Anything Worth Buying Once is Worth Buying 10 Times?

I cringe every time I hear someone tell me they went into human resources work because “I’m a people person!” HR obviously supports an organization’s people programs, but there is far more to it than liking people. In many respects, HR is more about program design, development and management than it is about individuals. One aspect of HR program management that needs a lot of work is HR information technology. Agencies are spending billions of dollars on HR IT, yet they are not getting billions of dollars in value. Why is that? And can it be made better?

The answer to the second question is yes, but it will take a change in how IT requirements are identified and systems are funded and acquired. That change is more consequential than it may seem, both in terms of impact and difficulty. The difficulty arises from the answer to the “Why” question about problems with HR IT.

HR systems in most agencies are under the purview of HR offices. In many Departments, they are under the purview of multiple HR organizations at the Departmental headquarters and at the bureau/component/agency/service level. On top of that, there is often no clear process for defining requirements and funding them. When I arrived at the Department of Homeland Security in 2009, I found the Department had more than 300 HR systems of varying sizes and cost. Many of them did the same thing, and in some cases, involved purchasing the same software multiple times. For example, we found more than a half dozen Learning Management Systems (LMS), including two that had each been purchased by different components. Rather than using its combined purchasing power to get the best possible price through enterprise buys, the components operated on their own. A few key systems were standardized across the enterprise, but many were not. The cost of that type of redundancy is substantial. It adds cost to licenses, development, and administration. In the best of circumstance, it results in slightly different implementations of the same system. Other times it results in multiple solutions for the same problem, with no interoperability and greatly increased costs. Just as one example, consider that consolidating to one LMS in DHS could result in savings of $5 million or more per year.

Former DHS CIO Richard Spires and I decided the laissez-faire approach simply could not continue. We found our own organizations were part of the problem. An ongoing turf battle over who controlled HR IT meant key staff who should have been working on common solutions were distracted by constant discussions about who should do what. These were not problem people – they were good people stuck in a bad organizational construct. We addressed the problem by placing the most technical HR IT folks under control of the CIO, but housed in the space of the Office of the Chief Human Capital Officer. We put a seasoned SES in charge of HR IT, and created a program management process to identify requirements, determine funding requirements, and oversee headquarters and component HR IT. We asked for and got a letter from the Deputy Secretary (Jane Holl Lute), requiring that any new HR IT systems be approved by the DHS CHCO and CIO. As a result, DHS is moving toward streamlining and reducing the cost of HR IT.

The issues DHS faced are not unique. They are repeated at the Department and Bureau levels in many places in the government. If every Department and independent agency got complete control of HR IT, eliminated redundancy and improved program management, they might find there is enough money in existing budgets to fund everything they need to provide the best HR technology available. It seems like a simple proposition, but our experience at DHS showed that it upsets too many parochial interests to be easy. Those interests are not limited just to people within the organization. In the case of DHS, much of the HR IT money is in component appropriations rather than at the DHS level. That makes moving the money around to manage smartly a bit more difficult, but not impossible.

So – am I recommending that happen? Should Departments and Agencies manage all of their HR IT at the highest level of the organization? Yes, but only for HR IT that is not common. I believe there is another solution that is better, but admittedly is much more difficult to implement, for systems that multiple agencies use. The solution is not to define the enterprise as a Department or Agency, but rather to define it for core HR systems as the entire Federal government. Core systems, like the LMS I mentioned above, payroll, benefits, timekeeping and record keeping, are so similar from one agency to another that there is little reason to buy them over and over and over and over. One great example is the core HR system that agencies use to process personnel actions. In the Department of Defense, it is a customized version of Oracle HR. In other agencies it is Peoplesoft or a COBOL based system from the Department of Agriculture National Finance Center. Peoplesoft and Oracle HR are both owned by Oracle. Why does the government buy the same system multiple times? Would it not be better to enter into one agreement with Oracle to buy one of their two systems? Even having one agreement to buy each of their two systems would be better than running multiple instances that are each tweaked just enough to make them harder to maintain and not easily used by multiple agencies. If the government used only one such system, it would be far easier to ensure accuracy of transactions when employees move from one agency to another, and it could make it more likely that OPM would have complete and accurate records when employees retire.

Talent acquisition systems are another great example. Agencies are using systems from Monster,, Taleo, Avue Technologies and a home-grown system from the Office of Personnel Mangement. All of those systems have their supporters and detractors. OPM has made great strides in making it possible for those systems to work with USAJobs, the government’s job board, but that still means applying for jobs is different from one agency to another. It might seem the easy solution is to go to just one system, but there has been so much invested in those systems that changing from one to another is far more costly than one might imagine. That doesn’t mean there are not better ways to buy them. Why not have one agency (such as GSA or OPM) enter into an enterprise agreement with each of the providers, with agencies being able to use a modern talent acquisition system without having to go through the process of buying it and recompeting it every few years. While they are at it, establishing a blanket purchase agreement that allows agencies to buy assessment services from a variety of providers would make the process even better.

If the core systems were acquired using a more consistent enterprise approach, agencies could be free to use the resources they save to address other requirements for HR IT that truly are peculiar to their agency mission and workforce. If HR IT becomes a partnership between the CHCO and CIO in every agency where that is not now the case, they will find they can get more done with the same or less resources. The only losers in the process are those who want to protect their parochial interests at the taxpayers’ expense. Not a bad tradeoff if you ask me.


Everything Old is New Again

15 years ago, the Department of Commerce proposed abolishing the National Technical Information Service (NTIS), an agency that serves as a repository for government-funded scientific, technical, engineering and business-related information and provides a variety of services to other Federal agencies. In 1999 their view was that NTIS had outlived its usefulness.  Much of the material they cataloged and sold was becoming available on the Internet, and they believed the agencies whose material NTIS houses could simply maintain their own material. I was the Deputy Director of Human Resources for Commerce at the time, so I became very involved in the discussion. Like several other senior executives and political appointees in the Department, I believed there was still value in the services NTIS offers. While it was true that much of what they sell was available online, much of it was not. There are also a lot of people who want to consume information in printed form. Given a choice of online and free or printed for a fee, they will go with printed. After much internal debate, discussion and arguing, the Department agreed that NTIS would be downsized, but would remain in place. The decision was based in part on the idea that the government invests a tremendous amount of money in producing information and cannot rely on countless web sites, government and non-government, to maintain the information. Much of what is maintained online is driven by traffic. If it gets enough hits, it stays up. If not, it may be deleted. The NTIS decision was based on sound logic, resulted in savings for the taxpayers, and ensured the wealth of information they maintained would be available to the people (you) who paid for it. Fast forward 15 years. There is a Senate bill called the “Let Me Google That For You Act” that proposes abolishing NTIS for the same reasons stated 15 years ago.

I have to give the bill’s sponsors credit for a catchy name for the bill, but beyond that I think there is more to consider than just how the Internet has changed the storage and retrieval of information.

  • The bill references a 2012 GAO report that said 74% of the information NTIS sells is available free on agency websites and in other locations. That means a quarter of such information is not available and it says nothing about how long the information will be on those web sites.
  • NTIS is not the only agency that sells printed versions of documents that are available free from various Sites on the Internet. The Legislative Branch Government Printing Office sells print copies of many documents, including budget reports and others, that are freely available as electronic versions. The simple fact is some people want or need print editions. As much as many of us have embraced technology, some people have not or cannot.
  • “Free” isn’t free. Information on agency web sites has a cost. They pay for hosting, security, programming, site design, maintenance, and other operating costs. Because the money is not charged directly to consumers of the information, we think of it as free when it is not.
  • NTIS operates using a revolving fund rather than appropriated money. Because it has to earn its keep, they sell products and services (full disclosure – my employer ICF International is an NTIS Joint Venture Partner). Agencies such as NTIS that operate using revolving funds have to function much more like businesses. They identify products and services they can sell, determine how they can add value to customers, market their services, and add or subtract staff based on revenues. If they fail to bring in adequate revenue, they suffer the consequences, much like anyone in the private sector. Thus, the market can and does decide whether NTIS should exist.

My views on NTIS have not changed since I was involved with them 15 years ago. I am concerned that much of the information available from NTIS will cease to be available if NTIS is abolished and a well-established service provider will go away.

If there is going to be a public policy discussion regarding the mission and existence of NTIS, my recommendation is that it be expanded into a comprehensive look at how the government maintains and provides access to the vast amounts of information it pays for. Findings resulting from government grants for scientific research, technical data, reports from government agencies, and countless other types of data are produced every day. Taxpayers should have access to the information their tax dollars support. Relying on commercial search engines such as Google, Yahoo and Bing to find the information and hoping agencies will maintain it on their web sites for years doesn’t seem to be the best public policy.

We would be better off if one or more agencies are charged with maintaining and making that information available. The funding mechanism should be part of that debate. Some would argue it should be funded with appropriated money, while others would argue it should be funded by the people and organizations that consume the information. Both ideas have merit.

The issues the sponsors of this bill have raised are valid topics for a public policy discussion. Let’s have that discussion on a broad basis rather than focusing on one small agency. The result could be substantially more access to information at a reasonable cost.