This year's Workday Innovation Summit was renamed the Workday AI/ML Innovation Summit. Over the course of two content-packed days, the Workday team made its AI/ML innovation case to the assembled analysts in San Francisco.
In an upcoming video filmed by Constellation's Holger Mueller, we hashed out the themes of the show. Yes, Workday's AI/ML plans took center stage, but there was more. Workday is also serious about moving beyond their HR vendor roots, to being equally relevant to the CHRO, CFO, and CIO as well. A cynic could argue that this is a branding exercise, but to me, discussions across process areas are welcome - silos are always counterproductive.
Leadership was also a big theme - for many analysts, this was the first time to engage with Co-CEO Carl Eschenbach (announced December 2022). When overly-opinionated analysts gather, debates are plentiful. This year's included:
- The problem of bias/misuse in enterprise AI - and the role of the vendor and customer in addressing it.
- How to bridge the gap between HR and Finance, and address the concerns of both sides. Can collaborative planning help?
- Workday's take on generative AI, and how disruptive generative AI will be for the enterprise.
Some think enterprise tech is doomed to lag behind the ferocious pace of consumer tech. But in this week's Enterprise hits and misses, I explained why I think the enterprise can, shockingly enough, take the lead in responsible/effective approaches to generative AI - and why Workday's approach to AI, with built-in guardrails, quality data sources, increased explainability and iterative reinforcement learning, is categorically different than the "ChatGPT unleashed" experience we've had, for better and often for worse, in the consumer tech domain this year.
My pick for underrated Workday AI use case: Skills Cloud
For Mueller's video, I also picked my underrated themes of the show. My top pick? The impact of Workday's Skills Cloud. Amidst the sexy generative AI talk, why Skills Cloud? Because I am ultimately swayed by two things: customer impact and customer adoption.
To me, building an enterprise around skills goes beyond HR. It is the future - and some Workday customers are far enough along to share instructive stories. Oh, and this is not far from cutting edge AI anyhow. Behind a skills-based approach to workforce development is a seemingly mundane thing called a skills ontology. But a proper skills ontology is loaded with relevant data. AI is only as potent as the data that powers it. But why skills, and why now? After all, skills ontologies are not new. Historically, a big problem was populating them, and keeping them updated.
Start with the adoption: Workday launched Skills Cloud in 2018. Now, out of 4,000+ Workday customers, 1,500 are on Skills Cloud, with five billion skills in use (more than half of Workday's HR customers are using it). As Workday points out, treating this as a skills management exercise only takes you so far. The big shift happens when you manage your talent via a skills-based approach:
Skills management sees skills as part of the work equation, whereas a skills-based approach sees skills as foundational to running an organization and achieving business goals, not just one more thing. It should be clear by now that to attract and retain workers, companies will need to focus on skills.
Accenture and Workday - how Accenture shifted to a skill-based approach to talent
I'm probably a tad idealistic about how far organizations should go to compete for talent. We need proof points - at scale. How about Accenture? If a services firm like Accenture gets talent wrong, they're done. Colin Anderson, Managing Director at Accenture and Accenture's Digital HR, Planning and People Analytics lead, explained their talent approach to Workday Chief Customer Officer Sheri Rhodes:
Exactly as the title says, I run a digital HR. So all of our people technologies, planning, forecasting - that includes what we call our talent supply chain. All of our staffing, and basically how we match people with clients and project demands - Workday is at the core of how we think about enabling our people and our business.
Accenture definitely puts Workday to my scale test. Anderson:
We have 738,000 employees, all of which are on Workday. We do a lot of acquisitions as well, all of which also fold into Workday... We've really built a playbook to say, 'How can we make these acquisitions, fold them and keep the culture? Keep what's unique about them?' And so basically, 'How do we have a flexible, dynamic set of technologies?'
Accenture's skills transformation traces back nine years:
Over the last two days, seeing all the focus on Skills Cloud, that's something that we really built together - it's been a lot of fun to see the excitement from you all. We started our skills journey almost a decade ago. 2014 was when we declared that we wanted to be a leader in conditional talent.
What dose that entail?
That means there's new technologies all the time; that means new things are coming and going all the time. We needed a way to work that wasn't job-based, because that wasn't going to work for 'How do you become the leader in digital talent'?
But there was a major obstacle. Yep, Accenture had that good 'ol skills ontology problem:
We knew that we needed skills, but we needed a skills-based way of working that was totally dynamic, where we can add skills, skills-to-go, and they could change all the time. And so we built that ourselves. We built a skills library and ontology; we went through the tremendous work of stitching skills into all of our systems.
Now live on Workday Skills Cloud globally, Accenture's skills-based approach is in high gear. This is where Workday's embedded AI approach comes into the skills picture:
We have amazing opportunities at Accenture; with 738,000 people; we do almost everything somewhere. So for our people to be able to say, 'I aspire to do this.' And then for us to be able to have learning recommendations and project recommendations that help them get better. That means, again, better career opportunities for them, a more engaging place for them to be - a need for us to hire [more experienced] hires is certainly a lot less, and the ability to unlock the potential of those people. Again, we can understand what might they be good at. Being skills-based is what allows us to do that.
But Anderson's comment on degree requirements was my sit-up-in-chair moment:
One of the things that we're really proud of: 45% of our roles in the US, just as an example, don't require a four year degree. Skills are the way that we've been able to get ourselves comfortable with that, where we can say, 'Oh, let's look at the potential of people instead of their pedigree.' 20% of those entry level people we hire don't have a degree... Again, that's because we're able to really understand the potential in the skills of people in a way that's different from the university.
To me, this is what AI will look like in the enterprise. It won't look like AI at all. It will just be a way of doing our jobs better. As Anderson put it:
We basically have this approach of skills-based everything... So you can start with recruiting, or someone planning: 'How many people am I going to need?' One of the things that we've done is: we've worked a lot with our sales teams to say, 'Let's not have skills just be an attribute of a person. Let it be an attribute of our demand of what we're selling as well.' Now as people are solutioning, and pricing in these types of things, they're thinking about the skills that people are going to need. And that allows us basically real-time to say, 'Oh, great, well, how do we merge our algorithms to say, what we're selling solution is going to need this. Therefore, what we need to be thinking about developing, hiring, training, etc, needs to be that way.'
My take - generative AI is a small part of enterprise AI
Talk all you want about generative AI. There will be good enterprise use cases; Workday has a number of them underway (see Workday Co-President Sayan Chakraborty's How AI and ML Are Powering the Future of Work). Some of Workday's generative AI use cases are not yet public; some were shared at the AI/ML Innovation Summit. What grabs me are the AI use cases already changing work for the better. Given the potential for AI misuse and overhype, this is actually a pretty high bar.
I get why Workday wants to emphasize its enterprise platform well beyond HR; a Fannie Mae customer told the story of moving to Workday Financials at a massive scale. Colin Anderson said that Workday Adaptive Planning brought Accenture's finance and HR leaders into much closer collaboration, addressing one of the key concerns raised by analysts at the show (bridging the HR/Finance gap). But to me, the Skills Cloud story stood out - not just as an HR story, or a Workday story, but as a notable enterprise AI story. I'm not the only one. As Josh Bersin wrote after the event:
The Workday HCM suite is in the strongest shape I’ve seen in years. The Workday Skills Cloud is maturing into a skills intelligence platform and it now has features that make it almost essential for a Workday customer. It can import data from any vertical or specialized skills database, it gives companies multiple ways to infer or assess skills, and it gives you dozens of ways to report on skills gaps, predict skills deficiencies, and create upskilling pathways for each employee or workforce group. I’ve watched this technology grow over the years and never before have I seen it so well put together and positioned to do what companies want.
During our soon-to-be-released video, Holger Mueller noted the main concern about anything that relies on skills ontologies: keeping the data accurate and completely up to date. Fall behind even a bit, and you're planning inaccurately. The system must intelligently auto-populate where possible, and users should be able to easily update and even submit skills overrides. From what I saw last week, Workday is delivering that, but customers need to evaluate such things closely, to make sure it will work for their users. I remain interested in hearing more on VNDLY - Skills Cloud integration. I see possibilities managing skills across both employee and contingent workforces. Though as VNDLY has pointed out to me, there are risks to address there too, in terms of the legal employment differences between employee and contractor.
As Anderson said, we could take skills further: he'd like to see a day when employees had a "skills wallet" they could take with them, from employer to employer. For years, I've carried on about the limitations of imposing degree requirements on talent. It's such a blunt force, exclusionary screening tactic. To hear Accenture talk of moving beyond that is welcome. As Anderson told us:
Becoming a skills-based organization is not a project. It is a new way of working.
There is another fascinating angle to Accenture's use of Workday, one that I wrote about last fall at Workday Rising: the potential use of Workday Peakon Employee Voice for monitoring Accenture's project sentiment in (near) real-time. Anderson told me that this use case is progressing. It does require the customer to have a Peakon license from the project kickoff, but Anderson said from the Workday side, Workday is helping to make this work well for customers.
Whether there could ever be a freemium version of Peakon to support this use case is an intriguing question; I believe it could help Workday (and Accenture) change the monitoring of project quality in an even broader way. But, like any good event, you run out of time before you run out of questions.