Almost every government in Africa has committed to good governance. The commitments appear in constitutions, party manifestos, national development plans, ministerial speeches and the founding charters of regional bodies. Very few of them are stated in a form that anyone can verify.

That is the heart of the problem. Transparency, accountability and integrity are dispositions, and dispositions cannot be audited. Practices can.

The independent indices that track governance on the continent tell a consistent story about what happens when the two are confused. The Mo Ibrahim Foundation’s Index of African Governance has reported that continental averages improved over the past decade in areas such as human development and economic opportunity while deteriorating in security and the rule of law — meaning that visible progress has coexisted with erosion in the foundations. The World Justice Project’s Rule of Law Index and Transparency International’s Corruption Perceptions Index show a similar divergence between countries that installed specific institutional machinery and countries that adopted the vocabulary without it.

The distinction matters for anyone actually trying to reform an institution. You cannot improve a value. You can improve a practice, on a schedule, with a published number attached to it. Here are six that carry most of the weight.

Publishing the budget in a form citizens can read

Fiscal opacity is the precondition for almost every other governance failure, and it is also the easiest to correct, because the documents usually already exist.

The practice is to publish, on a fixed calendar, the core budget documents that the International Budget Partnership’s Open Budget Survey assesses — from the pre-budget statement through the enacted budget, in-year and mid-year reports, the year-end accounts and the audit report — together with a plain-language citizens’ budget. What separates a serious system from a decorative one is timing and format: documents released before the decision they are meant to inform rather than after it, published as machine-readable data rather than scanned images, and classified consistently from year to year so that trends can be constructed.

Where allocation and release data are genuinely public, journalists, legislators and civil society perform a large share of the oversight work at no cost to the state. Where they are not, no oversight body is large enough to compensate.

Making procurement open by default

Procurement is where the largest share of public money meets private incentive, and competition remains the cheapest anti-corruption instrument available. But competition requires that potential bidders know an opportunity exists.

The practice is to publish tender notices, bid evaluations, awards, amendments and completion records in a standard structured format — the Open Contracting Data Standard is the established schema — and to make publication a condition of a contract’s validity rather than a courtesy extended after the fact. Ukraine’s ProZorro platform is the case most often cited, and the reason is not the website. It is that the system inverted the default: publication became automatic and secrecy became the exception requiring justification.

The measures that reveal whether it is working are straightforward. What share of procurement value passes through the open platform. How many bidders appear per tender. What proportion of contracts are awarded single-source. And how many awards ever receive a published completion record.

Verifying asset declarations rather than merely collecting them

Nigeria and Ghana both require declarations of assets from senior public officials. The gap is almost never at the filing stage.

A declaration regime with full compliance and no verification is a filing cabinet. The deterrent effect comes entirely from the credible possibility of being checked, which means that a serious system verifies a random sample every year, cross-references declarations against tax, land and corporate registries, publishes the verification rate, and attaches automatic consequences to non-filing that do not require a political decision to trigger.

The numbers worth publishing are the filing compliance rate, the proportion of declarations sampled for verification, the number of discrepancies referred onward, and the median time from referral to resolution. The last of these is usually the most revealing.

Answering requests for information

Nigeria’s Freedom of Information Act and Ghana’s Right to Information Act put the legal foundation in place. The practice that determines whether the law means anything is administrative: every public body designates an information officer, logs every request received, and publishes quarterly statistics showing how many requests were granted, how many refused, and on what stated grounds.

Good performance looks like response within the statutory window as the norm rather than the exception, refusals reasoned in writing against a specified exemption, and an appeal route that does not require litigation to use. The published statistics also serve a diagnostic purpose beyond information access itself: an unresponsive information desk is a reliable proxy for wider administrative dysfunction, and the pattern across agencies tells reformers where to look first.

Closing the loop between audit and consequence

Most countries already have an audit function. What they frequently lack is follow-through.

The full practice has two halves. On one side, an Auditor-General with security of tenure, an independent budget and unrestricted access to records. On the other, a Public Accounts Committee that holds hearings on every audit report and maintains a public register of its recommendations showing who is responsible for each, by when, and whether it has been implemented.

What matters is measurable: how many days pass between the end of the financial year and the tabling of the audit report, what share of audit queries receive a formal response, and what proportion of committee recommendations are implemented within twelve months. Audit findings that produce no consequence teach an institution that the audit is a ritual, and that lesson is learned very quickly.

Measuring service delivery at the counter

Citizens do not experience governance in the aggregate. They experience it at one counter, on one morning.

The practice is to publish service standards for the transactions people actually undertake — passport issuance, land title registration, business incorporation, court case listing, connection to the electricity grid — showing the target time, the actual median time and the official fee. Crucially, performance should be published by office rather than only nationally, refreshed monthly, alongside complaint volumes and resolution rates.

Disaggregating to the office level does two useful things at once. It converts an unmanageable national problem into a set of manageable local ones. And it makes visible the offices that are already performing well, whose methods can then be copied rather than invented.

Alignment with the global goals

These six practices map directly onto Goal 16 of the Sustainable Development Goals, which calls for a substantial reduction in corruption and bribery, for effective and accountable institutions at all levels, and for public access to information. They also underpin progress elsewhere. Investment follows contract enforceability and predictable administration, which is Goal 8. Infrastructure delivery depends on procurement integrity, which is Goal 9. And opaque systems allocate resources to the well-connected, which is why Goal 10 on reduced inequalities cannot be separated from governance reform.

Sequencing the work

Reform programmes tend to fail when they attempt all six at once. A more workable sequence runs in three phases.

In the first six months, publish what you already hold. Most institutions have more publishable material than they release — budget documents, tender awards, audit reports, fee schedules. Making them public requires a decision rather than a project.

Over the following year, build the pipelines: an open contracting platform, a logging system for information requests, digitised asset declarations, office-level service dashboards. This is the capital-intensive phase and the one that demands sustained sponsorship from the top.

In the second and third years, close the consequence loop. Verification sampling with real referrals. Recommendation tracking with real deadlines. Sanction and recovery figures published annually. This is the phase in which governance reform either becomes real or becomes decorative, and it is also the phase most often abandoned.

Any institution can begin by scoring itself honestly across the six practices — absent, declared but unimplemented, operating internally, or published and consequential. Most bodies attempting the exercise for the first time discover it is the first honest internal assessment they have had. The score matters far less than the movement in it from one year to the next.

A message of hope

Good governance is not a moral quality that some nations possess and others lack. It is a set of practices that some institutions have installed and others have not. Each one has a documented method, a measurable output and a known failure mode, and none of them requires waiting for a change in culture. Culture follows the practices rather than preceding them.

Nigeria, Ghana and their neighbours have the legal frameworks, the professional expertise and the civic energy needed to make each of these six practices routine. What has been missing is the decision to be measured.

At Transforming Nations, we believe transparency, accountability and ethical leadership are advanced not by declaration but by installation — one publishable number at a time.