Flyzik: Cheaper and better

So I've been hanging out in the private sector — some call it the dark side — for about a year now. Although I still have a lot to learn, I've noticed a few fundamental things that make me want to offer some ideas on how the government and private sector can work together to make progress in many areas.

I understand from the government's point of view the many constraints and obstacles that have slowed major initiatives in the past. But I am now gaining an appreciation from the private sector's perspective of why some of the intuitively obvious things government officials want to do aren't so simple.

That being said, I believe there are some things that can be done to save the government considerable money and engage the private sector in new ways.

I call it my "Top 10 Ways to Save a Billion...or Two."

Drum roll, please...

No. 10: Make enterprise license agreements.

We need to create enterprise agreements in phases.

The first step is to determine which of an agency's bureaus has the best deal and negotiate it for the entire agency. As each organization's license expires, it would be required to automatically convert to the agencywide agreement.

The second step is for Office of Management and Budget and General Services Administration officials to review agencies' enterprise licenses and determine which has the best deal. Then, GSA can negotiate it as a governmentwide license.

Over three to five years, this will result in savings of several hundred million dollars.

One important note: This approach does not stymie competition in the software industry. In fact, it encourages agencies and vendors to be more creative. Smaller companies can play just as easily as the big guys.

No. 9: Use code converters to modernize code and Web-enabled applications.

Haven't these companies been around since the Year 2000 preparations? Don't they convert about 90 percent of the code and then leave you struggling to complete the rest — at a much higher cost? Well, I was skeptical about this one, too, until I spent some time with a few companies doing this kind of work.

If we can reduce the costs of all the ongoing government modernization programs by a factor of 10, we can save money.

No. 8: Use network storage approaches.

We all lived through the days of decentralizing the computer environment, which we refer to as end-user computing or client/server approaches, and now the trend seems to be back toward centralization.

But there is no reason to centralize information storage. Look at the cost, performance and continuity advantages of network-based storage approaches vs. centralized storage.

No. 7: Consolidate the telecommunications

infrastructure.

Easier said than done, right? But we can save a lot here. There are many redundant circuits out there. Consolidation has not been done yet because it disrupts the culture, people and processes that exist in government. Agency officials want to control their destinies and make their own decisions.

But this can be done today quite simply because it has to be. It is no longer a question of turf. It is an international economic competitiveness issue and a national security matter. The war on terrorism and e-government programs require information sharing at all levels of government.

As GSA embarks on the next phase of its FTS 2001 program, I recommend establishing an interagency committee to make this a priority. OMB needs to enforce compliance.

No. 6: Implement wireless solutions.

Does anyone doubt that wireless Internet access is on the verge of taking off? The Internet grew exponentially when we were tied to a wire in the wall. Today, with broadband speeds being realized on wireless networks, anywhere, anytime computing is here.

Regarding security, wireless may not be appropriate for all requirements, but short of classified operations, the time has come to implement wireless in ways that dramatically reduce costs.

Let me give you an example: I have four computers in my house. One is connected to a wire and the other three communicate wirelessly. It cost me $29.99 to implement the wireless network. What would it have cost me to run wires through the house or order more telecommunications lines? Project this scenario to a large government agency.

No. 5: Use automated tools to certify and accredit systems.

I don't know if I can call this savings, but it is cost avoidance.

Agencies need to get systems certified and accredited, yet they continue to put this off even though every malicious code attack requires that they spend billions of dollars in recovery costs. The problems are the costs of certification and the way we do it.

Traditionally, agencies bring in consulting firms that deal with systems sequentially and run up labor-hour costs. But there are tools that dramatically reduce certification time and costs. You have to pay for licenses to use the tools, but compare that to the costs of the sequential approach — or the costs of running vulnerable systems. Certification and accreditation represent a small fraction of those costs.

No. 4: Use multilevel security approaches to

eliminate the need for multiple workstations.

How many classified applications are running today using a physical security approach with two or more machines at each desk? Agencies need to get serious about using multilevel security architectures to eliminate this added expense. It is a security architecture issue, not a product issue.

No. 3: Automate the paperwork associated with

human resources and other administrative systems.

There are examples where this has worked. New York City did it. Officials invested $7 million and saved $56 million. Unfortunately, federal agencies have been unable to get the money necessary to automate administrative systems because mission systems get the lion's share of funding.

So what should agency officials do? Take advantage of private-sector resources.

No. 2: TALK, TALK, TALK!

If I have learned anything in my first year in the private sector, it is the difference in how government and industry officials buy things. In both cases, procurement processes start the same, but they progress in remarkably different manners. In the public sector, the level of interaction between the government and vendors is reduced as the award date nears.

Contrast this with a private-sector acquisition. The bidders and buyers meet daily and have detailed conversations about the requirements and the proposed solutions. The parties meet informally around a table and make sure they get the solution right before the award is made.

How about a government procurement that ends like this: Two weeks prior to the award, each of three bidders is given two hours every day to meet with government program managers and contracting officers to talk through their strategies and specific approaches. The feds freely interact with the bidders, and both sides make sure they understand what is required and what is being bid.

We can save some money if we get it right the first time.

And No. 1...

Get someone willing to do all these things at no cost to the government, perhaps a share-in-savings program to review your existing environment, and apply these 10 steps.

Flyzik is a partner with the government technology consulting company Guerra, Kiviat, Flyzik and Associates Inc. He left government in 2002 after 28 years of service.