Flowers, rice, Mendelssohn, cake. It's a shotgun wedding between the information technology and human resources functions. Neither party is sure it wants to tie the knot with the other, but they have no choice. In this case, they are being brought together by a powerful weapon called learning technologies. Let's review the lineage of these two families. HR is the home of the people people -- those helpful individuals whose job it is to assist in recruiting, compensating, evaluating, retaining and eventually removing people from organisations. IT, on the other hand, is focused on machines -- procuring, installing, programming, maintaining and eventually unplugging them. On the surface, you couldn't find a less likely set of partners. In some ways, however, HR and IT have a lot in common. HR uses machines to manage people and IT uses people to manage machines, and both have to deal with a mixture of the two worlds. Both are staff functions and enjoy relatively low regard within the organisation; when they get together, they can console each other. Perhaps their reputation problems are based on the fact that both are responsible for something (people and technology/information) that is really too big, too pervasive and too important for a single business function to address. HR and IT have so many similarities that my old boss Jim Champy (this was in his pre-guru stage) once read an article by Peter Drucker called "Goodbye to the Old Personnel Function" to a conference of IT managers, and the audience thought he was reading about IT.
There have also long been similarities between these two groups on the information transaction side. Now, you know what interest the IT organisation has in information transactions, but you may not realise that HR is hardly a stranger to transaction processing. The first IT application in business was, after all, payroll (way back in 1953). Some pretty large companies, including PeopleSoft, made their mark in HR transaction systems. A very large group of people work on HR transactions systems; they even have their own association, the International Association for Human Resource Information Management, with tens of thousands of members. Dave Ulrich, a University of Michigan professor and HR expert, argues that HR functions don't win the right to address higher-order issues unless they have first got the transaction correct. Of course, the same might be said about IT.
But neither HR nor IT are solely concerned with transactions. HR, in addition to being on the line to ensure that you get paid and are charged the right amount for dental insurance, is supposed to develop new skills and expertise among employees. Forward-thinking HR people, who generally resided in something called management development within the function, began to get excited about a concept called organisational learning (OL). Many of them became devotees of Peter Senge and his mysterious-sounding book, The Fifth Discipline: The Art and Practice of the Learning Organisation (Currency/Doubleday, 1994), full of esoteric concepts such as systems thinking and personal mastery. The OL movement was an upscaling breath of fresh air compared with running a bunch of training courses at the local hotel (so much so that many HR people, according to my friend Jenifer Lippincott, who is director of business development at Consultec, now refuse to use the "T" word). The only problem is that these noble ideas proved to be difficult to implement on a large scale. HR people were left with a desire to radically upgrade employee learning, but the OL movement didn't specify the tools that would make it feasible. The good old IT function has worked for the last four decades or so to provide lots of transaction data for the organisation. It has largely succeeded in its mission -- to such a degree that now most people have far more data than they can manage. Increasingly, it is becoming apparent that if no one is actually informed by or learns from the information provided over the IT infrastructure, then there's not much point to the whole endeavour. IT people have, over the past couple of years, been jazzed about a concept called knowledge management, in which information technologies have been harnessed for the purpose of helping knowledge workers do their jobs.
They build repositories of knowledge and value-added forms of information. But how do you ensure that people actually receive value -- perhaps even learn -- from the knowledge embedded in Web sites and Notes databases? Some argue that knowledge management has stalled because senior managers find it difficult to determine its value and its relationship to improved performance.
The Urge to Merge
Clearly, both HR and IT have moved beyond transactions to focus on such highfalutin' concerns as knowledge and learning. Just as clearly, both face critical challenges in converting these concepts into improved human and organisational performance. They're both in the church, but the preacher, the vows and the compelling urge to merge are missing.
Ah, here's the preacher: it's Gloria Gery, a pleasant-looking cleric with a strong following among the "integrated performance support" faithful. She's been giving sermons on the topic for more than a decade; the bible of the movement is her book Electronic Performance Support System (originally from Digital Equipment Press but for obvious reasons now offered by Gery Associates). As explained to me by Phil Tierney, a Gery devotee at Intel who designs performance support systems at Intel and who also forswears the "T" word, the concept requires technology to support learning on the job at the time when the learner needs it. The learning should not remove the learner from the business transaction but should be integrated with it through the system interface. The knowledge presented to the learner must be appropriate to the task and the worker. Gery is strict about these vows and will demand that any pairing of HR and IT adhere to them. However, performance support has been around for a while and hasn't led HR and IT to the altar. Perhaps it's because the concept has been applied primarily to transaction rather than to knowledge work. Gery and performance support advocates have railed, for example, against technologies that cover up the transaction screen to deliver learning. This is a real concern with transaction work, but knowledge workers -- a group that is surely growing in our companies and societies -- need learning too, and they may not even have transaction systems on their computer screens.
The real shotgun for this marriage is what Lippincott calls learning technology infrastructures -- typically Web-based platforms that contain a variety of knowledge types. These infrastructures might include performance support systems that are closely integrated with the job but could also incorporate reference information, product specifications, customer knowledge and so on.
This latter sort of online knowledge is extrinsic to the job and the transaction system and thus, anathema to performance support purists. But for knowledge workers, who are used to searching widely for knowledge and occasionally have a few moments to reflect on their learnings, an extrinsic knowledge delivery system may not be so problematic. Both Tierney and Lippincott respect Gery's work and took the vows of performance support, but like others who are attempting to work out the marriage of HR and IT on a daily basis, they've relaxed some of the vows in practice (as in my own marriage, where my wife always loves me, occasionally honours me, but seldom, if ever, obeys me -- come to think of it, she edited that last word out of her vows). At first glance, some of these Web-based learning infrastructures look a lot like knowledge management systems. They contain the same types of knowledge content and have similar search and retrieval capabilities. They may even involve some personalisation and push capabilities.
The difference between these learning technologies and most knowledge management applications, however, is that the former are designed with reference to a particular job or set of jobs. They may not be as tightly integrated with the job as traditional performance support systems are, but they are almost always built to support a specific business process.
Saying "I Do"
If the IT half of the marriage (probably the groom, since IT departments are disproportionately male) brought new tools and a strong interest in knowledge management, the HR half (more likely statistically to be the bride) brought a strong orientation to improving job performance and a focus on knowledge use.
Knowledge managers, and the IT profession more broadly, have often attempted to put as much knowledge into repositories as possible, without regard to how the knowledge will be used in the context of particular jobs. As a result, the repositories don't support any particular job very well. Performance support's emphasis on particular job contexts and knowledge management's reliance on knowledge workers and a wide variety of learning objects make for a healthy, diverse relationship. So these two functions are joined in holy matrimony until death do them part. They've merged to such a degree that it's often difficult to tell which camp particular practitioners came from. Lippincott works with IT people, HR people and business sponsors to sell and deliver these systems. Although Tierney's background is in the "T" word and performance support, he's now part of an IT group at Intel that's developing knowledge delivery systems for the company's SAP R/3 implementation. At recent knowledge management conferences, I've heard people from IT functions speaking about organisational learning and performance support and I've heard people from HR functions advocating knowledge repositories and intranets. CIOs talk to me about learning, and chief learning officers talk to me about knowledge management and technology.
Like all marriages, however (except mine, of course), there are occasional setbacks. The IT world, for example, is buzzing with enthusiasm about corporate portals -- one-stop shopping destinations for all information and knowledge content within the organisation. This idea seems to have worked pretty well on the Internet; billions of dollars in market value for companies like Yahoo and Lycos tell the story. But I think it's a bad idea for internal corporate applications. It's the knowledge management fallacy run amok; there is nothing resembling a specific job context for the application of portal-based knowledge. It's one-size-fits-all knowledge delivery. Let's hope that IT's better half, the HR/OL/performance support crowd, kills this idea before it goes too far. There are also going to be persistent communication issues. IT is from Mars, HR is from Venus, and so forth. Many IT people will prattle on about Web sites, intranets, and access to information and knowledge. Most HR people will prefer terms like performance support, human performance and organisational learning. Good HR-IT hybrids will have to be "semantic integrators", attempting to communicate in terms the other side understands and eventually creating a language that's appealing to both.
The progeny of this marriage will be the real HR/IT hybrids. Future generations will draw from the best features of organisational learning, IT infrastructures, performance support and knowledge management, and their hybrid vigour will dramatically improve organisational and individual performance.
Thomas H Davenport is a professor of management information systems at Boston University School of Management and director of the Andersen Consulting Institute for Strategic Change
Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.