Basically, I told the commission that states should apply the principles of buying a home to procuring major IT systems. So how does that track?
When most of us buy a house, we don’t draw up a wish list and worry about the cost later. No, we figure out what we can afford, get pre-approved for a mortgage, and let the budget shape what we look for. We work through tradeoffs deliberately, whether that’s more space or better commute, new construction versus an established neighborhood, or what we buy now versus what we can add later.
We also hire a real estate agent who knows the market, helps us weigh priorities, does extensive house hunting to inform our search, negotiates our contract terms, and generally holds our hand throughout the homebuying experience. A good agent also asks hard questions, like whether you can maintain the home yourself or should consider a condo where upkeep is built into the homeowners association. These conversations force clarity about what you really need versus what you want before making a commitment.
We take all these steps because buying a home is one of the most consequential decisions a family makes. So it’s important to follow the right process, leverage expertise and refine what you want before you make that big call.
So why don’t we apply the same thinking to major IT systems? For projects that will serve a state for decades and cost hundreds of millions of dollars, we often skip the budget conversation until requirements have already expanded beyond what is realistic. We issue solicitations without a clear price signal, leading vendors to submit proposals based on entirely different assumptions, resulting in bids that were never designed to be comparable. Rather than one trusted adviser with a view of the whole life cycle, we hire different consultants for each phase, and the institutional knowledge built during planning evaporates by the time implementation begins. A thoughtful homebuyer would never accept these conditions; governments shouldn’t accept them for large IT projects either.
Here are four foundational concepts that any state should examine seriously. None of them are flashy or novel. They are common sense, and they build on approaches that governments already know how to use.
FINANCE IT AS INFRASTRUCURE
You don’t pay for a house with cash saved in the bank. Rather, you take out a loan and pay it off over time. Major IT systems, whether benefits platforms, payroll systems or tax administration, serve a state for decades, yet we typically fund them year-to-year through the annual budget process. This creates stop-and-start disruption, erodes team continuity mid-project and pushes departments to over-scope their requests because they fear this may be their only realistic chance to modernize.
States already finance roads and buildings through bonds with milestone-based disbursement. We should do the same for large IT investments. I made this case in a previous GovTech article where I wrote that IT infrastructure warrants long-term financing, with oversight conducted by a board delegated authority by the governor and Legislature. California’s State Public Works Board offers one model worth studying. That kind of structure, adapted for IT, would give technology investments the same continuity we already give physical infrastructure.
REQUIRE AN OWNER'S REPRESENTATIVE FOR EVERY LARGE IT PROJECT
Just as a good realtor stays with the homebuyer from search through close, a qualified owner’s representative should be engaged for the full IT project life cycle, from pre-discovery through project completion.
California already applies this for state construction. At the California Department of General Services (DGS), where I served as director, every major DGS building project began with the hiring of a criteria architect, an independent architectural and engineering firm engaged from requirements development through construction completion, working alongside DGS’ own construction managers. The criteria architect helps the state define what it actually needs, helps develop the design criteria that go into the solicitation, and provides oversight and counsel to the department to ensure the contractor delivers what was promised. These firms come from private industry, know the latest innovations and can write RFPs well because they respond to them for other projects. Because the criteria architect cannot bid on the construction contract themselves, their only incentive is to represent the state’s interests. A pool for qualified IT owner’s representative firms would make it easy for any state’s departments to access this expertise quickly, without running a full competitive procurement each time.
BUILD STRUCTURED DISCOVERY INTO THE SOLICITATION PROCESS
You do not buy the first house that meets your minimum requirements, and your priorities change the more you look. Similarly, state departments should engage vendors in a structured way to solicit input before finalizing an RFP. For example, a state can issue a multistage process where finalists are first selected on qualifications, then submit their proposals, followed by multiple meetings to better clarify requirements, estimate costs and generally reduce uncertainty before the final solicitation is released.
California’s Franchise Tax Board (FTB) used this kind of approach on its Enterprise Data to Revenue 2 project, a multiyear modernization of the systems that support tax compliance and revenue collection. FTB held structured confidential sessions with vendors, refined their requirements based on what they heard and amended the solicitation transparently so every bidder had equal access to the same updated information. The result was a more lengthy and rigorous solicitation process, but one that ultimately resulted in better bids and fewer surprises during project implementation.
ESTABLISH IT GOVERNANCE THAT KEEPS THE RIGHT PEOPLE ENGAGED
When you buy a home, you make the big decisions yourself. You do not delegate the tradeoff between a larger house and a shorter commute to your agent. In state IT, the equivalent decisions — meaning what a system must do, what gets phased for later, what business process must change to fit the platform — too often drift down to project managers who lack the authority and organizational standing to make them well. The FTB experience is instructive here too. FTB structured its governance so that subject matter experts (SMEs) had a formal role in surfacing requirements throughout the process, and department executives were actively making and owning the key tradeoff decisions rather than deferring to consultants. That kind of governance structure, where SME input is built in and executive decision-making is expected and timely, is what separates projects that deliver from projects that drift.
Buying a home is a big decision. Consequently, we approach it with serious discipline because the stakes are real, and we live with the consequence for decades. Major IT procurement is also a big decision, with even higher stakes and longer consequences. The concepts I have described are not a new theory of government technology. They are the foundational practices that careful buyers apply to important purchases, and I hope states across the country will see them as a basis for better procurement of large, complex IT projects.
