The Work Nobody Sees: Why Good Government Depends on Administration

When people think about government, they usually think about the services they can see.

A police officer responding to a call. A teacher in a classroom. A sanitation truck collecting garbage. A correction officer supervising a housing area. A firefighter responding to an emergency. A newly paved street or a public building under construction.

What most people rarely see is the administrative machinery behind those services.

Someone developed the budget.

Someone established the funding.

Someone wrote the solicitation.

Someone reviewed the contract.

Someone processed the purchase order.

Someone verified the invoice.

Someone recruited the employee.

Someone reviewed the legal requirements.

Someone documented the decision.

And somewhere, someone is probably maintaining a spreadsheet because two government systems that were never designed to communicate with each other still don’t.

After more than two decades working in New York City government, I have come to believe that administration may be one of the most underestimated functions in the public sector.

It isn’t glamorous.

But when administration fails, almost everything else eventually does too.

The Distance Between an Idea and Reality

I currently work in IT procurement and budget for the New York City Department of Correction.

Technology provides an unusually good illustration of the importance of administration because identifying a technical solution is often only the beginning.

Imagine that an agency determines that an aging network switch needs to be replaced.

From a purely technical perspective, the answer might seem straightforward: determine what equipment is required and buy it.

Government doesn’t work that way—and for good reason.

What procurement method is legally permissible?

Is there an existing contract?

Is competition required?

Is the price fair and reasonable?

Is this an expense or capital expenditure?

What fiscal year should fund it?

Does sufficient budget exist?

Does another oversight entity need to approve it?

Are there Minority- and Women-Owned Business Enterprise considerations?

Are there cybersecurity requirements?

Does the scope accurately describe what the agency is purchasing?

Does the invoice ultimately match what was delivered?

Multiply those questions across hundreds of procurement actions, contracts and projects, and you begin to understand the administrative infrastructure underneath a modern government agency.

The distance between saying “we need this” and actually having it available for employees to use can be enormous.

Administration occupies that distance.

Bureaucracy Exists for a Reason

Government employees themselves frequently complain about bureaucracy.

I certainly have.

There are processes I would simplify tomorrow if given the opportunity.

But after spending years dealing with public money, I also understand why many of those controls exist.

If I spend my own money unwisely, I suffer the consequences.

If government spends money unwisely, taxpayers do.

That distinction matters.

Competitive procurement, approvals, documentation, segregation of duties and financial controls aren’t inherently bureaucratic obstacles. They are mechanisms designed to protect public resources.

The problem occurs when we confuse control with complexity.

A good control reduces risk.

A bad process merely adds steps.

Those aren’t the same thing.

Public administrators should constantly ask whether each approval, form and workflow is actually mitigating a meaningful risk.

If five people are reviewing the same information without adding anything to the decision, we haven’t necessarily created stronger oversight.

We may simply have created five opportunities for delay.

Administrative Capacity Is Operational Capacity

One of the biggest mistakes organizations make is treating administrative staffing as overhead that can be reduced without affecting frontline operations.

Eventually, the consequences become obvious.

Reduce procurement capacity and purchases take longer.

Reduce budget capacity and forecasts become less reliable.

Reduce contract-management capacity and renewals become emergencies.

Reduce human-resources capacity and vacancies remain open.

Reduce accounts-payable capacity and vendors wait longer for payment.

Reduce analytical capacity and managers make decisions with poorer information.

The frontline service may initially appear unaffected.

That can make administrative reductions tempting.

But administrative debt accumulates much like deferred maintenance on a building.

For a while, everything looks fine.

Then the failures begin appearing elsewhere.

The contract expires before its replacement is ready.

The funding isn’t available when a project reaches procurement.

An invoice sits unresolved.

A vendor stops performing.

A purchase becomes an emergency because nobody had sufficient capacity to begin it six months earlier.

An audit identifies documentation that should have existed but doesn’t.

What appears to be an operational failure may actually have begun months or years earlier as an administrative-capacity problem.

Technology Can Help- But It Won’t Fix a Bad Process

Artificial intelligence and workflow automation create enormous opportunities for government administration.

I use technology constantly and believe government should be much more aggressive about eliminating repetitive administrative work.

Software can route approvals.

Systems can flag expiring contracts.

Dashboards can identify budget variances.

Artificial intelligence can help analyze documents, compare proposals, summarize contracts and draft routine material.

But there is an important caveat.

Automating a bad process does not make it a good process.

It makes the bad process faster.

Before government digitizes a workflow, someone should ask a very simple question:

Why are we doing this step at all?

If nobody can explain the risk being controlled, the information being created or the decision being made, perhaps that step shouldn’t survive modernization.

Government transformation should therefore begin with process mapping, not software procurement.

What information enters the process?

Who touches it?

What decision does each person make?

Where does it wait?

Where is information entered twice?

Where does one system create a document that someone manually retypes into another?

Those are the places where administrative modernization can produce enormous returns.

We Also Need Better Administrative Career Paths

There is another problem government should address.

We frequently celebrate specialists while underestimating generalists.

Yet some of the most valuable public employees are people who understand how multiple systems intersect.

They understand procurement well enough to talk to buyers.

Budget well enough to talk to finance.

Technology well enough to talk to engineers.

Contracts well enough to talk to lawyers.

Operations well enough to understand why something is urgent.

And organizational politics well enough to know how to actually get something completed.

Those employees become institutional translators.

Government needs them.

We should develop clearer career paths for administrative professionals who acquire that kind of cross-functional knowledge.

The person processing a purchase today may eventually become the person redesigning the agency’s entire procurement operation.

The analyst reviewing expenditures may eventually manage a multimillion-dollar budget.

But only if we treat administration as a profession worthy of development rather than simply a collection of support tasks.

Measure Administration by Outcomes

Administrative organizations also need better measures of success.

Counting activity isn’t enough.

Processing 500 transactions tells me something about workload.

It tells me very little about performance.

I would rather know:

How long did transactions take?

How many were returned for correction?

How much competition did we obtain?

How much money did negotiation save?

How many contracts were renewed before they became emergencies?

How accurate were our spending forecasts?

How many invoices exceeded payment targets?

How many manual steps did we eliminate?

Those measures turn administration into management.

They also help answer an important question when resources are constrained: where will another employee or another technology investment produce the greatest return?

Government cannot manage what it refuses to measure.

Administration Is Public Service

I come from a family with generations of government employment, and I have spent more than 22 years in New York City government myself.

One thing that experience has taught me is that public service takes many forms.

Some are highly visible.

Others happen behind computer screens, inside conference rooms and buried within budget systems.

The employee who negotiates a better technology contract may never meet the taxpayer whose money was saved.

The analyst who catches an incorrect invoice may never receive public recognition.

The procurement professional who gets equipment delivered before an operational deadline may never appear in a press release.

That’s fine.

Government isn’t supposed to exist for our recognition.

It exists to produce outcomes for the public.

But leaders should recognize something internally that the public understandably may not:

Administration is not separate from service delivery. Administration makes service delivery possible.

The objective should never be bureaucracy for bureaucracy’s sake.

It should be the smallest, smartest and most accountable administrative structure capable of protecting public resources while allowing the people delivering services to do their jobs.

After more than two decades inside government, that is increasingly how I define good administration.

Not more process.

Not fewer controls.

Better ones.

And when we get that right, most people will never notice.

In government administration, that may be the best evidence that it is working.