is successful digital transformation still elusive?
Rodney Zemmel: It’s become fashionable to say that many digital transformations fail, that it’s hard to get value out of them, and so on. I think that’s created an impression that digital transformation is elusive. The reality is, most big companies have undertaken a digital transformation, and most big companies get some value from that digital transformation.
The point of digital transformation isn’t to become digital. It’s actually to generate value for the business. And having a clear, integrated, top-down road map of where that value is is one of the biggest gaps between companies that get the full value and companies that get something that is just a shadow approximation of the full value.
Kate Smaje: Another important part of a successful digital transformation is how you define what “good” really looks like here. Because there isn’t a point in time when digital transformation is done. It’s much more about, “How am I building a real muscle for the organization to continue getting better and better as I go on?”
The point of digital transformation isn’t to become digital. It’s actually to generate value for the business.
Where leaders go wrong
Lucia Rahilly: The two of you are talking to leaders on this topic every day. In your work helping leaders reinvent themselves digitally, what are some concrete examples of the kinds of challenges you see?
Rodney Zemmel: A common failure mode, or an insufficient success mode, is a chief executive saying, “We’re going to go digital,” and then making public statements about digital strategy. Then every person on their leadership team creates their own digital road map. What you end up with, six months or a year later, are many digital pilot projects across the organization.
Instead, the chief executive needs to focus on getting the full leadership team to be able to talk from the same page and have the same set of priorities, and then focus on having a talent and capability road map that is as detailed as their technology road map—and getting that team to move as one against a clear set of technology priorities and people priorities.
Kate Smaje: The key here is being able to articulate how value is created in a company. And the challenge that lots of companies find is they don’t have a consistent and aligned way of either identifying or measuring that value, and therefore it becomes harder to go after it.
What you typically see 12 months, 18 months, or two years into many transformations is this notion of, “I don’t feel I’m getting as much as I should be.” The root cause of that is often a lack of alignment on where and how that value is going to be created.
How to get started
Lucia Rahilly: Let’s go a little deeper into the how—for example, how leaders assess where transformative value is possible within an organization. What do you suggest there?
Rodney Zemmel: There are three rules of thumb that seem to be evolving. First is that companies that get the most value from this actually spend a lot of effort thinking about, “What are the new digital businesses to launch? How can we create new value with new products and new customers versus transforming the existing business processes?” There’s sort of a duality—you should spend as much focus on new digital business building as you do on transforming the current business.
Rule of thumb number two is, you’ve got to focus on things that are big enough. And maybe that’s obvious, but it sometimes surprises us how many people will call something a digital transformation, and you add up the total economic impact, and it’s less than, say, 15 or 20 percent of the company’s overall EBITDA. If you’re not targeting at least 15 or 20 percent, in our mind it’s hard to call that a transformation and to sustain the level of organizational focus around it.
And then the third rule of thumb is, it’s best to start with a concentration in a particular area rather than sprinkle a little bit of digital or a handful of analytics use cases broadly across the organization. Pick one area of the business and really focus on building some momentum there first and then on growing from there on out. We think those three rules of thumb have emerged from the companies that have been more successful relative to others.
The metrics that matter
Lucia Rahilly: What role do metrics play in assessing the value of digital-transformation efforts?
Kate Smaje: Metrics are super important. But it’s important to define those metrics. What you want to get a sense of is, “Is my digital transformation really working here?” And, yes, some of this is financial benefits: there should be a set of outcomes across the business that you’re driving to, whether that’s operational or financial metrics.
But it’s also: Is this working in terms of really building this muscle that we were talking about? Are my capabilities being enhanced over time? Are there more people within my organization that understand how to use technology or data better in their day-to-day working? Is it starting to transform the culture? Can I start to see the metabolic rate of the organization speeding up? Are we making decisions faster as a result of this? There are lots of different indicators that you’re trying to build up to get a real sense of, “Is this working?”
If you just focus on the financial or the operational pieces—they are important, don’t get me wrong, but they’re not the be-all and end-all. And they often miss whether you’re really creating that long-term, sustainable muscle versus just executing against a bunch of very good initiatives for today.
Lucia Rahilly: Do successful CEOs share a best-practice governance model for digital transformations?
Rodney Zemmel: In terms of how the transformation is led day to day, we’ve seen a model based on a single transformation leader, and we’ve seen a coleader model—a technology executive and a business executive. Both of those models can work. But digital transformation does really need to be a standing item at the top of the company to make sure it stays on the CEO’s agenda and to show that it’s aligned across the full company agenda.
The talent opportunity
Lucia Rahilly: Let’s talk a bit now about talent. McKinsey has published a lot about ways that “the Great Attrition,” or what others call “the Great Resignation,” is altering dynamics in the talent marketplace, including exacerbating the talent gap. How is that affecting tech talent in particular?
Kate Smaje: This, in many ways, is a wonderful opportunity for tech talent, given how many companies out there are trying to upgrade. And I use the word purposefully. Because this isn’t all about hiring; it’s often about upgrading existing talent, as well.
Given there is so much demand for that talent, the fact that some people are taking stock and saying, “What’s the best place for me going forward?” or “Am I getting as much out of my career as I really want to?” is not necessarily a bad thing.
What I think folks miss on this, however, is that the hiring part is perhaps the easier part. What people miss is, “How do I make that talent that I’m going to bring in wildly successful once they are here?” And they miss the ways in which you need to think this through: What does this mean for how HR works? What does it mean for spans and layers? What does it mean for comp models or procurement thresholds? And so on. It’s those kinds of issues that, in many ways, are harder than the actual hiring. The folks that don’t fix for those issues ahead of time are, unfortunately, probably going to experience more of the effects of the Great Attrition.
Rodney Zemmel: What we’re finding is, the top talent for these technology topics is much more dispersed than it was a few years ago. We’ve seen companies attract the right talent in pretty much any geography you can think of.
If you’ve got the right mission, the right set of leaders who are going to put effort behind it, and if you are prepared to make the right technology and career path investments to make those people successful, you’re going to attract them in a much more location-independent manner. That’s piece one.
Piece two is the range of talent that you need. A while back, it was focused on just a handful of what we call “technical guilds.” It was all about getting the data scientists or the software engineers. Now, it’s a broader groove—it’s not just data scientists, it’s also data engineers and machine learning engineers. If you’re not in-sourcing agile coaches and leadership for a strong product management function and so on, you’re not going to build the capacity that you need. So a broader array of guilds.
And then the third piece, as Kate was mentioning, is this huge potential in reskilling your existing talent. The extent to which companies—even companies in some pretty traditional industries, from engineering backgrounds and so on—have been particularly good at this, in their ability to actually reskill and turn many of their existing people into really strong digital and technical talent, is quite remarkable.
Lucia Rahilly: What role do ecosystems beyond the organization play in widening or deepening that talent pool?
Kate Smaje: Partnerships are becoming a much more significant way of accelerating the talent upgrade that we’re talking about here. That no longer means outsourcing large swaths of folks, but it might mean partnering on specific capability areas or augmenting talent in the near term in order to get that level of acceleration.
For some folks, it’s also about being able to show the real step-up in the bar for quality for that talent. And can we see it, live it, and feel it by bringing in some of those folks, even if it’s for a short period of time, as you upskill and upgrade the talent base that you have?
In my mind, technology organizations are becoming more porous. But it is less about outsourcing. It’s not a simple line in the sand on this anymore. It’s much more about how to collaborate in an ecosystem of partners in order to really accelerate and upgrade that talent.
Rodney Zemmel: More in-sourcing and smaller teams of higher-skilled people, rather than large numbers of technology arms and legs, seems to be the dominant trend.
Technology organizations are becoming more porous. But it is less about outsourcing and more about how to collaborate in an ecosystem of partners to accelerate and upgrade that talent.
Getting better—faster—on diversity
Lucia Rahilly: We’ve seen a lot in the media about the lack of diversity in tech, both in the sector and as a functional capability. Do you see that changing or playing a role in attracting tech talent?
Kate Smaje: Yes. It is clearly a perennial issue. We did some research fairly recently around why gender diversity is particularly elusive among tech talent teams. And one of the things we found is, it isn’t so much about a glass ceiling, as we see in many other walks of life, but much more about a broken first or second rung. It’s getting that first promotion, that second promotion, that is affecting the pipeline quite significantly.
Rodney Zemmel: Part of the solution on the diversity challenge is how much a company is willing to bet on early-career talent. I’ll give you two facts. Number one is, if you look at how universities have done at getting to gender parity in computer science courses and data science courses and various technical disciplines, that varies by geography, obviously. But there has been a huge success, and many, many top schools have reached parity, or close to parity.
And number two is, if you look at the areas where companies are ramping up hiring the most, it’s, of course, in digital and technology areas. So if you combine those two thoughts and say, “Where are we prepared to bet on early-career talent?,” you should be able to use this as an accelerator for diversity in your company, not a decelerator.
Modernizing your infrastructure
Lucia Rahilly: How should CEOs think about tech and data as part of a successful digital transformation?
Rodney Zemmel: If you’re trying to launch just one or two or three different analytics use cases in your company, you don’t need to worry too much about having a data strategy or data architecture. Because one or two or three use cases is normally pretty manageable.
By the time you’re at the inflection point, something really transformational, you need to take the topic of data governance extremely seriously. It’s a federated model, where you have a small, probably central, organization that sets standards, sets rules, sets clear ownership, sets privacy, sets strategy for what needs to be internal, external, and so on.
And then you have clear business owners across the business for each data stream within the company, and a view on how you’re actually going to use data to get competitive advantage. So, like many things we’re talking about today, it becomes more of an organization problem than a technology problem. Thinking through that data governance is critical for anyone who has ambition to do something that’s truly transformational, rather than just launch a few use cases.
Kate Smaje: And there’s an agenda here around modernization. Often, people get bogged down in the technology road map or the data road map and so on, and miss that this is actually about modernizing the infrastructure of the company.
We saw many of these several years ago, where people just wanted to get excited about the shiny new thing: “Build me a new front-end app,” something that we can touch and feel and pat ourselves on the back for. Not understanding how much your technical debt is constraining what you are able to do is becoming a real blocker to execs being successful. So for me, before we go into the tech, the data, and so on, it’s about modernization: modernizing the way in which that infrastructure is going to help you do all the things the business wants to be able to do.
Rodney Zemmel: Where I’d be really careful is, there’s often a temptation for companies to say, “Let’s make the technology investment, and then we’ll figure out all the great things we’ll be able to do in the business.” If you’re not working backward from a business case, then you’re unlikely to get the value from it that you’re intending to get.
So you think you’re agile?
Lucia Rahilly: Particularly during the pandemic, we have talked a lot about speed and the need to maintain an accelerated pace in the digital world. Has the pandemic proved the case on agile ways of working?
Kate Smaje: Oh, a wholehearted yes from me on this one. And the reason I say so is because in many ways, it was an overnight experiment in agile. Not just in terms of the speed—many organizations had to work in cross-functional ways that they hadn’t had to do before, because they were asking fundamental questions that required small teams of great people to come together and really solve for them across the siloes of the company.
Many employees had to work at a speed, at a metabolic rate, that was very different from their regular day-to-day operations. And we needed to get things done fast—we needed to figure out the resilience of the supply chain, we needed to figure out how we were going to contact customers in new and different ways.
Rodney Zemmel: The reality is, although we’ve been talking about agile for a decade or more, very few companies are really doing it at a broad scale across the enterprise. We have a lot of companies that say they’re doing agile, and when you look, they might be doing agile on how they deploy their technology teams. But whether they really have technology and the business working together, in multiple pods across the business—not just in one or two places that has the CEO’s attention on it—that’s rare.
What’s even rarer is when you can get the control functions in there as well. And often, if you have technology and business together, but you don’t have legal, regulatory, compliance, finance, whatever the relevant control function is in your context, in the agile pod as well, then you’re not really getting the speed acceleration that you need to get.
Working together to solve problems
Lucia Rahilly: Let’s talk about adoption. We’ve all had the microexperience of rolling out new technology and encountering a sort of wide-ranging, passive resistance. Change is hard—folks may have an innate bias in favor of the status quo. And CEOs are functioning at a very high level within organizations. How do successful leaders ensure that these initiatives actually get traction?
Kate Smaje: First is, to get traction this has to be linked to business owners from the start. This can’t be technology pushing out. It also can’t be business throwing requirements over the wall to technology. The integration between business and technology needs to be really, really seamless.
The second is, you need to think about adoption from the get-go. This is not something that you do in the linear fashion of creating the product and then figuring out how it will be adopted. It needs to be pulled up front, in terms of designing that product, to make sure that you optimize and maximize adoption from the get-go.
The third is, almost as a rule of thumb: for every dollar that is spent on digital, on development, and so on, at least another dollar should be spent on adoption. And then going into this thinking about it as an investment in the success of it as much as an investment in the technology itself.
Rodney Zemmel: I want to come back on one word Kate used there: “requirements.” In an IT culture, the business sets the requirements. In a digital culture, you’re not in a requirements world. Instead, there’s a team that is addressing a business problem, or a customer problem. And that team of businesspeople, technologists, and the control function work together to solve that problem. And the word “requirements” is not in there. It’s solving a problem versus working against a sort of order list from the business.
To me, there’s one question that matters on adoption, and that is, “Who is responsible for adoption?” That answer should be, “It’s the business leader.” The business leader, the business owners, need to be responsible for the adoption of the initiatives that they are sponsoring to drive profit improvements or growth in their business.
Trends to watch
Lucia Rahilly: Last question. Any particularly hot or fast-moving digital trends you’re keeping an eye on in 2022?
Kate Smaje: Go on, Rodney, you go first. I know what you’re going to say.
Rodney Zemmel: I’ll give you two. Fast moving and here right now is MLOps. If you don’t know that word, you should. It’s “machine learning operations.” It’s the operating system—both the technology and the people and processes—that you need to make machine learning work at scale in your organization.
As companies or organizations go from doing one or two things in AI or machine learning to really having a powerful enterprise, they need an operating system. They need a way of doing it—an end-to-end operating system for how you’re going to make that work across your company. It’s a pretty complicated topic and it’s moving quickly, but your company, your organization, needs to have one if you are going to get value, at scale, from machine learning or AI.
My over-the-horizon trend is quantum computing, or quantum technology, more broadly. If it works—and it is still a science bet—it will change the entire landscape that we’re talking about. So for many companies, it’s worth it just to have an eye on it to understand where it’s up to and what the implications could be.
Kate Smaje: Let me offer two more that look slightly different. The first for me is, it’s not about any individual technology as a trend going forward. It’s about the combinatorial power of two or three or four—collide them together and say, “Now, what’s the art of the possible?” Because it’s when you bring them together that the real magic happens.
And a second one that I’m seeing enter the conversation more and more, for good and right reasons, is “digital trust”—how we as organizations and as a society are thinking about the interrelated issues of model bias, model explainability, privacy concerns, data residency, and so on.
Rodney Zemmel: It’s worth remembering what we said right at the beginning: That the point of a digital transformation isn’t to become digital and to be able to drop technology buzzwords in your investor presentation. It’s to generate value for the business.
It sounds really obvious, but just thinking hard about: “What are the key points in your business where you can accelerate value?” And then, “What are the technologies?” or “How do you mobilize the organization against those?” rather than taking a technology-first view is the right way for most organizations.
Segment two: Author Talks excerpt with Amy Webb
Roberta Fusaro: And now, from digital transformation to DNA transformation: let’s hear from author and futurist Amy Webb about her new book, featured in McKinsey’s Author Talks series, The Genesis Machine: Our Quest to Rewrite Life in the Age of Synthetic Biology [Hachette Book Group, February 2022], coauthored by microbiologist Andrew Hessel.
Amy Webb: Some biotech veterans recently raised $3 billion to create a new company on the premise that the fundamental machinery of living cells can be reprogrammed. There’s a flood of capital being directed at synthetic biology right now.
This is going to have some important knock-on effects on different industries, including pharmaceuticals and healthcare. But we’re also going to start seeing changes in agriculture, industrial materials, even space, because we’ve now proven that it’s possible to reprogram the fundamental units of life.
In the book, we identify nine risks, and they aren’t insignificant. The most worrying data security breaches could actually involve our DNA. This means that this biological era that we’re entering could be an information security problem.
There are easy ways to scrape somebody’s genetic code, and that could have widespread implications if that person is a politician or if that person is the CEO of a big company. Bio cybersecurity is another important emerging problem on a much grander level. We think that synthetic biology, because it promises so much, could actually lead to new geopolitical conflicts.
The reason that we wrote the book actually has to do with solutions. It’s hard to get people to change.
We just saw at COP26, the big climate change conference, that world leaders are just not going to act fast enough to mitigate the climate crisis, especially when what we’re asking countries to do is to stop contributing to their economy by scaling back some of their manufacturing or changing it.
We have to develop alternatives. And in this case, synthetic biology gives us optionality. We’re going to have to make personal choices going forward. We should do that when we’re informed, not under duress.
In this book, we created five scenarios as a way of helping make all of this research and all of this information much more real to people so that they could envision what the future could look like.
The first scenario has to do with a fertility center set in the future. How could synthetic biology influence how we create children in the future? Some of the questions we had included things like, if you could select the genes for your offspring, what would you select?
The second scenario has to do with aging. If people are able to live much longer, much healthier lives, how does that change the future of work? How does that change the relationship between a CEO and an executive team, or a CEO and a board of directors? How does that start to shape what a board of directors might be doing? If you’re a family company, how does that shift the decisions that you make? What does succession planning look like in a world in which people can live much longer than they do today?
The third scenario of the book is a where-to-eat guide. Synthetic biology will influence how and where we get our food, whether that’s meat produced from cells or plants grown in a bioreactor rather than on a farm. Someday, you might have the freshest sushi that you’ve ever had in your life from a bioreactor in Nebraska versus off the coastal waters of Nagoya in Japan.
In the fourth scenario, we explore groups of people who, in an effort to figure out what it would be like to live off-planet, wind up moving underground. It’s a really interesting alternative viewpoint into ways that we might mitigate climate change and live differently.
The fifth scenario we found in a paper written by a couple of academics who were curious to know what could happen with having genetic code in one academic lab somewhere and sending it off to China, which is where the sequences are oftentimes put together and sent back. If you send computer code back and forth, there are always vulnerabilities and possibilities for somebody to inject malware. What if somebody injected malware into genetic code but it was undetected?
What if something like that actually happened? What would be the response? I’m not going to give it away, but the answer is not a good one.
I recognize that some of what’s here is going to be too radical for a general audience, and potentially too radical for CEOs, and too radical even for an audience of scientists. Because what we’re really asking is: What happens when we remove our current evolutionary constraints? What happens when we explicitly view biology as a technology platform?
I think the biggest and the most durable inventions of the 21st century are going to be at the nexus of biology and technology. For that reason, I cannot think of an industry that synthetic biology will not have some impact on over the next decade.
What gives me hope is that between human ingenuity and the science and technology that we have access to, we can right some of our wrongs and can create better futures. But we can’t approach that with outright fear or a complete utopian ideal. We have to take a pragmatic approach.
If we can do that, I think we can live longer, better, healthier, happier lives. I really do. It’s going to be hard work, though.